Merge samba.org:/scratch/anton/linux-2.5

into samba.org:/scratch/anton/tmp3

arch/ppc64/kernel/Makefile

@@ -9,7 +9,7 @@ obj-y               :=	setup.o entry.o traps.o irq.o idle.o \
 			align.o semaphore.o bitops.o stab.o htab.o pacaData.o \
 			udbg.o binfmt_elf32.o sys_ppc32.o ioctl32.o \
 			ptrace32.o signal32.o pmc.o rtc.o init_task.o \
-			lmb.o pci.o pci_dn.o pci_dma.o
+			lmb.o pci.o pci_dn.o pci_dma.o cputable.o
 
 obj-$(CONFIG_PPC_ISERIES) += iSeries_pci.o       iSeries_pci_reset.o \
 			     iSeries_IoMmTable.o iSeries_irq.o \
@@ -19,11 +19,11 @@ obj-$(CONFIG_PPC_ISERIES) += iSeries_pci.o       iSeries_pci_reset.o \
 			     mf.o HvLpEvent.o iSeries_proc.o 
 
 obj-$(CONFIG_PPC_PSERIES) += pSeries_pci.o pSeries_lpar.o pSeries_hvCall.o \
-			     eeh.o rtasd.o nvram.o
+			     eeh.o rtasd.o nvram.o ras.o
 
 # Change this to pSeries only once we've got iSeries up to date
 obj-y			  += open_pic.o xics.o pSeries_htab.o rtas.o \
-			     chrp_setup.o i8259.o ras.o prom.o 
+			     chrp_setup.o i8259.o prom.o 
 
 obj-$(CONFIG_PROC_FS)		+= proc_ppc64.o
 obj-$(CONFIG_RTAS_FLASH)	+= rtas_flash.o
@@ -31,3 +31,5 @@ obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_MODULES)		+= module.o ppc_ksyms.o
 obj-$(CONFIG_PPC_RTAS)		+= rtas-proc.o
 obj-$(CONFIG_SCANLOG)		+= scanlog.o
+
+CFLAGS_ioctl32.o += -Ifs/

arch/ppc64/kernel/align.c

@@ -20,6 +20,7 @@
 #include <asm/uaccess.h>
 #include <asm/system.h>
 #include <asm/cache.h>
+#include <asm/cputable.h>
 
 void disable_kernel_fp(void); /* asm function from head.S */
 
@@ -238,12 +239,11 @@ fix_alignment(struct pt_regs *regs)
 
 	dsisr = regs->dsisr;
 
-	/* Power4 doesn't set DSISR for an alignment interrupt */
-	if (!cpu_alignexc_sets_dsisr()) {
+	if (cur_cpu_spec->cpu_features & CPU_FTR_NODSISRALIGN) {
 	    unsigned int real_instr;
 	    if (__get_user(real_instr, (unsigned int *)regs->nip))
 		return 0;
-		dsisr = make_dsisr(real_instr);
+	    dsisr = make_dsisr(*((unsigned *)regs->nip));
 	}
 
 	/* extract the operation and registers from the dsisr */

arch/ppc64/kernel/asm-offsets.c

@@ -34,6 +34,7 @@
 #include <asm/iSeries/HvLpEvent.h>
 #include <asm/prom.h>
 #include <asm/rtas.h>
+#include <asm/cputable.h>
 
 #define DEFINE(sym, val) \
 	asm volatile("\n->" #sym " %0 " #val : : "i" (val))
@@ -159,5 +160,12 @@ int main(void)
 	DEFINE(CLONE_VM, CLONE_VM);
 	DEFINE(CLONE_UNTRACED, CLONE_UNTRACED);
 
+	/* About the CPU features table */
+	DEFINE(CPU_SPEC_ENTRY_SIZE, sizeof(struct cpu_spec));
+	DEFINE(CPU_SPEC_PVR_MASK, offsetof(struct cpu_spec, pvr_mask));
+	DEFINE(CPU_SPEC_PVR_VALUE, offsetof(struct cpu_spec, pvr_value));
+	DEFINE(CPU_SPEC_FEATURES, offsetof(struct cpu_spec, cpu_features));
+	DEFINE(CPU_SPEC_SETUP, offsetof(struct cpu_spec, cpu_setup));
+
 	return 0;
 }

arch/ppc64/kernel/chrp_setup.c

@@ -62,13 +62,13 @@
 #include "open_pic.h"
 #include <asm/xics.h>
 #include <asm/ppcdebug.h>
+#include <asm/cputable.h>
 
 extern volatile unsigned char *chrp_int_ack_special;
 
 void chrp_progress(char *, unsigned short);
 
 extern void openpic_init_IRQ(void);
-extern void init_ras_IRQ(void);
 
 extern void find_and_init_phbs(void);
 
