Merge git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc-merge

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

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
 

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 */

arch/powerpc/kernel/cpu_setup_power4.S

@@ -114,11 +114,11 @@ _GLOBAL(__setup_cpu_ppc970)
 
 	.data
 	.balign	L1_CACHE_BYTES,0
-cpu_state_storage:	
+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
@@ -151,7 +151,7 @@ _GLOBAL(__save_cpu_setup)
 	std	r3,CS_HID4(r5)
 	mfspr	r3,SPRN_HID5
 	std	r3,CS_HID5(r5)
-	
+
 2:
 	mtcr	r7
 	blr
@@ -213,7 +213,7 @@ _GLOBAL(__restore_cpu_setup)
 	mtspr	SPRN_HID1,r3
 	sync
 	isync
-	
+
 	/* Restore HID4 */
 	ld	r3,CS_HID4(r5)
 	sync

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,

arch/powerpc/kernel/firmware.c

@@ -1,6 +1,4 @@
 /*
- *  arch/ppc64/kernel/firmware.c
- *
  *  Extracted from cputable.c
  *
  *  Copyright (C) 2001 Ben. Herrenschmidt (benh@kernel.crashing.org)

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)
+	PPC_LL	r4,OFF(last_task_used_math)(r3)
-	CMPI	0,r4,0
+	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)
+	PPC_LL	r5,PT_REGS(r3)
-	CMPI	0,r5,0
+	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
 

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
+	/* Do early setup for that CPU (stab, slb, hash table pointer) */
-	/* Initialize the page table pointer register. */
+	bl	.early_setup_secondary
-	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
 
 	/* 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	 */
+	/* Clear backchain so we get nice backtraces */
-	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
 	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

arch/powerpc/kernel/ioctl32.c

@@ -1,6 +1,6 @@
-/* 
+/*
  * ioctl32.c: Conversion between 32bit and 64bit native ioctls.
- * 
+ *
  * Based on sparc64 ioctl32.c by:
  *
  * Copyright (C) 1997-2000  Jakub Jelinek  (jakub@redhat.com)

arch/powerpc/kernel/irq.c

@@ -5,12 +5,12 @@
  *    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@cs.nmt.edu)
+ *  Updated and modified by Cort Dougan <cort@fsmlabs.com>
- *    Copyright (C) 1996 Cort Dougan
+ *    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
@@ -21,6 +21,14 @@
  * 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>
@@ -29,6 +37,7 @@
 #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>
@@ -40,9 +49,13 @@
 #include <linux/irq.h>
 #include <linux/proc_fs.h>
 #include <linux/random.h>
-#include <linux/kallsyms.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>
@@ -52,35 +65,54 @@
 #include <asm/cache.h>
 #include <asm/prom.h>
 #include <asm/ptrace.h>
-#include <asm/iseries/it_lp_queue.h>
 #include <asm/machdep.h>
+#ifdef CONFIG_PPC64
+#include <asm/iseries/it_lp_queue.h>
 #include <asm/paca.h>
-
-#ifdef CONFIG_SMP
-extern void iSeries_smp_message_recv( struct pt_regs * );
 #endif
 
-extern irq_desc_t irq_desc[NR_IRQS];
+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;
-int ppc_spurious_interrupts;
 u64 ppc64_interrupt_controller;
+#endif /* CONFIG_PPC64 */
 
 int show_interrupts(struct seq_file *p, void *v)
 {
-	int i = *(loff_t *) v, j;
+	int i = *(loff_t *)v, j;
-	struct irqaction * action;
+	struct irqaction *action;
 	irq_desc_t *desc;
 	unsigned long flags;
 
 	if (i == 0) {
-		seq_printf(p, "           ");
+		seq_puts(p, "           ");
-		for (j=0; j<NR_CPUS; j++) {
+		for_each_online_cpu(j)
-			if (cpu_online(j))
+			seq_printf(p, "CPU%d       ", j);
-				seq_printf(p, "CPU%d       ",j);
-		}
 		seq_putc(p, '\n');
 	}
 
@@ -92,26 +124,41 @@ int show_interrupts(struct seq_file *p, void *v)
 			goto skip;
 		seq_printf(p, "%3d: ", i);
 #ifdef CONFIG_SMP
-		for (j = 0; j < NR_CPUS; j++) {
+		for_each_online_cpu(j)
-			if (cpu_online(j))
+			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
-				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 );
+			seq_printf(p, " %s ", desc->handler->typename);
 		else
-			seq_printf(p, "  None      ");
+			seq_puts(p, "  None      ");
 		seq_printf(p, "%s", (desc->status & IRQ_LEVEL) ? "Level " : "Edge  ");
-		seq_printf(p, "    %s",action->name);
+		seq_printf(p, "    %s", action->name);
-		for (action=action->next; action; action = action->next)
+		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)
+	} 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;
 }
 
@@ -144,126 +191,6 @@ void fixup_irqs(cpumask_t map)
 }
 #endif
 
-extern int noirqdebug;
-
-/*
- * Eventually, this should take an array of interrupts and an array size
- * so it can dispatch multiple interrupts.
- */
-void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq)
-{
-	int status;
-	struct irqaction *action;
-	int cpu = smp_processor_id();
-	irq_desc_t *desc = get_irq_desc(irq);
-	irqreturn_t action_ret;
-#ifdef CONFIG_IRQSTACKS
-	struct thread_info *curtp, *irqtp;
-#endif
-
-	kstat_cpu(cpu).irqs[irq]++;
-
-	if (desc->status & IRQ_PER_CPU) {
-		/* no locking required for CPU-local interrupts: */
-		ack_irq(irq);
-		action_ret = handle_IRQ_event(irq, regs, desc->action);
-		desc->handler->end(irq);
-		return;
-	}
-
-	spin_lock(&desc->lock);
-	ack_irq(irq);	
-	/*
-	   REPLAY is when Linux resends an IRQ that was dropped earlier
-	   WAITING is used by probe to mark irqs that are being tested
-	   */
-	status = desc->status & ~(IRQ_REPLAY | IRQ_WAITING);
-	status |= IRQ_PENDING; /* we _want_ to handle it */
-
-	/*
-	 * If the IRQ is disabled for whatever reason, we cannot
-	 * use the action we have.
-	 */
-	action = NULL;
-	if (likely(!(status & (IRQ_DISABLED | IRQ_INPROGRESS)))) {
-		action = desc->action;
-		if (!action || !action->handler) {
-			ppc_spurious_interrupts++;
-			printk(KERN_DEBUG "Unhandled interrupt %x, disabled\n", irq);
-			/* We can't call disable_irq here, it would deadlock */
-			if (!desc->depth)
-				desc->depth = 1;
-			desc->status |= IRQ_DISABLED;
-			/* This is not a real spurrious interrupt, we
-			 * have to eoi it, so we jump to out
-			 */
-			mask_irq(irq);
-			goto out;
-		}
-		status &= ~IRQ_PENDING; /* we commit to handling */
-		status |= IRQ_INPROGRESS; /* we are handling it */
-	}
-	desc->status = status;
-
-	/*
-	 * If there is no IRQ handler or it was disabled, exit early.
-	   Since we set PENDING, if another processor is handling
-	   a different instance of this same irq, the other processor
-	   will take care of it.
-	 */
-	if (unlikely(!action))
-		goto out;
-
-	/*
-	 * Edge triggered interrupts need to remember
-	 * pending events.
-	 * This applies to any hw interrupts that allow a second
-	 * instance of the same irq to arrive while we are in do_IRQ
-	 * or in the handler. But the code here only handles the _second_
-	 * instance of the irq, not the third or fourth. So it is mostly
-	 * useful for irq hardware that does not mask cleanly in an
-	 * SMP environment.
-	 */
-	for (;;) {
-		spin_unlock(&desc->lock);
-
-#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;
-			action_ret = call_handle_IRQ_event(irq, regs, action, irqtp);
-			irqtp->task = NULL;
-			if (irqtp->flags)
-				set_bits(irqtp->flags, &curtp->flags);
-		} else
-#endif
-			action_ret = handle_IRQ_event(irq, regs, action);
-
-		spin_lock(&desc->lock);
-		if (!noirqdebug)
-			note_interrupt(irq, desc, action_ret, regs);
-		if (likely(!(desc->status & IRQ_PENDING)))
-			break;
-		desc->status &= ~IRQ_PENDING;
-	}
-out:
-	desc->status &= ~IRQ_INPROGRESS;
-	/*
-	 * The ->end() handler has to deal with interrupts which got
-	 * disabled while the handler was running.
-	 */
-	if (desc->handler) {
-		if (desc->handler->end)
-			desc->handler->end(irq);
-		else if (desc->handler->enable)
-			desc->handler->enable(irq);
-	}
-	spin_unlock(&desc->lock);
-}
-
 #ifdef CONFIG_PPC_ISERIES
 void do_IRQ(struct pt_regs *regs)
 {
@@ -310,8 +237,11 @@ void do_IRQ(struct pt_regs *regs)
 void do_IRQ(struct pt_regs *regs)
 {
 	int irq;
+#ifdef CONFIG_IRQSTACKS
+	struct thread_info *curtp, *irqtp;
+#endif
 
-	irq_enter();
+        irq_enter();
 
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 	/* Debugging check for stack overflow: is there less than 2KB free? */
@@ -328,20 +258,44 @@ void do_IRQ(struct pt_regs *regs)
 	}
 #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)
+	if (irq >= 0) {
-		ppc_irq_dispatch_handler(regs, irq);
+#ifdef CONFIG_IRQSTACKS
-	else
+		/* Switch to the irq stack to handle this */
-		/* That's not SMP safe ... but who cares ? */
+		curtp = current_thread_info();
-		ppc_spurious_interrupts++;
+		irqtp = hardirq_ctx[smp_processor_id()];
-
+		if (curtp != irqtp) {
-	irq_exit();
+			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)
@@ -349,10 +303,14 @@ void __init init_IRQ(void)
 
 	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.
@@ -517,3 +475,4 @@ static int __init setup_noirqdistrib(char *str)
 }
 
 __setup("noirqdistrib", setup_noirqdistrib);
+#endif /* CONFIG_PPC64 */

arch/powerpc/kernel/lparcfg.c

@@ -35,6 +35,7 @@
 #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"
@@ -96,7 +97,7 @@ static unsigned long get_purr(void)
 
 #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)
@@ -168,7 +169,7 @@ static int lparcfg_data(struct seq_file *m, void *v)
 #endif				/* CONFIG_PPC_ISERIES */
 
 #ifdef CONFIG_PPC_PSERIES
-/* 
+/*
  * Methods used to fetch LPAR data when running on a pSeries platform.
  */
 
@@ -177,7 +178,7 @@ static int lparcfg_data(struct seq_file *m, void *v)
  *  entitled_capacity,unallocated_capacity,
  *  aggregation, resource_capability).
  *
- *  R4 = Entitled Processor Capacity Percentage. 
+ *  R4 = Entitled Processor Capacity Percentage.
  *  R5 = Unallocated Processor Capacity Percentage.
  *  R6 (AABBCCDDEEFFGGHH).
  *      XXXX - reserved (0)
@@ -190,7 +191,7 @@ static int lparcfg_data(struct seq_file *m, void *v)
  *          XX - variable processor Capacity Weight
  *            XX - Unallocated Variable Processor Capacity Weight.
  *              XXXX - Active processors in Physical Processor Pool.
- *                  XXXX  - Processors active on platform. 
+ *                  XXXX  - Processors active on platform.
  */
 static unsigned int h_get_ppp(unsigned long *entitled,
 			      unsigned long *unallocated,
@@ -273,7 +274,7 @@ static void parse_system_parameter_string(struct seq_file *m)
 		if (!workbuffer) {
 			printk(KERN_ERR "%s %s kmalloc failure at line %d \n",
 			       __FILE__, __FUNCTION__, __LINE__);
-			kfree(local_buffer);			
+			kfree(local_buffer);
 			return;
 		}
 #ifdef LPARCFG_DEBUG
@@ -371,7 +372,7 @@ static int lparcfg_data(struct seq_file *m, void *v)
 	lrdrp = (int *)get_property(rtas_node, "ibm,lrdr-capacity", NULL);
 
 	if (lrdrp == NULL) {
-		partition_potential_processors = systemcfg->processorCount;
+		partition_potential_processors = _systemcfg->processorCount;
 	} else {
 		partition_potential_processors = *(lrdrp + 4);
 	}

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*

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)
+	stdu	r1,THREAD_SIZE-112(r5)
-	mr	r1,r6
+	mr	r1,r5
-	bl	.handle_IRQ_event
+	bl	.__do_IRQ
 	ld	r1,0(r1)
 	ld	r0,16(r1)
 	mtlr	r0

arch/powerpc/kernel/paca.c

@@ -15,7 +15,7 @@
 #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>
@@ -24,15 +24,14 @@ static union {
 	struct systemcfg	data;
 	u8			page[PAGE_SIZE];
 } systemcfg_store __attribute__((__section__(".data.page.aligned")));
-struct systemcfg *systemcfg = &systemcfg_store.data;
+struct systemcfg *_systemcfg = &systemcfg_store.data;
-EXPORT_SYMBOL(systemcfg);
 
 
 /* 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 
+/* 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.