@@ -238,7 +238,6 @@ chrp_init(unsigned long r3, unsigned long r4, unsigned long r5,
 		ppc_md.init_IRQ       = xics_init_IRQ;
 		ppc_md.get_irq        = xics_get_irq;
 	}
-	ppc_md.init_ras_IRQ = init_ras_IRQ;
 
 	ppc_md.init           = chrp_init2;
 
@@ -253,6 +252,34 @@ chrp_init(unsigned long r3, unsigned long r4, unsigned long r5,
 
 	ppc_md.progress = chrp_progress;
 
+        /* build up the firmware_features bitmask field
+         * using contents of device-tree/ibm,hypertas-functions.
+         * Ultimately this functionality may be moved into prom.c prom_init().
+         */
+	struct device_node * dn;
+	char * hypertas;
+	unsigned int len;
+	dn = find_path_device("/rtas");
+	cur_cpu_spec->firmware_features = 0;
+	hypertas = get_property(dn, "ibm,hypertas-functions", &len);
+	if (hypertas) {
+	    while (len > 0){
+		int i;
+		/* check value against table of strings */
+		for(i=0; i < FIRMWARE_MAX_FEATURES ;i++) {
+		    if ((firmware_features_table[i].name) && (strcmp(firmware_features_table[i].name,hypertas))==0) {
+			/* we have a match */
+			cur_cpu_spec->firmware_features |= (1UL << firmware_features_table[i].val);
+			break;
+		    } 
+		}
+		int hypertas_len = strlen(hypertas);
+		len -= hypertas_len +1;
+		hypertas+= hypertas_len +1;
+	    }
+	}
+	udbg_printf("firmware_features bitmask: 0x%x \n",
+		    cur_cpu_spec->firmware_features);
 }
 
 void

arch/ppc64/kernel/cputable.c

+/*
+ *  arch/ppc64/kernel/cputable.c
+ *
+ *  Copyright (C) 2001 Ben. Herrenschmidt (benh@kernel.crashing.org)
+ *
+ *  Modifications for ppc64:
+ *      Copyright (C) 2003 Dave Engebretsen <engebret@us.ibm.com>
+ * 
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/config.h>
+#include <linux/string.h>
+#include <linux/sched.h>
+#include <linux/threads.h>
+#include <linux/init.h>
+#include <asm/cputable.h>
+
+struct cpu_spec* cur_cpu_spec = NULL;
+
+extern void __setup_cpu_power3(unsigned long offset, struct cpu_spec* spec);
+extern void __setup_cpu_power4(unsigned long offset, struct cpu_spec* spec);
+
+
+/* We only set the altivec features if the kernel was compiled with altivec
+ * support
+ */
+#ifdef CONFIG_ALTIVEC
+#define CPU_FTR_ALTIVEC_COMP	CPU_FTR_ALTIVEC
+#else
+#define CPU_FTR_ALTIVEC_COMP	0
+#endif
+
+struct cpu_spec	cpu_specs[] = {
+    {	/* Power3 */
+	    0xffff0000, 0x00400000, "Power3 (630)",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power3,
+	    COMMON_PPC64_FW
+    },
+    {	/* Power3+ */
+	    0xffff0000, 0x00410000, "Power3 (630+)",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power3,
+	    COMMON_PPC64_FW
+    },
+    {	/* Northstar */
+	    0xffff0000, 0x00330000, "Northstar",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power3,
+	    COMMON_PPC64_FW
+    },
+    {	/* Pulsar */
+	    0xffff0000, 0x00340000, "Pulsar",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power3,
+	    COMMON_PPC64_FW
+    },
+    {	/* I-star */
+	    0xffff0000, 0x00360000, "I-star",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power3,
+	    COMMON_PPC64_FW
+    },
+    {	/* S-star */
+	    0xffff0000, 0x00370000, "S-star",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power3,
+	    COMMON_PPC64_FW
+    },
+    {	/* Power4 */
+	    0xffff0000, 0x00350000, "Power4",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
+	    CPU_FTR_PPCAS_ARCH_V2,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power4,
+	    COMMON_PPC64_FW
+    },
+    {	/* Power4+ */
+	    0xffff0000, 0x00380000, "Power4+",
+	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
+	    CPU_FTR_PPCAS_ARCH_V2,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power4,
+	    COMMON_PPC64_FW
+    },
+    {	/* default match */
+	    0x00000000, 0x00000000, "(Power4-Compatible)",
+  	    CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE |
+	    CPU_FTR_PPCAS_ARCH_V2,
+	    COMMON_USER_PPC64,
+	    128, 128,
+	    __setup_cpu_power4,
+	    COMMON_PPC64_FW
+    }
+};
+
+firmware_feature_t firmware_features_table[FIRMWARE_MAX_FEATURES] = {
+    {FW_FEATURE_PFT,		"hcall-pft"},
+    {FW_FEATURE_TCE,		"hcall-tce"},
+    {FW_FEATURE_SPRG0,		"hcall-sprg0"},
+    {FW_FEATURE_DABR,		"hcall-dabr"},
+    {FW_FEATURE_COPY,		"hcall-copy"},
+    {FW_FEATURE_ASR,		"hcall-asr"},
+    {FW_FEATURE_DEBUG,		"hcall-debug"},
+    {FW_FEATURE_PERF,		"hcall-perf"},
+    {FW_FEATURE_DUMP,		"hcall-dump"},
+    {FW_FEATURE_INTERRUPT,	"hcall-interrupt"},
+    {FW_FEATURE_MIGRATE,	"hcall-migrate"},
+};