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

arch/powerpc/kernel/proc_ppc64.c

@@ -1,18 +1,16 @@
 /*
- * arch/ppc64/kernel/proc_ppc64.c
- *
  * 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
@@ -53,7 +51,7 @@ static int __init proc_ppc64_create(void)
 	if (!root)
 		return 1;
 
-	if (!(systemcfg->platform & (PLATFORM_PSERIES | PLATFORM_CELL)))
+	if (!(platform_is_pseries() || _machine == PLATFORM_CELL))
 		return 0;
 
 	if (!proc_mkdir("rtas", root))
@@ -74,7 +72,7 @@ static int __init proc_ppc64_init(void)
 	if (!pde)
 		return 1;
 	pde->nlink = 1;
-	pde->data = systemcfg;
+	pde->data = _systemcfg;
 	pde->size = PAGE_SIZE;
 	pde->proc_fops = &page_map_fops;
 

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;
 }

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,
+	if (align == 0 && (OF_WORKAROUNDS & OF_WA_CLAIM)) {
-			(prom_arg_t)align);
+		/*
-	if (ret != -1 && _prom->mmumap != 0)
+		 * Old OF requires we claim physical and virtual separately
-		/* old pmacs need us to map as well */
+		 * 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);
+			  ADDR("map"), _prom->mmumap, 0x12, size, virt, virt);
-	return ret;
+		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,
+static void add_string(char **str, const char *q)
-			       void *value, size_t valuelen)
 {
-	return call_prom("setprop", 4, 1, node, ADDR(pname),
+	char *p = *str;
-			 (u32)(unsigned long) value, (u32) valuelen);
+
+	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, "/rtas", "linux,rtas-base",
-	prom_setprop(rtas_node, "linux,rtas-entry", &entry, sizeof(entry));
+		     &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",
+	prom_setprop(_prom->chosen, "/chosen", "linux,stdout-path",
-		     RELOC(of_stdout_device), strlen(RELOC(of_stdout_device))+1);
+		     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,
+		prom_setprop(_prom->chosen, "/chosen", "linux,initrd-start",
-			     sizeof(val));
+			     &val, sizeof(val));
 		val = RELOC(prom_initrd_end);
-		prom_setprop(_prom->chosen, "linux,initrd-end", &val,
+		prom_setprop(_prom->chosen, "/chosen", "linux,initrd-end",
-			     sizeof(val));
+			     &val, sizeof(val));
 
 		reserve_mem(RELOC(prom_initrd_start),
 			    RELOC(prom_initrd_end) - RELOC(prom_initrd_start));
@@ -1968,16 +2044,17 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4,
 	 */
 	prom_init_client_services(pp);
 
+	/*
+	 * See if this OF is old enough that we need to do explicit maps
+	 * and other workarounds
+	 */
+	prom_find_mmu();
+
 	/*
 	 * Init prom stdout device
 	 */
 	prom_init_stdout();
 
-	/*
-	 * See if this OF is old enough that we need to do explicit maps
-	 */
-	prom_find_mmu();
-
 	/*
 	 * 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

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");

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);

arch/powerpc/kernel/rtas_pci.c

@@ -5,19 +5,19 @@
  * 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
@@ -47,7 +47,7 @@ static int write_pci_config;
 static int ibm_read_pci_config;
 static int ibm_write_pci_config;
 
-static int config_access_valid(struct pci_dn *dn, int where)
+static inline int config_access_valid(struct pci_dn *dn, int where)
 {
 	if (where < 256)
 		return 1;
@@ -72,16 +72,14 @@ static int of_device_available(struct device_node * dn)
         return 0;
 }
 
-static int rtas_read_config(struct device_node *dn, int where, int size, u32 *val)
+static int rtas_read_config(struct pci_dn *pdn, int where, int size, u32 *val)
 {
 	int returnval = -1;
 	unsigned long buid, addr;
 	int ret;
-	struct pci_dn *pdn;
 
-	if (!dn || !dn->data)
+	if (!pdn)
 		return PCIBIOS_DEVICE_NOT_FOUND;
-	pdn = dn->data;
 	if (!config_access_valid(pdn, where))
 		return PCIBIOS_BAD_REGISTER_NUMBER;
 
@@ -90,7 +88,7 @@ static int rtas_read_config(struct device_node *dn, int where, int size, u32 *va
 	buid = pdn->phb->buid;
 	if (buid) {
 		ret = rtas_call(ibm_read_pci_config, 4, 2, &returnval,
-				addr, buid >> 32, buid & 0xffffffff, size);
+				addr, BUID_HI(buid), BUID_LO(buid), size);
 	} else {
 		ret = rtas_call(read_pci_config, 2, 2, &returnval, addr, size);
 	}
@@ -100,7 +98,7 @@ static int rtas_read_config(struct device_node *dn, int where, int size, u32 *va
 		return PCIBIOS_DEVICE_NOT_FOUND;
 
 	if (returnval == EEH_IO_ERROR_VALUE(size) &&
-	    eeh_dn_check_failure (dn, NULL))
+	    eeh_dn_check_failure (pdn->node, NULL))
 		return PCIBIOS_DEVICE_NOT_FOUND;
 
 	return PCIBIOS_SUCCESSFUL;
@@ -118,23 +116,23 @@ static int rtas_pci_read_config(struct pci_bus *bus,
 		busdn = bus->sysdata;	/* must be a phb */
 
 	/* Search only direct children of the bus */
-	for (dn = busdn->child; dn; dn = dn->sibling)
+	for (dn = busdn->child; dn; dn = dn->sibling) {
-		if (dn->data && PCI_DN(dn)->devfn == devfn
+		struct pci_dn *pdn = PCI_DN(dn);
+		if (pdn && pdn->devfn == devfn
 		    && of_device_available(dn))
-			return rtas_read_config(dn, where, size, val);
+			return rtas_read_config(pdn, where, size, val);
+	}
 
 	return PCIBIOS_DEVICE_NOT_FOUND;
 }
 
-int rtas_write_config(struct device_node *dn, int where, int size, u32 val)
+int rtas_write_config(struct pci_dn *pdn, int where, int size, u32 val)
 {
 	unsigned long buid, addr;
 	int ret;
-	struct pci_dn *pdn;
 
-	if (!dn || !dn->data)
+	if (!pdn)
 		return PCIBIOS_DEVICE_NOT_FOUND;
-	pdn = dn->data;
 	if (!config_access_valid(pdn, where))
 		return PCIBIOS_BAD_REGISTER_NUMBER;
 
@@ -142,7 +140,8 @@ int rtas_write_config(struct device_node *dn, int where, int size, u32 val)
 		(pdn->devfn << 8) | (where & 0xff);
 	buid = pdn->phb->buid;
 	if (buid) {
-		ret = rtas_call(ibm_write_pci_config, 5, 1, NULL, addr, buid >> 32, buid & 0xffffffff, size, (ulong) val);
+		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);
 	}
@@ -165,10 +164,12 @@ static int rtas_pci_write_config(struct pci_bus *bus,
 		busdn = bus->sysdata;	/* must be a phb */
 
 	/* Search only direct children of the bus */
-	for (dn = busdn->child; dn; dn = dn->sibling)
+	for (dn = busdn->child; dn; dn = dn->sibling) {
-		if (dn->data && PCI_DN(dn)->devfn == devfn
+		struct pci_dn *pdn = PCI_DN(dn);
+		if (pdn && pdn->devfn == devfn
 		    && of_device_available(dn))
-			return rtas_write_config(dn, where, size, val);
+			return rtas_write_config(pdn, where, size, val);
+	}
 	return PCIBIOS_DEVICE_NOT_FOUND;
 }
 
@@ -221,7 +222,7 @@ static void python_countermeasures(struct device_node *dev,
 	/* 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
 	 */
@@ -292,7 +293,7 @@ static int phb_set_bus_ranges(struct device_node *dev,
 	if (bus_range == NULL || len < 2 * sizeof(int)) {
 		return 1;
  	}
- 
+
 	phb->first_busno =  bus_range[0];
 	phb->last_busno  =  bus_range[1];
 

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 &&
+	if (_machine == PLATFORM_PSERIES_LPAR &&
-				(dn = of_find_node_by_path("/rtas"))) {
+	    (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

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 */

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
+#ifdef CONFIG_XMON_DEFAULT
-	xmon_map_scc();
+	xmon_init(1);
-	if (strstr(cmd_line, "xmon")) {
+#endif
-		xmon_init(1);
-		debugger(NULL);
-	}
-#endif /* CONFIG_XMON */
-	if ( ppc_md.progress ) ppc_md.progress("setup_arch: enter", 0x3eab);
 
 #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);

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",
+	DBG("Probing machine type for platform %x...\n", _machine);
-	    systemcfg->platform);
 
 	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_size = ppc64_caches.dsize = size;
-			systemcfg->dcache_line_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_size = ppc64_caches.isize = size;
-			systemcfg->icache_line_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");
+	strcpy(_systemcfg->eye_catcher, "SYSTEMCFG:PPC64");
-	systemcfg->version.major = SYSTEMCFG_MAJOR;
+	_systemcfg->version.major = SYSTEMCFG_MAJOR;
-	systemcfg->version.minor = SYSTEMCFG_MINOR;
+	_systemcfg->version.minor = SYSTEMCFG_MINOR;
-	systemcfg->processor = mfspr(SPRN_PVR);
+	_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                     = 0x%p\n", _systemcfg);
-	printk("systemcfg->platform           = 0x%x\n", systemcfg->platform);
+	printk("systemcfg->platform           = 0x%x\n", _systemcfg->platform);
-	printk("systemcfg->processorCount     = 0x%lx\n", systemcfg->processorCount);
+	printk("systemcfg->processorCount     = 0x%lx\n", _systemcfg->processorCount);
-	printk("systemcfg->physicalMemorySize = 0x%lx\n", systemcfg->physicalMemorySize);
+	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)

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>

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();
 }

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 */

arch/powerpc/kernel/sysfs.c

@@ -232,7 +232,7 @@ static void register_cpu_online(unsigned int cpu)
 		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);
 }

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->tb_orig_stamp = new_tb_stamp;
-	systemcfg->stamp_xsec = new_stamp_xsec;
+	_systemcfg->stamp_xsec = new_stamp_xsec;
-	systemcfg->tb_to_xs = new_tb_to_xs;
+	_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_ticks_per_sec =
-				systemcfg->tb_to_xs = tb_to_xs;
+					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_minuteswest = sys_tz.tz_minuteswest;
-	systemcfg->tz_dsttime = sys_tz.tz_dsttime;
+	_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_orig_stamp = tb_last_jiffy;
-	systemcfg->tb_update_count = 0;
+	_systemcfg->tb_update_count = 0;
-	systemcfg->tb_ticks_per_sec = tb_ticks_per_sec;
+	_systemcfg->tb_ticks_per_sec = tb_ticks_per_sec;
-	systemcfg->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC;
+	_systemcfg->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC;
-	systemcfg->tb_to_xs = tb_to_xs;
+	_systemcfg->tb_to_xs = tb_to_xs;
 #endif
 
 	time_freq = 0;

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;

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:

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);
+	mem_size = lmb_phys_mem_size();
-	if (rnd_mem_size < systemcfg->physicalMemorySize)
+	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
  */

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;
 
 /*

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] = {
+static const unsigned int pgtable_cache_size[3] = {
-	PTE_TABLE_SIZE, PGD_TABLE_SIZE
+	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 |

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;
 

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))
+				      mmu_virtual_psize)) {
-			panic("Can't map bolted IO mapping");
+			printk(KERN_ERR "Failed to do bolted mapping IO "
+			       "memory at %016lx !\n", pa);
+			return -ENOMEM;
+		}
 	}
 	return 0;
 }

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);
 }

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) &&
+	if (platform_is_lpar() && (mmcra & MMCRA_SIHV))
-	    (mmcra & MMCRA_SIHV))
 		/* function descriptor madness */
 		return *((unsigned long *)hypervisor_bucket);
 

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)

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
+ * This does nothing because there is not enough information
- * enable_IRQ at the end of interrupt handling.  However, this does
+ * provided to do the EOI HvCall.  This is done by XmPciLpEvent.c
- * 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)
 {

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
 

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;
+	_systemcfg->processor = xIoHriProcessorVpd[procIx].xPVR;
-	printk("Processor version = %x\n", systemcfg->processor);
+	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);
 }

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");
 }
 

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 */

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).

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)
+	 * This is necessary because OF doesn't know about the
-		smp_hw_index[i] = i;
+	 * 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
+};

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

arch/powerpc/platforms/pseries/eeh.c

@@ -1,39 +1,37 @@
 /*
  * 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/bootmem.h>
+#include <linux/delay.h>
 #include <linux/init.h>
 #include <linux/list.h>
-#include <linux/mm.h>
-#include <linux/notifier.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/rtas.h>
-#include <asm/atomic.h>
-#include <asm/systemcfg.h>
 #include <asm/ppc-pci.h>
+#include <asm/rtas.h>
 
 #undef DEBUG
 
@@ -49,8 +47,8 @@
  *  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 dust, vibration, humidity,
+ *  which in turn might be caused by low voltage on the bus, dust,
- *  radioactivity or plain-old failed hardware.
+ *  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,
@@ -71,26 +69,15 @@
  *  and sent out for processing.
  */
 
-/** Bus Unit ID macros; get low and hi 32-bits of the 64-bit BUID */
+/* If a device driver keeps reading an MMIO register in an interrupt
-#define BUID_HI(buid) ((buid) >> 32)
-#define BUID_LO(buid) ((buid) & 0xffffffff)
-
-/* EEH event workqueue setup. */
-static DEFINE_SPINLOCK(eeh_eventlist_lock);
-LIST_HEAD(eeh_eventlist);
-static void eeh_event_handler(void *);
-DECLARE_WORK(eeh_event_wq, eeh_event_handler, NULL);
-
-static struct notifier_block *eeh_notifier_chain;
-
-/*
- * 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	1000
+#define EEH_MAX_FAILS	100000
-static atomic_t eeh_fail_count;
+
+/* Misc forward declaraions */
+static void eeh_save_bars(struct pci_dev * pdev, struct pci_dn *pdn);
 
 /* RTAS tokens */
 static int ibm_set_eeh_option;
@@ -101,12 +88,19 @@ 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);
@@ -224,9 +218,9 @@ pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo,
 	while (*p) {
 		parent = *p;
 		piar = rb_entry(parent, struct pci_io_addr_range, rb_node);
-		if (alo < piar->addr_lo) {
+		if (ahi < piar->addr_lo) {
 			p = &parent->rb_left;
-		} else if (ahi > piar->addr_hi) {
+		} else if (alo > piar->addr_hi) {
 			p = &parent->rb_right;
 		} else {
 			if (dev != piar->pcidev ||
@@ -245,6 +239,11 @@ pci_addr_cache_insert(struct pci_dev *dev, unsigned long alo,
 	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);
 
@@ -260,18 +259,17 @@ static void __pci_addr_cache_insert_device(struct pci_dev *dev)
 
 	dn = pci_device_to_OF_node(dev);
 	if (!dn) {
-		printk(KERN_WARNING "PCI: no pci dn found for dev=%s\n",
+		printk(KERN_WARNING "PCI: no pci dn found for dev=%s\n", pci_name(dev));
-			pci_name(dev));
 		return;
 	}
 
 	/* Skip any devices for which EEH is not enabled. */
-	pdn = dn->data;
+	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\n",
+		printk(KERN_INFO "PCI: skip building address cache for=%s - %s\n",
-		       pci_name(dev));
+		       pci_name(dev), pdn->node->full_name);
 #endif
 		return;
 	}
@@ -307,7 +305,7 @@ static void __pci_addr_cache_insert_device(struct pci_dev *dev)
  * we maintain a cache of devices that can be quickly searched.
  * This routine adds a device to that cache.
  */
-void pci_addr_cache_insert_device(struct pci_dev *dev)
+static void pci_addr_cache_insert_device(struct pci_dev *dev)
 {
 	unsigned long flags;
 
@@ -350,7 +348,7 @@ restart:
  * the tree multiple times (once per resource).
  * But so what; device removal doesn't need to be that fast.
  */
-void pci_addr_cache_remove_device(struct pci_dev *dev)
+static void pci_addr_cache_remove_device(struct pci_dev *dev)
 {
 	unsigned long flags;
 
@@ -370,8 +368,12 @@ void pci_addr_cache_remove_device(struct pci_dev *dev)
  */
 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) {
@@ -380,6 +382,10 @@ void __init pci_addr_cache_build(void)
 			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
@@ -391,22 +397,26 @@ void __init pci_addr_cache_build(void)
 /* --------------------------------------------------------------- */
 /* Above lies the PCI Address Cache. Below lies the EEH event infrastructure */
 
-/**
+void eeh_slot_error_detail (struct pci_dn *pdn, int severity)
- * eeh_register_notifier - Register to find out about EEH events.
- * @nb: notifier block to callback on events
- */
-int eeh_register_notifier(struct notifier_block *nb)
 {
-	return notifier_chain_register(&eeh_notifier_chain, nb);
+	unsigned long flags;
-}
+	int rc;
 
-/**
+	/* Log the error with the rtas logger */
- * eeh_unregister_notifier - Unregister to an EEH event notifier.
+	spin_lock_irqsave(&slot_errbuf_lock, flags);
- * @nb: notifier block to callback on events
+	memset(slot_errbuf, 0, eeh_error_buf_size);
- */
+
-int eeh_unregister_notifier(struct notifier_block *nb)
+	rc = rtas_call(ibm_slot_error_detail,
-{
+	               8, 1, NULL, pdn->eeh_config_addr,
-	return notifier_chain_unregister(&eeh_notifier_chain, nb);
+	               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);
 }
 
 /**
@@ -414,16 +424,16 @@ int eeh_unregister_notifier(struct notifier_block *nb)
  * @dn: device node to read
  * @rets: array to return results in
  */
-static int read_slot_reset_state(struct device_node *dn, int rets[])
+static int read_slot_reset_state(struct pci_dn *pdn, int rets[])
 {
 	int token, outputs;
-	struct pci_dn *pdn = dn->data;
 
 	if (ibm_read_slot_reset_state2 != RTAS_UNKNOWN_SERVICE) {
 		token = ibm_read_slot_reset_state2;
 		outputs = 4;
 	} else {
 		token = ibm_read_slot_reset_state;
+		rets[2] = 0; /* fake PE Unavailable info */
 		outputs = 3;
 	}
 
@@ -431,76 +441,9 @@ static int read_slot_reset_state(struct device_node *dn, int rets[])
 			 BUID_HI(pdn->phb->buid), BUID_LO(pdn->phb->buid));
 }
 
-/**
- * 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)
-{
-	/*
-	 * XXX We should create a separate sysctl for this.
-	 *
-	 * 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 {
-		__get_cpu_var(ignored_failures)++;
-		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 void eeh_event_handler(void *dummy)
-{
-	unsigned long flags;
-	struct eeh_event	*event;
-
-	while (1) {
-		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: MMIO failure (%d), notifiying device "
-		       "%s\n", event->reset_state,
-		       pci_name(event->dev));
-
-		atomic_set(&eeh_fail_count, 0);
-		notifier_call_chain (&eeh_notifier_chain,
-				     EEH_NOTIFY_FREEZE, event);
-
-		__get_cpu_var(slot_resets)++;
-
-		pci_dev_put(event->dev);
-		kfree(event);
-	}
-}
-
 /**
  * eeh_token_to_phys - convert EEH address token to phys address
- * @token i/o token, should be address in the form 0xE....
+ * @token i/o token, should be address in the form 0xA....
  */
 static inline unsigned long eeh_token_to_phys(unsigned long token)
 {
@@ -515,6 +458,70 @@ static inline unsigned long eeh_token_to_phys(unsigned long token)
 	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
@@ -526,7 +533,7 @@ static inline unsigned long eeh_token_to_phys(unsigned long token)
  * 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 solt isolation event notification.
+ * a non-zero value and queues up a slot isolation event notification.
  *
  * It is safe to call this routine in an interrupt context.
  */
@@ -535,42 +542,59 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
 	int ret;
 	int rets[3];
 	unsigned long flags;
-	int rc, reset_state;
-	struct eeh_event  *event;
 	struct pci_dn *pdn;
+	int rc = 0;
 
 	__get_cpu_var(total_mmio_ffs)++;
 
 	if (!eeh_subsystem_enabled)
 		return 0;
 
-	if (!dn)
+	if (!dn) {
+		__get_cpu_var(no_dn)++;
 		return 0;
-	pdn = dn->data;
+	}
+	pdn = PCI_DN(dn);
 
 	/* Access to IO BARs might get this far and still not want checking. */
-	if (!pdn->eeh_capable || !(pdn->eeh_mode & EEH_MODE_SUPPORTED) ||
+	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
-	 * If we already have a pending isolation event for this
+	 * slot, we know it's bad already, we don't need to check.
-	 * 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) {
-		atomic_inc(&eeh_fail_count);
+		pdn->eeh_check_count ++;
-		if (atomic_read(&eeh_fail_count) >= EEH_MAX_FAILS) {
+		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(dn, rets) != 0)
+			if (read_slot_reset_state(pdn, rets) != 0)
 				rets[0] = -1;	/* reset state unknown */
-			eeh_panic(dev, rets[0]);
+
+			/* 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));
 		}
-		return 0;
+		goto dn_unlock;
 	}
 
 	/*
@@ -580,66 +604,69 @@ int eeh_dn_check_failure(struct device_node *dn, struct pci_dev *dev)
 	 * function zero of a multi-function device.
 	 * In any case they must share a common PHB.
 	 */
-	ret = read_slot_reset_state(dn, rets);
+	ret = read_slot_reset_state(pdn, rets);
-	if (!(ret == 0 && rets[1] == 1 && (rets[0] == 2 || rets[0] == 4))) {
+
+	/* 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)++;
-		return 0;
+		rc = 0;
+		goto dn_unlock;
 	}
 
-	/* prevent repeated reports of this failure */
+	/* If EEH is not supported on this device, punt. */
-	pdn->eeh_mode |= EEH_MODE_ISOLATED;
+	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;
+	}
 
-	reset_state = rets[0];
+	/* 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;
+	}
 