arch/ppc64/kernel/head.S

@@ -34,6 +34,7 @@
 #include <asm/ppc_asm.h>
 #include <asm/offsets.h>
 #include <asm/bug.h>
+#include <asm/cputable.h>
 
 #ifdef CONFIG_PPC_ISERIES
 #define DO_SOFT_DISABLE
@@ -1267,6 +1268,11 @@ _GLOBAL(__start_initialization_iSeries)
 	li	r0,0
 	stdu	r0,-STACK_FRAME_OVERHEAD(r1)
 
+	LOADADDR(r3,cpu_specs)
+	LOADADDR(r4,cur_cpu_spec)
+	li	r5,0
+	bl	.identify_cpu
+
 	LOADADDR(r2,__toc_start)
 	addi	r2,r2,0x4000
 	addi	r2,r2,0x4000
@@ -1730,6 +1736,13 @@ _STATIC(start_here_pSeries)
 	li	r0,0
 	stdu	r0,-STACK_FRAME_OVERHEAD(r1)
 
+	LOADADDR(r3,cpu_specs)
+	sub	r3,r3,r26
+	LOADADDR(r4,cur_cpu_spec)
+	sub	r4,r4,r26
+	mr	r5,r26
+	bl	.identify_cpu
+
 	/* set up the TOC (physical address) */
 	LOADADDR(r2,__toc_start)
 	addi    r2,r2,0x4000
@@ -1888,6 +1901,11 @@ _STATIC(start_here_common)
 
 	bl .start_kernel
 
+_GLOBAL(__setup_cpu_power3)
+	blr
+_GLOBAL(__setup_cpu_power4)
+	blr
+
 _GLOBAL(hmt_init)
 #ifdef CONFIG_HMT
 	LOADADDR(r5, hmt_thread_data)

arch/ppc64/kernel/htab.c

@@ -46,6 +46,7 @@
 #include <asm/eeh.h>
 #include <asm/tlb.h>
 #include <asm/cacheflush.h>
+#include <asm/cputable.h>
 
 /*
  * Note:  pte   --> Linux PTE
@@ -165,7 +166,8 @@ htab_initialize(void)
 	mode_rw = _PAGE_ACCESSED | _PAGE_COHERENT | PP_RWXX;
 
 	/* XXX we currently map kernel text rw, should fix this */
-	if (cpu_has_largepage() && systemcfg->physicalMemorySize > 256*MB) {
+	if ((cur_cpu_spec->cpu_features & CPU_FTR_16M_PAGE)
+	    && systemcfg->physicalMemorySize > 256*MB) {
 		create_pte_mapping((unsigned long)KERNELBASE, 
 				   KERNELBASE + 256*MB, mode_rw, 0);
 		create_pte_mapping((unsigned long)KERNELBASE + 256*MB, 
@@ -279,7 +281,8 @@ int __hash_page(unsigned long ea, unsigned long access, unsigned long vsid,
 #define PPC64_HWNOEXEC (1 << 2)
 