-	spin_lock_irqsave(&slot_errbuf_lock, flags);
+	/* Note that config-io to empty slots may fail;
-	memset(slot_errbuf, 0, eeh_error_buf_size);
+	 * 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;
+	}
 
-	rc = rtas_call(ibm_slot_error_detail,
+	__get_cpu_var(slot_resets)++;
-	               8, 1, NULL, pdn->eeh_config_addr,
+ 
-	               BUID_HI(pdn->phb->buid),
+	/* Avoid repeated reports of this failure, including problems
-	               BUID_LO(pdn->phb->buid), NULL, 0,
+	 * with other functions on this device, and functions under
-	               virt_to_phys(slot_errbuf),
+	 * bridges. */
-	               eeh_error_buf_size,
+	eeh_mark_slot (dn, EEH_MODE_ISOLATED);
-	               1 /* Temporary Error */);
+	spin_unlock_irqrestore(&confirm_error_lock, flags);
-
-	if (rc == 0)
-		log_error(slot_errbuf, ERR_TYPE_RTAS_LOG, 0);
-	spin_unlock_irqrestore(&slot_errbuf_lock, flags);
-
-	printk(KERN_INFO "EEH: MMIO failure (%d) on device: %s %s\n",
-	       rets[0], dn->name, dn->full_name);
-	event = kmalloc(sizeof(*event), GFP_ATOMIC);
-	if (event == NULL) {
-		eeh_panic(dev, reset_state);
-		return 1;
- 	}
-
-	event->dev = dev;
-	event->dn = dn;
-	event->reset_state = reset_state;
-
-	/* 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);
 
+	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. */
-	dump_stack();
+	if (rets[0] != 5) dump_stack();
-	schedule_work(&eeh_event_wq);
+	return 1;
 
-	return 0;
+dn_unlock:
+	spin_unlock_irqrestore(&confirm_error_lock, flags);
+	return rc;
 }
 
-EXPORT_SYMBOL(eeh_dn_check_failure);
+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.
  * 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
@@ -656,8 +683,10 @@ unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned lon
 	/* 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)
+	if (!dev) {
+		__get_cpu_var(no_device)++;
 		return val;
+	}
 
 	dn = pci_device_to_OF_node(dev);
 	eeh_dn_check_failure (dn, dev);
@@ -668,6 +697,217 @@ unsigned long eeh_check_failure(const volatile void __iomem *token, unsigned lon
 
 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;
@@ -684,9 +924,11 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
 	u32 *device_id = (u32 *)get_property(dn, "device-id", NULL);
 	u32 *regs;
 	int enable;
-	struct pci_dn *pdn = dn->data;
+	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 */
@@ -723,8 +965,9 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
 		/* 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,
+		                regs[0], info->buid_hi, info->buid_lo,
-				EEH_ENABLE);
+		                EEH_ENABLE);
+
 		if (ret == 0) {
 			eeh_subsystem_enabled = 1;
 			pdn->eeh_mode |= EEH_MODE_SUPPORTED;
@@ -736,7 +979,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
 
 			/* This device doesn't support EEH, but it may have an
 			 * EEH parent, in which case we mark it as supported. */
-			if (dn->parent && dn->parent->data
+			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;
@@ -749,7 +992,7 @@ static void *early_enable_eeh(struct device_node *dn, void *data)
 		       dn->full_name);
 	}
 
-	return NULL; 
+	return NULL;
 }
 
 /*
@@ -770,6 +1013,9 @@ 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;
@@ -797,13 +1043,11 @@ void __init eeh_init(void)
 	for (phb = of_find_node_by_name(NULL, "pci"); phb;
 	     phb = of_find_node_by_name(phb, "pci")) {
 		unsigned long buid;
-		struct pci_dn *pci;
 
 		buid = get_phb_buid(phb);
-		if (buid == 0 || phb->data == NULL)
+		if (buid == 0 || PCI_DN(phb) == NULL)
 			continue;
 
-		pci = phb->data;
 		info.buid_lo = BUID_LO(buid);
 		info.buid_hi = BUID_HI(buid);
 		traverse_pci_devices(phb, early_enable_eeh, &info);
@@ -832,11 +1076,13 @@ void eeh_add_device_early(struct device_node *dn)
 	struct pci_controller *phb;
 	struct eeh_early_enable_info info;
 
-	if (!dn || !dn->data)
+	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\n");
+		printk(KERN_WARNING "EEH: Expected buid but found none for %s\n",
+		       dn->full_name);
+		dump_stack();
 		return;
 	}
 
@@ -844,7 +1090,7 @@ void eeh_add_device_early(struct device_node *dn)
 	info.buid_lo = BUID_LO(phb->buid);
 	early_enable_eeh(dn, &info);
 }
-EXPORT_SYMBOL(eeh_add_device_early);
+EXPORT_SYMBOL_GPL(eeh_add_device_early);
 
 /**
  * eeh_add_device_late - perform EEH initialization for the indicated pci device
@@ -855,6 +1101,9 @@ EXPORT_SYMBOL(eeh_add_device_early);
  */
 void eeh_add_device_late(struct pci_dev *dev)
 {
+	struct device_node *dn;
+	struct pci_dn *pdn;
+
 	if (!dev || !eeh_subsystem_enabled)
 		return;
 
@@ -862,9 +1111,15 @@ void eeh_add_device_late(struct pci_dev *dev)
 	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(eeh_add_device_late);
+EXPORT_SYMBOL_GPL(eeh_add_device_late);
 
 /**
  * eeh_remove_device - undo EEH setup for the indicated pci device
@@ -875,6 +1130,7 @@ EXPORT_SYMBOL(eeh_add_device_late);
  */
 void eeh_remove_device(struct pci_dev *dev)
 {
+	struct device_node *dn;
 	if (!dev || !eeh_subsystem_enabled)
 		return;
 
@@ -883,20 +1139,29 @@ void eeh_remove_device(struct pci_dev *dev)
 	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(eeh_remove_device);
+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) {
@@ -904,13 +1169,17 @@ static int proc_eeh_show(struct seq_file *m, void *v)
 		seq_printf(m, "eeh_total_mmio_ffs=%ld\n", ffs);
 	} else {
 		seq_printf(m, "EEH Subsystem is enabled\n");
-		seq_printf(m, "eeh_total_mmio_ffs=%ld\n"
+		seq_printf(m,
-			   "eeh_false_positives=%ld\n"
+				"no device=%ld\n"
-			   "eeh_ignored_failures=%ld\n"
+				"no device node=%ld\n"
-			   "eeh_slot_resets=%ld\n"
+				"no config address=%ld\n"
-				"eeh_fail_count=%d\n",
+				"check not wanted=%ld\n"
-			   ffs, positives, failures, resets,
+				"eeh_total_mmio_ffs=%ld\n"
-				eeh_fail_count.counter);
+				"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;
@@ -932,7 +1201,7 @@ static int __init eeh_init_proc(void)
 {
 	struct proc_dir_entry *e;
 
-	if (systemcfg->platform & PLATFORM_PSERIES) {
+	if (platform_is_pseries()) {
 		e = create_proc_entry("ppc64/eeh", 0, NULL);
 		if (e)
 			e->proc_fops = &proc_eeh_operations;

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 ******************************/

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;

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");

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);

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;
 	}
 

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;
 }

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
 

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)

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;

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

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);
+}

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))
+#define printf	xmon_printf
+#define putchar	xmon_putchar
+
+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);

arch/powerpc/xmon/setjmp.S

@@ -14,61 +14,61 @@
 
 _GLOBAL(xmon_setjmp)
 	mflr	r0
-	STL	r0,0(r3)
+	PPC_STL	r0,0(r3)
-	STL	r1,SZL(r3)
+	PPC_STL	r1,SZL(r3)
-	STL	r2,2*SZL(r3)
+	PPC_STL	r2,2*SZL(r3)
 	mfcr	r0
-	STL	r0,3*SZL(r3)
+	PPC_STL	r0,3*SZL(r3)
-	STL	r13,4*SZL(r3)
+	PPC_STL	r13,4*SZL(r3)
-	STL	r14,5*SZL(r3)
+	PPC_STL	r14,5*SZL(r3)
-	STL	r15,6*SZL(r3)
+	PPC_STL	r15,6*SZL(r3)
-	STL	r16,7*SZL(r3)
+	PPC_STL	r16,7*SZL(r3)
-	STL	r17,8*SZL(r3)
+	PPC_STL	r17,8*SZL(r3)
-	STL	r18,9*SZL(r3)
+	PPC_STL	r18,9*SZL(r3)
-	STL	r19,10*SZL(r3)
+	PPC_STL	r19,10*SZL(r3)
-	STL	r20,11*SZL(r3)
+	PPC_STL	r20,11*SZL(r3)
-	STL	r21,12*SZL(r3)
+	PPC_STL	r21,12*SZL(r3)
-	STL	r22,13*SZL(r3)
+	PPC_STL	r22,13*SZL(r3)
-	STL	r23,14*SZL(r3)
+	PPC_STL	r23,14*SZL(r3)
-	STL	r24,15*SZL(r3)
+	PPC_STL	r24,15*SZL(r3)
-	STL	r25,16*SZL(r3)
+	PPC_STL	r25,16*SZL(r3)
-	STL	r26,17*SZL(r3)
+	PPC_STL	r26,17*SZL(r3)
-	STL	r27,18*SZL(r3)
+	PPC_STL	r27,18*SZL(r3)
-	STL	r28,19*SZL(r3)
+	PPC_STL	r28,19*SZL(r3)
-	STL	r29,20*SZL(r3)
+	PPC_STL	r29,20*SZL(r3)
-	STL	r30,21*SZL(r3)
+	PPC_STL	r30,21*SZL(r3)
-	STL	r31,22*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)
+1:	PPC_LL	r13,4*SZL(r3)
-	LDL	r14,5*SZL(r3)
+	PPC_LL	r14,5*SZL(r3)
-	LDL	r15,6*SZL(r3)
+	PPC_LL	r15,6*SZL(r3)
-	LDL	r16,7*SZL(r3)
+	PPC_LL	r16,7*SZL(r3)
-	LDL	r17,8*SZL(r3)
+	PPC_LL	r17,8*SZL(r3)
-	LDL	r18,9*SZL(r3)
+	PPC_LL	r18,9*SZL(r3)
-	LDL	r19,10*SZL(r3)
+	PPC_LL	r19,10*SZL(r3)
-	LDL	r20,11*SZL(r3)
+	PPC_LL	r20,11*SZL(r3)
-	LDL	r21,12*SZL(r3)
+	PPC_LL	r21,12*SZL(r3)
-	LDL	r22,13*SZL(r3)
+	PPC_LL	r22,13*SZL(r3)
-	LDL	r23,14*SZL(r3)
+	PPC_LL	r23,14*SZL(r3)
-	LDL	r24,15*SZL(r3)
+	PPC_LL	r24,15*SZL(r3)
-	LDL	r25,16*SZL(r3)
+	PPC_LL	r25,16*SZL(r3)
-	LDL	r26,17*SZL(r3)
+	PPC_LL	r26,17*SZL(r3)
-	LDL	r27,18*SZL(r3)
+	PPC_LL	r27,18*SZL(r3)
-	LDL	r28,19*SZL(r3)
+	PPC_LL	r28,19*SZL(r3)
-	LDL	r29,20*SZL(r3)
+	PPC_LL	r29,20*SZL(r3)
-	LDL	r30,21*SZL(r3)
+	PPC_LL	r30,21*SZL(r3)
-	LDL	r31,22*SZL(r3)
+	PPC_LL	r31,22*SZL(r3)
-	LDL	r0,3*SZL(r3)
+	PPC_LL	r0,3*SZL(r3)
 	mtcrf	0x38,r0
-	LDL	r0,0(r3)
+	PPC_LL	r0,0(r3)
-	LDL	r1,SZL(r3)
+	PPC_LL	r1,SZL(r3)
-	LDL	r2,2*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)
+	PPC_STL	r0,0*SZL(r3)
-	STL	r2,2*SZL(r3)
+	PPC_STL	r2,2*SZL(r3)
-	STL	r3,3*SZL(r3)
+	PPC_STL	r3,3*SZL(r3)
-	STL	r4,4*SZL(r3)
+	PPC_STL	r4,4*SZL(r3)
-	STL	r5,5*SZL(r3)
+	PPC_STL	r5,5*SZL(r3)
-	STL	r6,6*SZL(r3)
+	PPC_STL	r6,6*SZL(r3)
-	STL	r7,7*SZL(r3)
+	PPC_STL	r7,7*SZL(r3)
-	STL	r8,8*SZL(r3)
+	PPC_STL	r8,8*SZL(r3)
-	STL	r9,9*SZL(r3)
+	PPC_STL	r9,9*SZL(r3)
-	STL	r10,10*SZL(r3)
+	PPC_STL	r10,10*SZL(r3)
-	STL	r11,11*SZL(r3)
+	PPC_STL	r11,11*SZL(r3)
-	STL	r12,12*SZL(r3)
+	PPC_STL	r12,12*SZL(r3)
-	STL	r13,13*SZL(r3)
+	PPC_STL	r13,13*SZL(r3)
-	STL	r14,14*SZL(r3)
+	PPC_STL	r14,14*SZL(r3)
-	STL	r15,15*SZL(r3)
+	PPC_STL	r15,15*SZL(r3)
-	STL	r16,16*SZL(r3)
+	PPC_STL	r16,16*SZL(r3)
-	STL	r17,17*SZL(r3)
+	PPC_STL	r17,17*SZL(r3)
-	STL	r18,18*SZL(r3)
+	PPC_STL	r18,18*SZL(r3)
-	STL	r19,19*SZL(r3)
+	PPC_STL	r19,19*SZL(r3)
-	STL	r20,20*SZL(r3)
+	PPC_STL	r20,20*SZL(r3)
-	STL	r21,21*SZL(r3)
+	PPC_STL	r21,21*SZL(r3)
-	STL	r22,22*SZL(r3)
+	PPC_STL	r22,22*SZL(r3)
-	STL	r23,23*SZL(r3)
+	PPC_STL	r23,23*SZL(r3)
-	STL	r24,24*SZL(r3)
+	PPC_STL	r24,24*SZL(r3)
-	STL	r25,25*SZL(r3)
+	PPC_STL	r25,25*SZL(r3)
-	STL	r26,26*SZL(r3)
+	PPC_STL	r26,26*SZL(r3)
-	STL	r27,27*SZL(r3)
+	PPC_STL	r27,27*SZL(r3)
-	STL	r28,28*SZL(r3)
+	PPC_STL	r28,28*SZL(r3)
-	STL	r29,29*SZL(r3)
+	PPC_STL	r29,29*SZL(r3)
-	STL	r30,30*SZL(r3)
+	PPC_STL	r30,30*SZL(r3)
-	STL	r31,31*SZL(r3)
+	PPC_STL	r31,31*SZL(r3)
 	/* go up one stack frame for SP */
-	LDL	r4,0(r1)
+	PPC_LL	r4,0(r1)
-	STL	r4,1*SZL(r3)
+	PPC_STL	r4,1*SZL(r3)
 	/* get caller's LR */
-	LDL	r0,LRSAVE(r4)
+	PPC_LL	r0,LRSAVE(r4)
-	STL	r0,_NIP-STACK_FRAME_OVERHEAD(r3)
+	PPC_STL	r0,_NIP-STACK_FRAME_OVERHEAD(r3)
-	STL	r0,_LINK-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