 	/* We do lazy icache flushing on cpus that support it */
-	if (unlikely(cpu_has_noexecute() && pfn_valid(pte_pfn(new_pte)))) {
+	if (unlikely((cur_cpu_spec->cpu_features & CPU_FTR_NOEXECUTE)
+		     && pfn_valid(pte_pfn(new_pte)))) {
 		struct page *page = pte_page(new_pte);
 
 		/* page is dirty */

arch/ppc64/kernel/iSeries_setup.c

@@ -33,6 +33,7 @@
 #include <asm/mmu.h>
 #include <asm/pgtable.h>
 #include <asm/mmu_context.h>
+#include <asm/cputable.h>
 
 #include <asm/time.h>
 #include "iSeries_setup.h"
@@ -254,7 +255,7 @@ unsigned long iSeries_process_mainstore_vpd( struct MemoryBlock *mb_array, unsig
 {
 	unsigned long i;
 	unsigned long mem_blocks = 0;
-	if (__is_processor(PV_POWER4) || __is_processor(PV_POWER4p))
+	if (cur_cpu_spec->cpu_features & CPU_FTR_SLB)
 		mem_blocks = iSeries_process_Regatta_mainstore_vpd( mb_array, max_entries );
 	else
 		mem_blocks = iSeries_process_Condor_mainstore_vpd( mb_array, max_entries );
@@ -311,7 +312,6 @@ iSeries_init_early(void)
 	ppc_md.setup_residual	 	= iSeries_setup_residual;
 	ppc_md.get_cpuinfo	 	= iSeries_get_cpuinfo;
 	ppc_md.init_IRQ		 	= iSeries_init_IRQ;
-	ppc_md.init_ras_IRQ		= NULL;
 	ppc_md.get_irq		 	= iSeries_get_irq;
 	ppc_md.init		 	= NULL;
 

arch/ppc64/kernel/irq.c

@@ -596,7 +596,6 @@ void __init init_IRQ(void)
 		once++;
 	
 	ppc_md.init_IRQ();
-	if(ppc_md.init_ras_IRQ) ppc_md.init_ras_IRQ(); 
 }
 
 static struct proc_dir_entry * root_irq_dir;

arch/ppc64/kernel/misc.S

@@ -27,6 +27,7 @@
 #include <asm/cache.h>
 #include <asm/ppc_asm.h>
 #include <asm/offsets.h>
+#include <asm/cputable.h>
 