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
+void xmon_map_scc(void)
-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)
 {
 #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
+int xmon_write(void *ptr, int nb)
-xmon_write(void *handle, 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
+static int xmon_get_adb_key(void)
-xmon_get_adb_key(void)
 {
 	int k, t, on;
 
@@ -350,32 +320,21 @@ xmon_get_adb_key(void)
 }
 #endif /* CONFIG_BOOTX_TEXT */
 
-int
+int xmon_readchar(void)
-xmon_read(void *handle, void *ptr, int nb)
 {
-    char *p = ptr;
-    int i;
-
 #ifdef CONFIG_BOOTX_TEXT
-    if (use_screen) {
+	if (use_screen)
-	for (i = 0; i < nb; ++i)
+		return xmon_get_adb_key();
-	    *p++ = xmon_get_adb_key();
-	return i;
-    }
 #endif
-    if (!scc_initialized)
+	if (!scc_initialized)
-	xmon_init_scc();
+		xmon_init_scc();
-    for (i = 0; i < nb; ++i) {
 	while ((*sccc & RXRDY) == 0)
-	    do_poll_adb();
+		do_poll_adb();
 	buf_access();
-	*p++ = *sccd;
+	return *sccd;
-    }
-    return i;
 }
 
-int
+int xmon_read_poll(void)
-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
+void xmon_init_scc(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();
+			t0 = get_tbl();
-			while (readtb() - t0 < 3*TB_SPEED)
+			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_enter(void)
-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)
 {
 #ifdef CONFIG_ADB_PMU
 	if (_machine == _MACH_Pmac) {
@@ -613,8 +431,7 @@ xmon_enter(void)
 #endif
 }
 
-void
+void xmon_leave(void)
-xmon_leave(void)
 {
 #ifdef CONFIG_ADB_PMU
 	if (_machine == _MACH_Pmac) {

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
+void xmon_map_scc(void)
-
-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 = 
+int xmon_write(void *ptr, int nb)
-{
-	.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)
 {
 	return udbg_write(ptr, nb);
 }
 
-int
+int xmon_readchar(void)
-xmon_read(void *handle, void *ptr, int nb)
 {
-	return udbg_read(ptr, nb);
+	if (udbg_getc)
+		return udbg_getc();
+	return -1;
 }
 
-int
+int xmon_read_poll(void)
-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;
-}

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 */
+void xmon_map_scc(void)
-
-#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)
-{
-
 	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
+int xmon_write(void *ptr, int nb)
-xmon_write(void *handle, 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_readchar(void)
-
-int
-xmon_read(void *handle, void *ptr, int nb)
 {
-	char *p = ptr;
+	return xmon_8xx_read_char();
-	int i;
-
-	if (!scc_initialized)
-		xmon_init_scc();
-
-	for (i = 0; i < nb; ++i) {
-		*p++ = xmon_8xx_read_char();
-	}
-	return i;
 }
 
-int
+int xmon_read_poll(void)
-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);
-}

arch/powerpc/xmon/subr_prf.c

@@ -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);
-}
-

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)
+		        if ((m & (sizeof(long) - 1)) == 0 && m > 0)
-			    putchar(' ');
+				putchar(' ');
 			if (m < nr)
 				printf("%.2x", temp[m]);
 			else
 				printf("%s", fault_chars[fault_type]);
 		}
-		if (m <= 8)
+		for (; m < 16; ++m) {
-			printf(" ");
+		        if ((m & (sizeof(long) - 1)) == 0)
-		for (; m < 16; ++m)
+				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 */

arch/ppc/boot/include/of1275.h

@@ -19,6 +19,9 @@ extern prom_entry of_prom_entry;
 
 /* function declarations */
 
+int	call_prom(const char *service, int nargs, int nret, ...);
+int	call_prom_ret(const char *service, int nargs, int nret,
+		      unsigned int *rets, ...);
 void *	claim(unsigned int virt, unsigned int size, unsigned int align);
 int	map(unsigned int phys, unsigned int virt, unsigned int size);
 void	enter(void);

arch/ppc/boot/of1275/Makefile

@@ -3,4 +3,4 @@
 #
 
 lib-y := claim.o enter.o exit.o finddevice.o getprop.o ofinit.o	\
-	 ofstdio.o read.o release.o write.o map.o
+	 ofstdio.o read.o release.o write.o map.o call_prom.o

arch/ppc/boot/of1275/call_prom.c

@@ -0,0 +1,74 @@
+/*
+ * 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 "of1275.h"
+#include <stdarg.h>
+
+int call_prom(const char *service, int nargs, int nret, ...)
+{
+	int i;
+	struct prom_args {
+		const char *service;
+		int nargs;
+		int nret;
+		unsigned int args[12];
+	} args;
+	va_list list;
+
+	args.service = service;
+	args.nargs = nargs;
+	args.nret = nret;
+
+	va_start(list, nret);
+	for (i = 0; i < nargs; i++)
+		args.args[i] = va_arg(list, unsigned int);
+	va_end(list);
+
+	for (i = 0; i < nret; i++)
+		args.args[nargs+i] = 0;
+
+	if (of_prom_entry(&args) < 0)
+		return -1;
+
+	return (nret > 0)? args.args[nargs]: 0;
+}
+
+int call_prom_ret(const char *service, int nargs, int nret,
+		  unsigned int *rets, ...)
+{
+	int i;
+	struct prom_args {
+		const char *service;
+		int nargs;
+		int nret;
+		unsigned int args[12];
+	} args;
+	va_list list;
+
+	args.service = service;
+	args.nargs = nargs;
+	args.nret = nret;
+
+	va_start(list, rets);
+	for (i = 0; i < nargs; i++)
+		args.args[i] = va_arg(list, unsigned int);
+	va_end(list);
+
+	for (i = 0; i < nret; i++)
+		args.args[nargs+i] = 0;
+
+	if (of_prom_entry(&args) < 0)
+		return -1;
+
+	if (rets != (void *) 0)
+		for (i = 1; i < nret; ++i)
+			rets[i-1] = args.args[nargs+i];
+
+	return (nret > 0)? args.args[nargs]: 0;
+}

arch/ppc/boot/of1275/claim.c

@@ -9,27 +9,84 @@
  */
 
 #include "of1275.h"
+#include "nonstdio.h"
 
-void *
+/*
-claim(unsigned int virt, unsigned int size, unsigned int align)
+ * Older OF's require that when claiming a specific range of addresses,
+ * we claim the physical space in the /memory node and the virtual
+ * space in the chosen mmu node, and then do a map operation to
+ * map virtual to physical.
+ */
+static int need_map = -1;
+static ihandle chosen_mmu;
+static phandle memory;
+
+/* returns true if s2 is a prefix of s1 */
+static int string_match(const char *s1, const char *s2)
 {
-    struct prom_args {
+	for (; *s2; ++s2)
-	char *service;
+		if (*s1++ != *s2)
-	int nargs;
+			return 0;
-	int nret;
+	return 1;
-	unsigned int virt;
+}
-	unsigned int size;
+
-	unsigned int align;
+static int check_of_version(void)
-	void *ret;
+{
-    } args;
+	phandle oprom, chosen;
-
+	char version[64];
-    args.service = "claim";
+
-    args.nargs = 3;
+	oprom = finddevice("/openprom");
-    args.nret = 1;
+	if (oprom == OF_INVALID_HANDLE)
-    args.virt = virt;
+		return 0;
-    args.size = size;
+	if (getprop(oprom, "model", version, sizeof(version)) <= 0)
-    args.align = align;
+		return 0;
-    args.ret = (void *) 0;
+	version[sizeof(version)-1] = 0;
-    (*of_prom_entry)(&args);
+	printf("OF version = '%s'\n", version);
-    return args.ret;
+	if (!string_match(version, "Open Firmware, 1.")
+	    && !string_match(version, "FirmWorks,3."))
+		return 0;
+	chosen = finddevice("/chosen");
+	if (chosen == OF_INVALID_HANDLE) {
+		chosen = finddevice("/chosen@0");
+		if (chosen == OF_INVALID_HANDLE) {
+			printf("no chosen\n");
+			return 0;
+		}
+	}
+	if (getprop(chosen, "mmu", &chosen_mmu, sizeof(chosen_mmu)) <= 0) {
+		printf("no mmu\n");
+		return 0;
+	}
+	memory = (ihandle) call_prom("open", 1, 1, "/memory");
+	if (memory == OF_INVALID_HANDLE) {
+		memory = (ihandle) call_prom("open", 1, 1, "/memory@0");
+		if (memory == OF_INVALID_HANDLE) {
+			printf("no memory node\n");
+			return 0;
+		}
+	}
+	printf("old OF detected\n");
+	return 1;
+}
+
+void *claim(unsigned int virt, unsigned int size, unsigned int align)
+{
+	int ret;
+	unsigned int result;
+
+	if (need_map < 0)
+		need_map = check_of_version();
+	if (align || !need_map)
+		return (void *) call_prom("claim", 3, 1, virt, size, align);
+	
+	ret = call_prom_ret("call-method", 5, 2, &result, "claim", memory,
+			    align, size, virt);
+	if (ret != 0 || result == -1)
+		return (void *) -1;
+	ret = call_prom_ret("call-method", 5, 2, &result, "claim", chosen_mmu,
+			    align, size, virt);
+	/* 0x12 == coherent + read/write */
+	ret = call_prom("call-method", 6, 1, "map", chosen_mmu,
+			0x12, size, virt, virt);
+	return virt;
 }

arch/ppc/boot/of1275/finddevice.c

@@ -10,22 +10,7 @@
 
 #include "of1275.h"
 
-phandle
+phandle finddevice(const char *name)
-finddevice(const char *name)
 {
-    struct prom_args {
+	return (phandle) call_prom("finddevice", 1, 1, name);
-	char *service;
-	int nargs;
-	int nret;
-	const char *devspec;
-	phandle device;
-    } args;
-
-    args.service = "finddevice";
-    args.nargs = 1;
-    args.nret = 1;
-    args.devspec = name;
-    args.device = OF_INVALID_HANDLE;
-    (*of_prom_entry)(&args);
-    return args.device;
 }

arch/ppc/boot/openfirmware/Makefile

@@ -80,8 +80,7 @@ $(obj)/note: $(utils)/mknote FORCE
 	$(call if_changed,mknote)
 
 
-$(obj)/coffcrt0.o: EXTRA_AFLAGS := -traditional -DXCOFF
+$(obj)/coffcrt0.o: EXTRA_AFLAGS := -DXCOFF
-$(obj)/crt0.o:     EXTRA_AFLAGS := -traditional
 targets += coffcrt0.o crt0.o
 $(obj)/coffcrt0.o $(obj)/crt0.o: $(common)/crt0.S FORCE
 	$(call if_changed_dep,as_o_S)

arch/ppc/kernel/Makefile

@@ -12,7 +12,7 @@ extra-$(CONFIG_6xx)		+= idle_6xx.o
 extra-$(CONFIG_POWER4)		+= idle_power4.o
 extra-y				+= vmlinux.lds
 
-obj-y				:= entry.o traps.o irq.o idle.o time.o misc.o \
+obj-y				:= entry.o traps.o idle.o time.o misc.o \
 					process.o align.o \
 					setup.o \
 					ppc_htab.o
@@ -38,8 +38,7 @@ endif
 # These are here while we do the architecture merge
 
 else
-obj-y				:= irq.o idle.o \
+obj-y				:= idle.o align.o
-					align.o
 obj-$(CONFIG_6xx)		+= l2cr.o cpu_setup_6xx.o
 obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o
 obj-$(CONFIG_MODULES)		+= module.o

arch/ppc/kernel/head_booke.h

@@ -358,6 +358,6 @@ label:
 	NORMAL_EXCEPTION_PROLOG;					      \
 	bne	load_up_fpu;		/* if from user, just load it up */   \
 	addi	r3,r1,STACK_FRAME_OVERHEAD;				      \
-	EXC_XFER_EE_LITE(0x800, KernelFP)
+	EXC_XFER_EE_LITE(0x800, kernel_fp_unavailable_exception)
 
 #endif /* __HEAD_BOOKE_H__ */

arch/ppc/kernel/irq.c

@@ -1,165 +0,0 @@
-/*
- *  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 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>
-
-#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>
-
-#define NR_MASK_WORDS	((NR_IRQS + 31) / 32)
-
-extern atomic_t ipi_recv;
-extern atomic_t ipi_sent;
-
-#define MAXCOUNT 10000000
-
-int ppc_spurious_interrupts = 0;
-struct irqaction *ppc_irq_action[NR_IRQS];
-unsigned long ppc_cached_irq_mask[NR_MASK_WORDS];
-unsigned long ppc_lost_interrupts[NR_MASK_WORDS];
-atomic_t ppc_n_lost_interrupts;
-
-#ifdef CONFIG_TAU_INT
-extern int tau_initialized;
-extern int tau_interrupts(int);
-#endif
-
-int show_interrupts(struct seq_file *p, void *v)
-{
-	int i = *(loff_t *) v, j;
-	struct irqaction * action;
-	unsigned long flags;
-
-	if (i == 0) {
-		seq_puts(p, "           ");
-		for (j=0; j<NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "CPU%d       ", j);
-		seq_putc(p, '\n');
-	}
-
-	if (i < NR_IRQS) {
-		spin_lock_irqsave(&irq_desc[i].lock, flags);
-		action = irq_desc[i].action;
-		if ( !action || !action->handler )
-			goto skip;
-		seq_printf(p, "%3d: ", i);
-#ifdef CONFIG_SMP
-		for (j = 0; j < NR_CPUS; j++)
-			if (cpu_online(j))
-				seq_printf(p, "%10u ",
-					   kstat_cpu(j).irqs[i]);
-#else
-		seq_printf(p, "%10u ", kstat_irqs(i));
-#endif /* CONFIG_SMP */
-		if (irq_desc[i].handler)
-			seq_printf(p, " %s ", irq_desc[i].handler->typename);
-		else
-			seq_puts(p, "  None      ");
-		seq_printf(p, "%s", (irq_desc[i].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(&irq_desc[i].lock, flags);
-	} else if (i == NR_IRQS) {
-#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
-		seq_printf(p, "BAD: %10u\n", ppc_spurious_interrupts);
-	}
-	return 0;
-}
-
-void do_IRQ(struct pt_regs *regs)
-{
-	int irq, first = 1;
-        irq_enter();
-
-	/*
-	 * 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.  But the first time
-	 * through the loop this means there wasn't and IRQ pending.
-	 * The value -2 is for buggy hardware and means that this IRQ
-	 * has already been handled. -- Tom
-	 */
-	while ((irq = ppc_md.get_irq(regs)) >= 0) {
-		__do_IRQ(irq, regs);
-		first = 0;
-	}
-	if (irq != -2 && first)
-		/* That's not SMP safe ... but who cares ? */
-		ppc_spurious_interrupts++;
-        irq_exit();
-}
-
-void __init init_IRQ(void)
-{
-	ppc_md.init_IRQ();
-}

arch/ppc/kernel/misc.S

@@ -497,9 +497,9 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SPLIT_ID_CACHE)
  * and invalidate the corresponding instruction cache blocks.
  * This is a no-op on the 601.
  *
- * flush_icache_range(unsigned long start, unsigned long stop)
+ * __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)

arch/ppc/kernel/ppc_ksyms.c

@@ -46,6 +46,7 @@
 #include <asm/btext.h>
 #include <asm/div64.h>
 #include <asm/xmon.h>
+#include <asm/signal.h>
 
 #ifdef  CONFIG_8xx
 #include <asm/commproc.h>
@@ -57,7 +58,6 @@ 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);
 
@@ -78,7 +78,6 @@ EXPORT_SYMBOL(program_check_exception);
 EXPORT_SYMBOL(single_step_exception);
 EXPORT_SYMBOL(sys_sigreturn);
 EXPORT_SYMBOL(ppc_n_lost_interrupts);
-EXPORT_SYMBOL(ppc_lost_interrupts);
 
 EXPORT_SYMBOL(ISA_DMA_THRESHOLD);
 EXPORT_SYMBOL(DMA_MODE_READ);
@@ -176,6 +175,7 @@ EXPORT_SYMBOL(pci_bus_to_phys);
 #endif /* CONFIG_PCI */
 
 #ifdef CONFIG_NOT_COHERENT_CACHE
+extern void flush_dcache_all(void);
 EXPORT_SYMBOL(flush_dcache_all);
 #endif
 
@@ -217,9 +217,6 @@ EXPORT_SYMBOL(adb_try_handler_change);
 EXPORT_SYMBOL(cuda_request);
 EXPORT_SYMBOL(cuda_poll);
 #endif /* CONFIG_ADB_CUDA */
-#ifdef CONFIG_PPC_MULTIPLATFORM
-EXPORT_SYMBOL(_machine);
-#endif
 #ifdef CONFIG_PPC_PMAC
 EXPORT_SYMBOL(sys_ctrler);
 EXPORT_SYMBOL(pmac_newworld);

arch/ppc/kernel/setup.c

@@ -76,6 +76,7 @@ unsigned int DMA_MODE_WRITE;
 
 #ifdef CONFIG_PPC_MULTIPLATFORM
 int _machine = 0;
+EXPORT_SYMBOL(_machine);
 
 extern void prep_init(unsigned long r3, unsigned long r4,
 		unsigned long r5, unsigned long r6, unsigned long r7);

arch/ppc/platforms/pmac_pic.c

@@ -75,6 +75,9 @@ 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).

arch/ppc/platforms/prep_setup.c

@@ -61,6 +61,15 @@
 #include <asm/pci-bridge.h>
 #include <asm/todc.h>
 
+/* prep registers for L2 */
+#define CACHECRBA       0x80000823      /* Cache configuration register address */
+#define L2CACHE_MASK	0x03	/* Mask for 2 L2 Cache bits */
+#define L2CACHE_512KB	0x00	/* 512KB */
+#define L2CACHE_256KB	0x01	/* 256KB */
+#define L2CACHE_1MB	0x02	/* 1MB */
+#define L2CACHE_NONE	0x03	/* NONE */
+#define L2CACHE_PARITY  0x08    /* Mask for L2 Cache Parity Protected bit */
+
 TODC_ALLOC();
 
 unsigned char ucSystemType;

arch/ppc64/Kconfig

@@ -297,6 +297,10 @@ 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

arch/ppc64/boot/addRamDisk.c

@@ -5,11 +5,59 @@
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <string.h>
+#include <elf.h>
 
 #define ElfHeaderSize  (64 * 1024)
 #define ElfPages  (ElfHeaderSize / 4096)
 #define KERNELBASE (0xc000000000000000)
+#define _ALIGN_UP(addr,size)	(((addr)+((size)-1))&(~((size)-1)))
 