 	.text
 
@@ -444,6 +445,95 @@ _GLOBAL(cvt_df)
 	stfd	0,0(r5)
 	blr
 
+/*
+ * identify_cpu,
+ * In:	r3 = base of the cpu_specs array
+ *	r4 = address of cur_cpu_spec
+ *	r5 = relocation offset
+ */
+_GLOBAL(identify_cpu)
+	mfpvr	r7
+1:
+	lwz	r8,CPU_SPEC_PVR_MASK(r3)
+	and	r8,r8,r7
+	lwz	r9,CPU_SPEC_PVR_VALUE(r3)
+	cmplw	0,r9,r8
+	beq	1f
+	addi	r3,r3,CPU_SPEC_ENTRY_SIZE
+	b	1b
+1:
+	add	r3,r3,r5
+	std	r3,0(r4)
+	blr
+
+/*
+ * do_cpu_ftr_fixups - goes through the list of CPU feature fixups
+ * and writes nop's over sections of code that don't apply for this cpu.
+ * r3 = data offset (not changed)
+ */
+_GLOBAL(do_cpu_ftr_fixups)
+	/* Get CPU 0 features */
+	LOADADDR(r6,cur_cpu_spec)
+	sub	r6,r6,r3
+	ld	r4,0(r6)
+	sub	r4,r4,r3
+	ld	r4,CPU_SPEC_FEATURES(r4)
+	/* Get the fixup table */
+	LOADADDR(r6,__start___ftr_fixup)
+	sub	r6,r6,r3
+	LOADADDR(r7,__stop___ftr_fixup)
+	sub	r7,r7,r3
+	/* Do the fixup */
+1:	cmpld	r6,r7
+	bgelr
+	addi	r6,r6,32
+	ld	r8,-32(r6)	/* mask */
+	and	r8,r8,r4
+	ld	r9,-24(r6)	/* value */
+	cmpld	r8,r9
+	beq	1b
+	ld	r8,-16(r6)	/* section begin */
+	ld	r9,-8(r6)	/* section end */
+	subf.	r9,r8,r9
+	beq	1b
+	/* write nops over the section of code */
+	/* todo: if large section, add a branch at the start of it */
+	srwi	r9,r9,2
+	mtctr	r9
+	sub	r8,r8,r3
+	lis	r0,0x60000000@h	/* nop */
+3:	stw	r0,0(r8)
+	andi.	r10,r4,CPU_FTR_SPLIT_ID_CACHE@l
+	beq	2f
+	dcbst	0,r8		/* suboptimal, but simpler */
+	sync
+	icbi	0,r8
+2:	addi	r8,r8,4
+	bdnz	3b
+	sync			/* additional sync needed on g4 */
+	isync
+	b	1b
+
+/*
+ * call_setup_cpu - call the setup_cpu function for this cpu
+ * r3 = data offset
+ *
+ * Setup function is called with:
+ *   r3 = data offset
+ *   r4 = ptr to CPU spec (relocated)
+ */
+_GLOBAL(call_setup_cpu)
+	LOADADDR(r4, cur_cpu_spec)
+	sub	r4,r4,r3
+	lwz	r4,0(r4)		# load pointer to cpu_spec
+	sub	r4,r4,r3		# relocate
+	lwz	r6,CPU_SPEC_SETUP(r4)	# load function pointer
+	sub	r6,r6,r3
+	mtctr	r6
+	bctr
+
+
+
 /*
  * Create a kernel thread
  *   kernel_thread(fn, arg, flags)

arch/ppc64/kernel/module.c

@@ -376,15 +376,57 @@ int apply_relocate_add(Elf64_Shdr *sechdrs,
 	return 0;
 }
 
+LIST_HEAD(module_bug_list);
+
 int module_finalize(const Elf_Ehdr *hdr,
 		const Elf_Shdr *sechdrs, struct module *me)
 {
+	char *secstrings;
+	unsigned int i;
+
+	me->arch.bug_table = NULL;
+	me->arch.num_bugs = 0;
+
+	/* Find the __bug_table section, if present */
+	secstrings = (char *)hdr + sechdrs[hdr->e_shstrndx].sh_offset;
+	for (i = 1; i < hdr->e_shnum; i++) {
+		if (strcmp(secstrings+sechdrs[i].sh_name, "__bug_table"))
+			continue;
+		me->arch.bug_table = (void *) sechdrs[i].sh_addr;
+		me->arch.num_bugs = sechdrs[i].sh_size / sizeof(struct bug_entry);
+		break;
+	}
+
+	/*
+	 * Strictly speaking this should have a spinlock to protect against
+	 * traversals, but since we only traverse on BUG()s, a spinlock
+	 * could potentially lead to deadlock and thus be counter-productive.
+	 */
+	list_add(&me->arch.bug_list, &module_bug_list);
+
 	sort_ex_table((struct exception_table_entry *)me->extable,
 		      (struct exception_table_entry *)me->extable +
 				me->num_exentries);
+
 	return 0;
 }
 
 void module_arch_cleanup(struct module *mod)
 {
+	list_del(&mod->arch.bug_list);
+}
+
+struct bug_entry *module_find_bug(unsigned long bugaddr)
+{
+	struct mod_arch_specific *mod;
+	unsigned int i;
+	struct bug_entry *bug;
+
+	list_for_each_entry(mod, &module_bug_list, bug_list) {
+		bug = mod->bug_table;
+		for (i = 0; i < mod->num_bugs; ++i, ++bug)
+			if (bugaddr == bug->bug_addr)
+				return bug;
+	}
+	return NULL;
 }

arch/ppc64/kernel/pSeries_htab.c

@@ -21,6 +21,7 @@
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 #include <asm/tlb.h>
+#include <asm/cputable.h>
 
 #define HPTE_LOCK_BIT 3
 
@@ -217,7 +218,7 @@ static long pSeries_hpte_updatepp(unsigned long slot, unsigned long newpp,
 	}
 