+struct addr_range {
+	unsigned long long addr;
+	unsigned long memsize;
+	unsigned long offset;
+};
+
+static int check_elf64(void *p, int size, struct addr_range *r)
+{
+	Elf64_Ehdr *elf64 = p;
+	Elf64_Phdr *elf64ph;
+
+	if (elf64->e_ident[EI_MAG0] != ELFMAG0 ||
+	    elf64->e_ident[EI_MAG1] != ELFMAG1 ||
+	    elf64->e_ident[EI_MAG2] != ELFMAG2 ||
+	    elf64->e_ident[EI_MAG3] != ELFMAG3 ||
+	    elf64->e_ident[EI_CLASS] != ELFCLASS64 ||
+	    elf64->e_ident[EI_DATA] != ELFDATA2MSB ||
+	    elf64->e_type != ET_EXEC || elf64->e_machine != EM_PPC64)
+		return 0;
+
+	if ((elf64->e_phoff + sizeof(Elf64_Phdr)) > size)
+		return 0;
+
+	elf64ph = (Elf64_Phdr *) ((unsigned long)elf64 +
+				  (unsigned long)elf64->e_phoff);
+
+	r->memsize = (unsigned long)elf64ph->p_memsz;
+	r->offset = (unsigned long)elf64ph->p_offset;
+	r->addr = (unsigned long long)elf64ph->p_vaddr;
+
+#ifdef DEBUG
+	printf("PPC64 ELF file, ph:\n");
+	printf("p_type   0x%08x\n", elf64ph->p_type);
+	printf("p_flags  0x%08x\n", elf64ph->p_flags);
+	printf("p_offset 0x%016llx\n", elf64ph->p_offset);
+	printf("p_vaddr  0x%016llx\n", elf64ph->p_vaddr);
+	printf("p_paddr  0x%016llx\n", elf64ph->p_paddr);
+	printf("p_filesz 0x%016llx\n", elf64ph->p_filesz);
+	printf("p_memsz  0x%016llx\n", elf64ph->p_memsz);
+	printf("p_align  0x%016llx\n", elf64ph->p_align);
+	printf("... skipping 0x%08lx bytes of ELF header\n",
+	       (unsigned long)elf64ph->p_offset);
+#endif
+
+	return 64;
+}
 void get4k(FILE *file, char *buf )
 {
 	unsigned j;
@@ -34,97 +82,92 @@ void death(const char *msg, FILE *fdesc, const char *fname)
 int main(int argc, char **argv)
 {
 	char inbuf[4096];
-	FILE *ramDisk = NULL;
+	struct addr_range vmlinux;
-	FILE *sysmap = NULL;
+	FILE *ramDisk;
-	FILE *inputVmlinux = NULL;
+	FILE *inputVmlinux;
-	FILE *outputVmlinux = NULL;
+	FILE *outputVmlinux;
-  
-	unsigned i = 0;
-	unsigned long ramFileLen = 0;
-	unsigned long ramLen = 0;
-	unsigned long roundR = 0;
-  
-	unsigned long sysmapFileLen = 0;
-	unsigned long sysmapLen = 0;
-	unsigned long sysmapPages = 0;
-	char* ptr_end = NULL; 
-	unsigned long offset_end = 0;
 
-	unsigned long kernelLen = 0;
+	char *rd_name, *lx_name, *out_name;
-	unsigned long actualKernelLen = 0;
+
-	unsigned long round = 0;
+	size_t i;
-	unsigned long roundedKernelLen = 0;
+	unsigned long ramFileLen;
-	unsigned long ramStartOffs = 0;
+	unsigned long ramLen;
-	unsigned long ramPages = 0;
+	unsigned long roundR;
-	unsigned long roundedKernelPages = 0;
+	unsigned long offset_end;
-	unsigned long hvReleaseData = 0;
+
+	unsigned long kernelLen;
+	unsigned long actualKernelLen;
+	unsigned long round;
+	unsigned long roundedKernelLen;
+	unsigned long ramStartOffs;
+	unsigned long ramPages;
+	unsigned long roundedKernelPages;
+	unsigned long hvReleaseData;
 	u_int32_t eyeCatcher = 0xc8a5d9c4;
-	unsigned long naca = 0;
+	unsigned long naca;
-	unsigned long xRamDisk = 0;
+	unsigned long xRamDisk;
-	unsigned long xRamDiskSize = 0;
+	unsigned long xRamDiskSize;
-	long padPages = 0;
+	long padPages;
   
   
 	if (argc < 2) {
 		fprintf(stderr, "Name of RAM disk file missing.\n");
 		exit(1);
 	}
+	rd_name = argv[1];
 
 	if (argc < 3) {
-		fprintf(stderr, "Name of System Map input file is missing.\n");
-		exit(1);
-	}
-  
-	if (argc < 4) {
 		fprintf(stderr, "Name of vmlinux file missing.\n");
 		exit(1);
 	}
+	lx_name = argv[2];
 
-	if (argc < 5) {
+	if (argc < 4) {
 		fprintf(stderr, "Name of vmlinux output file missing.\n");
 		exit(1);
 	}
+	out_name = argv[3];
 
 
-	ramDisk = fopen(argv[1], "r");
+	ramDisk = fopen(rd_name, "r");
 	if ( ! ramDisk ) {
-		fprintf(stderr, "RAM disk file \"%s\" failed to open.\n", argv[1]);
+		fprintf(stderr, "RAM disk file \"%s\" failed to open.\n", rd_name);
 		exit(1);
 	}
 
-	sysmap = fopen(argv[2], "r");
+	inputVmlinux = fopen(lx_name, "r");
-	if ( ! sysmap ) {
-		fprintf(stderr, "System Map file \"%s\" failed to open.\n", argv[2]);
-		exit(1);
-	}
-  
-	inputVmlinux = fopen(argv[3], "r");
 	if ( ! inputVmlinux ) {
-		fprintf(stderr, "vmlinux file \"%s\" failed to open.\n", argv[3]);
+		fprintf(stderr, "vmlinux file \"%s\" failed to open.\n", lx_name);
 		exit(1);
 	}
   
-	outputVmlinux = fopen(argv[4], "w+");
+	outputVmlinux = fopen(out_name, "w+");
 	if ( ! outputVmlinux ) {
-		fprintf(stderr, "output vmlinux file \"%s\" failed to open.\n", argv[4]);
+		fprintf(stderr, "output vmlinux file \"%s\" failed to open.\n", out_name);
 		exit(1);
 	}
-  
+
-  
+	i = fread(inbuf, 1, sizeof(inbuf), inputVmlinux);
-  
+	if (i != sizeof(inbuf)) {
+		fprintf(stderr, "can not read vmlinux file %s: %u\n", lx_name, i);
+		exit(1);
+	}
+
+	i = check_elf64(inbuf, sizeof(inbuf), &vmlinux);
+	if (i == 0) {
+		fprintf(stderr, "You must have a linux kernel specified as argv[2]\n");
+		exit(1);
+	}
+
 	/* Input Vmlinux file */
 	fseek(inputVmlinux, 0, SEEK_END);
 	kernelLen = ftell(inputVmlinux);
 	fseek(inputVmlinux, 0, SEEK_SET);
-	printf("kernel file size = %d\n", kernelLen);
+	printf("kernel file size = %lu\n", kernelLen);
-	if ( kernelLen == 0 ) {
-		fprintf(stderr, "You must have a linux kernel specified as argv[3]\n");
-		exit(1);
-	}
 
 	actualKernelLen = kernelLen - ElfHeaderSize;
 
-	printf("actual kernel length (minus ELF header) = %d\n", actualKernelLen);
+	printf("actual kernel length (minus ELF header) = %lu\n", actualKernelLen);
 
 	round = actualKernelLen % 4096;
 	roundedKernelLen = actualKernelLen;
@@ -134,39 +177,7 @@ int main(int argc, char **argv)
 	roundedKernelPages = roundedKernelLen / 4096;
 	printf("Vmlinux pages to copy = %ld/0x%lx \n", roundedKernelPages, roundedKernelPages);
 
-
+	offset_end = _ALIGN_UP(vmlinux.memsize, 4096);
-
-	/* Input System Map file */
-	/* (needs to be processed simply to determine if we need to add pad pages due to the static variables not being included in the vmlinux) */
-	fseek(sysmap, 0, SEEK_END);
-	sysmapFileLen = ftell(sysmap);
-	fseek(sysmap, 0, SEEK_SET);
-	printf("%s file size = %ld/0x%lx \n", argv[2], sysmapFileLen, sysmapFileLen);
-
-	sysmapLen = sysmapFileLen;
-
-	roundR = 4096 - (sysmapLen % 4096);
-	if (roundR) {
-		printf("Rounding System Map file up to a multiple of 4096, adding %ld/0x%lx \n", roundR, roundR);
-		sysmapLen += roundR;
-	}
-	printf("Rounded System Map size is %ld/0x%lx \n", sysmapLen, sysmapLen);
-  
-	/* Process the Sysmap file to determine where _end is */
-	sysmapPages = sysmapLen / 4096;
-	/* read the whole file line by line, expect that it doesn't fail */
-	while ( fgets(inbuf, 4096, sysmap) )  ;
-	/* search for _end in the last page of the system map */
-	ptr_end = strstr(inbuf, " _end");
-	if (!ptr_end) {
-		fprintf(stderr, "Unable to find _end in the sysmap file \n");
-		fprintf(stderr, "inbuf: \n");
-		fprintf(stderr, "%s \n", inbuf);
-		exit(1);
-	}
-	printf("Found _end in the last page of the sysmap - backing up 10 characters it looks like %s", ptr_end-10);
-	/* convert address of _end in system map to hex offset. */
-	offset_end = (unsigned int)strtol(ptr_end-10, NULL, 16);
 	/* calc how many pages we need to insert between the vmlinux and the start of the ram disk */
 	padPages = offset_end/4096 - roundedKernelPages;
 
@@ -194,7 +205,7 @@ int main(int argc, char **argv)
 	fseek(ramDisk, 0, SEEK_END);
 	ramFileLen = ftell(ramDisk);
 	fseek(ramDisk, 0, SEEK_SET);
-	printf("%s file size = %ld/0x%lx \n", argv[1], ramFileLen, ramFileLen);
+	printf("%s file size = %ld/0x%lx \n", rd_name, ramFileLen, ramFileLen);
 
 	ramLen = ramFileLen;
 
@@ -248,19 +259,19 @@ int main(int argc, char **argv)
 	/* fseek to the hvReleaseData pointer */
 	fseek(outputVmlinux, ElfHeaderSize + 0x24, SEEK_SET);
 	if (fread(&hvReleaseData, 4, 1, outputVmlinux) != 1) {
-		death("Could not read hvReleaseData pointer\n", outputVmlinux, argv[4]);
+		death("Could not read hvReleaseData pointer\n", outputVmlinux, out_name);
 	}
 	hvReleaseData = ntohl(hvReleaseData); /* Convert to native int */
-	printf("hvReleaseData is at %08x\n", hvReleaseData);
+	printf("hvReleaseData is at %08lx\n", hvReleaseData);
 
 	/* fseek to the hvReleaseData */
 	fseek(outputVmlinux, ElfHeaderSize + hvReleaseData, SEEK_SET);
 	if (fread(inbuf, 0x40, 1, outputVmlinux) != 1) {
-		death("Could not read hvReleaseData\n", outputVmlinux, argv[4]);
+		death("Could not read hvReleaseData\n", outputVmlinux, out_name);
 	}
 	/* Check hvReleaseData sanity */
 	if (memcmp(inbuf, &eyeCatcher, 4) != 0) {
-		death("hvReleaseData is invalid\n", outputVmlinux, argv[4]);
+		death("hvReleaseData is invalid\n", outputVmlinux, out_name);
 	}
 	/* Get the naca pointer */
 	naca = ntohl(*((u_int32_t*) &inbuf[0x0C])) - KERNELBASE;
@@ -269,13 +280,13 @@ int main(int argc, char **argv)
 	/* fseek to the naca */
 	fseek(outputVmlinux, ElfHeaderSize + naca, SEEK_SET);
 	if (fread(inbuf, 0x18, 1, outputVmlinux) != 1) {
-		death("Could not read naca\n", outputVmlinux, argv[4]);
+		death("Could not read naca\n", outputVmlinux, out_name);
 	}
 	xRamDisk = ntohl(*((u_int32_t *) &inbuf[0x0c]));
 	xRamDiskSize = ntohl(*((u_int32_t *) &inbuf[0x14]));
 	/* Make sure a RAM disk isn't already present */
 	if ((xRamDisk != 0) || (xRamDiskSize != 0)) {
-		death("RAM disk is already attached to this kernel\n", outputVmlinux, argv[4]);
+		death("RAM disk is already attached to this kernel\n", outputVmlinux, out_name);
 	}
 	/* Fill in the values */
 	*((u_int32_t *) &inbuf[0x0c]) = htonl(ramStartOffs);
@@ -285,15 +296,15 @@ int main(int argc, char **argv)
 	fflush(outputVmlinux);
 	fseek(outputVmlinux, ElfHeaderSize + naca, SEEK_SET);
 	if (fwrite(inbuf, 0x18, 1, outputVmlinux) != 1) {
-		death("Could not write naca\n", outputVmlinux, argv[4]);
+		death("Could not write naca\n", outputVmlinux, out_name);
 	}
-	printf("Ram Disk of 0x%lx pages is attached to the kernel at offset 0x%08x\n",
+	printf("Ram Disk of 0x%lx pages is attached to the kernel at offset 0x%08lx\n",
 	       ramPages, ramStartOffs);
 
 	/* Done */
 	fclose(outputVmlinux);
 	/* Set permission to executable */
-	chmod(argv[4], S_IRUSR|S_IWUSR|S_IXUSR|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH);
+	chmod(out_name, S_IRUSR|S_IWUSR|S_IXUSR|S_IRGRP|S_IXGRP|S_IROTH|S_IXOTH);
 
 	return 0;
 }

arch/ppc64/kernel/Makefile

@@ -11,12 +11,11 @@ obj-y               :=	misc.o prom.o
 
 endif
 
-obj-y               +=	irq.o idle.o dma.o \
+obj-y               +=	idle.o dma.o \
-			align.o pacaData.o \
+			align.o \
-			udbg.o ioctl32.o \
+			udbg.o \
 			rtc.o \
-			cpu_setup_power4.o \
+			iommu.o vdso.o
-			iommu.o sysfs.o vdso.o firmware.o
 obj-y += vdso32/ vdso64/
 
 pci-obj-$(CONFIG_PPC_MULTIPLATFORM)	+= pci_dn.o pci_direct_iommu.o
@@ -31,15 +30,10 @@ endif
 obj-$(CONFIG_PPC_PSERIES) += udbg_16550.o
 
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o
-obj-$(CONFIG_EEH)		+= eeh.o
-obj-$(CONFIG_PROC_FS)		+= proc_ppc64.o
 obj-$(CONFIG_MODULES)		+= module.o
 ifneq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_MODULES)		+= ppc_ksyms.o
 endif
-obj-$(CONFIG_PPC_RTAS)		+= rtas_pci.o
-obj-$(CONFIG_SCANLOG)		+= scanlog.o
-obj-$(CONFIG_LPARCFG)		+= lparcfg.o
 obj-$(CONFIG_HVC_CONSOLE)	+= hvconsole.o
 ifneq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_BOOTX_TEXT)	+= btext.o
@@ -52,8 +46,6 @@ obj-$(CONFIG_PPC_MAPLE)		+= udbg_16550.o
 
 obj-$(CONFIG_KPROBES)		+= kprobes.o
 
-CFLAGS_ioctl32.o += -Ifs/
-
 ifneq ($(CONFIG_PPC_MERGE),y)
 ifeq ($(CONFIG_PPC_ISERIES),y)
 arch/ppc64/kernel/head.o: arch/powerpc/kernel/lparmap.s

arch/ppc64/kernel/asm-offsets.c

@@ -74,7 +74,6 @@ 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 */

arch/ppc64/kernel/head.S

@@ -28,7 +28,6 @@
 #include <asm/processor.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>
@@ -1701,21 +1700,9 @@ _GLOBAL(__secondary_start)
 	HMT_MEDIUM			/* Set thread priority to MEDIUM */
 
 	ld	r2,PACATOC(r13)
-	li	r6,0
-	stb	r6,PACAPROCENABLED(r13)
 
-#ifndef CONFIG_PPC_ISERIES
+	/* Do early setup for that CPU */
-	/* Initialize the page table pointer register. */
+	bl	.early_setup_secondary
-	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
 
 	/* Initialize the kernel stack.  Just a repeat for iSeries.	 */
 	LOADADDR(r3,current_set)
@@ -1724,37 +1711,6 @@ 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
 	li	r7,0
 	mtlr	r7
 
@@ -1896,40 +1852,6 @@ _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 */
-	sub	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

arch/ppc64/kernel/idle.c

@@ -26,7 +26,6 @@
 #include <asm/processor.h>
 #include <asm/cputable.h>
 #include <asm/time.h>
-#include <asm/systemcfg.h>
 #include <asm/machdep.h>
 #include <asm/smp.h>
 

arch/ppc64/kernel/misc.S

@@ -78,12 +78,12 @@ _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)
+	stdu	r1,THREAD_SIZE-112(r5)
-	mr	r1,r6
+	mr	r1,r5
-	bl	.handle_IRQ_event
+	bl	.__do_IRQ
 	ld	r1,0(r1)
 	ld	r0,16(r1)
 	mtlr	r0

arch/ppc64/kernel/nvram.c

@@ -31,7 +31,6 @@
 #include <asm/rtas.h>
 #include <asm/prom.h>
 #include <asm/machdep.h>
-#include <asm/systemcfg.h>
 
 #undef DEBUG_NVRAM
 
@@ -167,7 +166,7 @@ static int dev_nvram_ioctl(struct inode *inode, struct file *file,
 	case IOC_NVRAM_GET_OFFSET: {
 		int part, offset;
 
-		if (systemcfg->platform != PLATFORM_POWERMAC)
+		if (_machine != PLATFORM_POWERMAC)
 			return -EINVAL;
 		if (copy_from_user(&part, (void __user*)arg, sizeof(part)) != 0)
 			return -EFAULT;
@@ -450,7 +449,7 @@ static int nvram_setup_partition(void)
 	 * in our nvram, as Apple defined partitions use pretty much
 	 * all of the space
 	 */
-	if (systemcfg->platform == PLATFORM_POWERMAC)
+	if (_machine == PLATFORM_POWERMAC)
 		return -ENOSPC;
 
 	/* see if we have an OS partition that meets our needs.

arch/ppc64/kernel/pci.c

@@ -548,6 +548,11 @@ static int __init pcibios_init(void)
 	if (ppc64_isabridge_dev != NULL)
 		printk("ISA bridge at %s\n", pci_name(ppc64_isabridge_dev));
 
+#ifdef CONFIG_PPC_MULTIPLATFORM
+	/* map in PCI I/O space */
+	phbs_remap_io();
+#endif
+
 	printk("PCI: Probing PCI hardware done\n");
 
 	return 0;
@@ -1277,12 +1282,9 @@ long sys_pciconfig_iobase(long which, unsigned long in_bus,
 	 * G5 machines... So when something asks for bus 0 io base
 	 * (bus 0 is HT root), we return the AGP one instead.
 	 */
-#ifdef CONFIG_PPC_PMAC
+	if (machine_is_compatible("MacRISC4"))
-	if (systemcfg->platform == PLATFORM_POWERMAC &&
-	    machine_is_compatible("MacRISC4"))
 		if (in_bus == 0)
 			in_bus = 0xf0;
-#endif /* CONFIG_PPC_PMAC */
 
 	/* That syscall isn't quite compatible with PCI domains, but it's
 	 * used on pre-domains setup. We return the first match

arch/ppc64/kernel/pci_dn.c

@@ -43,7 +43,7 @@ static void * __devinit update_dn_pci_info(struct device_node *dn, void *data)
 	u32 *regs;
 	struct pci_dn *pdn;
 
-	if (phb->is_dynamic)
+	if (mem_init_done)
 		pdn = kmalloc(sizeof(*pdn), GFP_KERNEL);
 	else
 		pdn = alloc_bootmem(sizeof(*pdn));
@@ -120,6 +120,14 @@ void *traverse_pci_devices(struct device_node *start, traverse_func pre,
 	return NULL;
 }
 
+/** 
+ * pci_devs_phb_init_dynamic - setup pci devices under this PHB
+ * phb: pci-to-host bridge (top-level bridge connecting to cpu)
+ *
+ * This routine is called both during boot, (before the memory
+ * subsystem is set up, before kmalloc is valid) and during the 
+ * dynamic lpar operation of adding a PHB to a running system.
+ */
 void __devinit pci_devs_phb_init_dynamic(struct pci_controller *phb)
 {
 	struct device_node * dn = (struct device_node *) phb->arch_data;
@@ -201,9 +209,14 @@ static struct notifier_block pci_dn_reconfig_nb = {
 	.notifier_call = pci_dn_reconfig_notifier,
 };
 
-/*
+/** 
- * Actually initialize the phbs.
+ * pci_devs_phb_init - Initialize phbs and pci devs under them.
- * The buswalk on this phb has not happened yet.
+ * 
+ * This routine walks over all phb's (pci-host bridges) on the
+ * system, and sets up assorted pci-related structures 
+ * (including pci info in the device node structs) for each
+ * pci device found underneath.  This routine runs once,
+ * early in the boot sequence.
  */
 void __init pci_devs_phb_init(void)
 {

arch/ppc64/kernel/prom.c

@@ -318,7 +318,7 @@ static int __devinit finish_node_interrupts(struct device_node *np,
 		}
 
 		/* 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;
@@ -1065,7 +1065,7 @@ 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;
-	systemcfg->platform = *prop;
+	_machine = *prop;
 
 	/* check if iommu is forced on or off */
 	if (of_get_flat_dt_prop(node, "linux,iommu-off", NULL) != NULL)
@@ -1230,11 +1230,8 @@ 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();
-	systemcfg->physicalMemorySize = lmb_phys_mem_size();
 	lmb_reserve(0, __pa(klimit));
 
-	DBG("Phys. mem: %lx\n", systemcfg->physicalMemorySize);
-
 	/* Reserve LMB regions used by kernel, initrd, dt, etc... */
 	early_reserve_mem();
 
@@ -1753,7 +1750,7 @@ 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 */

arch/ppc64/kernel/prom_init.c

@@ -1934,7 +1934,8 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, unsigned long
 	/*
 	 * 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();
 
 	/*

arch/ppc64/kernel/vdso.c

@@ -34,6 +34,7 @@
 #include <asm/machdep.h>
 #include <asm/cputable.h>
 #include <asm/sections.h>
+#include <asm/systemcfg.h>
 #include <asm/vdso.h>
 
 #undef DEBUG
@@ -179,7 +180,7 @@ static struct page * vdso_vma_nopage(struct vm_area_struct * vma,
 	 * Last page is systemcfg.
 	 */
 	if ((vma->vm_end - address) <= PAGE_SIZE)
-		pg = virt_to_page(systemcfg);
+		pg = virt_to_page(_systemcfg);
 	else
 		pg = virt_to_page(vbase + offset);
 
@@ -604,7 +605,7 @@ void __init vdso_init(void)
 		get_page(pg);
 	}
 
-	get_page(virt_to_page(systemcfg));
+	get_page(virt_to_page(_systemcfg));
 }
 
 int in_gate_area_no_task(unsigned long addr)

drivers/net/fs_enet/fs_enet-main.c

@@ -37,6 +37,7 @@
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/fs.h>
+#include <linux/platform_device.h>
 
 #include <linux/vmalloc.h>
 #include <asm/pgtable.h>

drivers/net/fs_enet/mac-fcc.c

@@ -34,6 +34,7 @@
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/fs.h>
+#include <linux/platform_device.h>
 
 #include <asm/immap_cpm2.h>
 #include <asm/mpc8260.h>

drivers/net/fs_enet/mac-fec.c

@@ -34,6 +34,7 @@
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/fs.h>
+#include <linux/platform_device.h>
 
 #include <asm/irq.h>
 #include <asm/uaccess.h>

drivers/net/fs_enet/mac-scc.c

@@ -34,6 +34,7 @@
 #include <linux/ethtool.h>
 #include <linux/bitops.h>
 #include <linux/fs.h>
+#include <linux/platform_device.h>
 
 #include <asm/irq.h>
 #include <asm/uaccess.h>

drivers/pci/hotplug/rpadlpar_core.c

@@ -306,7 +306,7 @@ static int dlpar_add_phb(char *drc_name, struct device_node *dn)
 {
 	struct pci_controller *phb;
 
-	if (PCI_DN(dn)->phb) {
+	if (PCI_DN(dn) && PCI_DN(dn)->phb) {
 		/* PHB already exists */
 		return -EINVAL;
 	}