 	/* Ensure it is out of the tlb too */
-	if (cpu_has_tlbiel() && !large && local) {
+	if ((cur_cpu_spec->cpu_features & CPU_FTR_TLBIEL) && !large && local) {
 		_tlbiel(va);
 	} else {
 		spin_lock_irqsave(&pSeries_tlbie_lock, flags);
@@ -283,7 +284,7 @@ static void pSeries_hpte_invalidate(unsigned long slot, unsigned long va,
 	}
 
 	/* Invalidate the tlb */
-	if (cpu_has_tlbiel() && !large && local) {
+	if ((cur_cpu_spec->cpu_features & CPU_FTR_TLBIEL) && !large && local) {
 		_tlbiel(va);
 	} else {
 		spin_lock_irqsave(&pSeries_tlbie_lock, flags);
@@ -346,7 +347,7 @@ static void pSeries_flush_hash_range(unsigned long context,
 		j++;
 	}
 
-	if (cpu_has_tlbiel() && !large && local) {
+	if ((cur_cpu_spec->cpu_features & CPU_FTR_TLBIEL) && !large && local) {
 		asm volatile("ptesync":::"memory");
 
 		for (i = 0; i < j; i++) {

arch/ppc64/kernel/process.c

@@ -45,6 +45,7 @@
 #include <asm/machdep.h>
 #include <asm/iSeries/HvCallHpt.h>
 #include <asm/hardirq.h>
+#include <asm/cputable.h>
 
 struct task_struct *last_task_used_math = NULL;
 
@@ -412,7 +413,7 @@ void initialize_paca_hardware_interrupt_stack(void)
 	 * __get_free_pages() might give us a page > KERNBASE+256M which
 	 * is mapped with large ptes so we can't set up the guard page.
 	 */
-	if (cpu_has_largepage())
+	if (cur_cpu_spec->cpu_features & CPU_FTR_16M_PAGE)
 		return;
 
 	for (i=0; i < NR_CPUS; i++) {

arch/ppc64/kernel/ras.c

@@ -58,7 +58,6 @@ static irqreturn_t ras_epow_interrupt(int irq, void *dev_id,
 					struct pt_regs * regs);
 static irqreturn_t ras_error_interrupt(int irq, void *dev_id,
 					struct pt_regs * regs);
-void init_ras_IRQ(void);
 
 /* #define DEBUG */
 
@@ -66,7 +65,8 @@ void init_ras_IRQ(void);
  * Initialize handlers for the set of interrupts caused by hardware errors
  * and power system events.
  */
-void init_ras_IRQ(void) {
+static int __init init_ras_IRQ(void)
+{
 	struct device_node *np;
 	unsigned int *ireg, len, i;
 
@@ -91,7 +91,10 @@ void init_ras_IRQ(void) {
 			ireg++;
 		}
 	}
+
+	return 1;
 }
+__initcall(init_ras_IRQ);
 
 /*
  * Handle power subsystem events (EPOW).

arch/ppc64/kernel/smp.c

@@ -47,6 +47,7 @@
 #include "open_pic.h"
 #include <asm/machdep.h>
 #include <asm/xics.h>
+#include <asm/cputable.h>
 
 int smp_threads_ready;
 unsigned long cache_decay_ticks;
@@ -583,7 +584,7 @@ int __devinit __cpu_up(unsigned int cpu)
 	paca[cpu].prof_multiplier = 1;
 	paca[cpu].default_decr = tb_ticks_per_jiffy / decr_overclock;
 
-	if (!cpu_has_slb()) {
+	if (!(cur_cpu_spec->cpu_features & CPU_FTR_SLB)) {
 		void *tmp;
 
 		/* maximum of 48 CPUs on machines with a segment table */

arch/ppc64/kernel/stab.c

@@ -21,6 +21,7 @@
 #include <asm/paca.h>
 #include <asm/naca.h>
 #include <asm/pmc.h>
+#include <asm/cputable.h>
 
 int make_ste(unsigned long stab, unsigned long esid, unsigned long vsid);
 void make_slbe(unsigned long esid, unsigned long vsid, int large,
@@ -38,7 +39,7 @@ void stab_initialize(unsigned long stab)
 	esid = GET_ESID(KERNELBASE);
 	vsid = get_kernel_vsid(esid << SID_SHIFT); 
 
-	if (cpu_has_slb()) {
+	if (cur_cpu_spec->cpu_features & CPU_FTR_SLB) {
 		/* Invalidate the entire SLB & all the ERATS */
 #ifdef CONFIG_PPC_ISERIES
 		asm volatile("isync; slbia; isync":::"memory");
@@ -222,7 +223,7 @@ void make_slbe(unsigned long esid, unsigned long vsid, int large,
 static inline void __ste_allocate(unsigned long esid, unsigned long vsid,
 				  int kernel_segment)
 {
-	if (cpu_has_slb()) {
+	if (cur_cpu_spec->cpu_features & CPU_FTR_SLB) {
 #ifndef CONFIG_PPC_ISERIES
 		if (REGION_ID(esid << SID_SHIFT) == KERNEL_REGION_ID)
 			make_slbe(esid, vsid, 1, kernel_segment); 
@@ -275,7 +276,7 @@ int ste_allocate(unsigned long ea)
 
 	esid = GET_ESID(ea);
 	__ste_allocate(esid, vsid, kernel_segment);
-	if (!cpu_has_slb()) {
+	if (!(cur_cpu_spec->cpu_features & CPU_FTR_SLB)) {
 		/* Order update */
 		asm volatile("sync":::"memory"); 
 	}
@@ -327,7 +328,7 @@ static void preload_stab(struct task_struct *tsk, struct mm_struct *mm)
 		}
 	}
 
-	if (!cpu_has_slb()) {
+	if (!(cur_cpu_spec->cpu_features & CPU_FTR_SLB)) {
 		/* Order update */
 		asm volatile("sync" : : : "memory"); 
 	}
@@ -336,7 +337,7 @@ static void preload_stab(struct task_struct *tsk, struct mm_struct *mm)
 /* Flush all user entries from the segment table of the current processor. */
 void flush_stab(struct task_struct *tsk, struct mm_struct *mm)
 {
-	if (cpu_has_slb()) {
+	if (cur_cpu_spec->cpu_features & CPU_FTR_SLB) {
 		/*
 		 * XXX disable 32bit slb invalidate optimisation until we fix
 		 * the issue where a 32bit app execed out of a 64bit app can

arch/ppc64/kernel/traps.c

@@ -28,6 +28,7 @@
 #include <linux/interrupt.h>
 #include <linux/config.h>
 #include <linux/init.h>
+#include <linux/module.h>
 
 #include <asm/pgtable.h>
 #include <asm/uaccess.h>
@@ -306,6 +307,56 @@ static void parse_fpe(struct pt_regs *regs)
 	_exception(SIGFPE, &info, regs);
 }
 
+/*
+ * Look through the list of trap instructions that are used for BUG(),
+ * BUG_ON() and WARN_ON() and see if we hit one.  At this point we know
+ * that the exception was caused by a trap instruction of some kind.
+ * Returns 1 if we should continue (i.e. it was a WARN_ON) or 0
+ * otherwise.
+ */
+extern struct bug_entry __start___bug_table[], __stop___bug_table[];
+
+#ifndef CONFIG_MODULES
+#define module_find_bug(x)	NULL
+#endif
+
+static struct bug_entry *find_bug(unsigned long bugaddr)
+{
+	struct bug_entry *bug;
+
+	for (bug = __start___bug_table; bug < __stop___bug_table; ++bug)
+		if (bugaddr == bug->bug_addr)
+			return bug;
+	return module_find_bug(bugaddr);
+}
+
+int
+check_bug_trap(struct pt_regs *regs)
+{
+	struct bug_entry *bug;
+	unsigned long addr;
+
+	if (regs->msr & MSR_PR)
+		return 0;	/* not in kernel */
+	addr = regs->nip;	/* address of trap instruction */
+	if (addr < PAGE_OFFSET)
+		return 0;
+	bug = find_bug(regs->nip);
+	if (bug == NULL)
+		return 0;
+	if (bug->line & BUG_WARNING_TRAP) {
+		/* this is a WARN_ON rather than BUG/BUG_ON */
+		printk(KERN_ERR "Badness in %s at %s:%d\n",
+		       bug->function, bug->file,
+		       bug->line & ~BUG_WARNING_TRAP);
+		dump_stack();
+		return 1;
+	}
+	printk(KERN_CRIT "kernel BUG in %s at %s:%d!\n",
+	       bug->function, bug->file, bug->line);
+	return 0;
+}
+
 void
 ProgramCheckException(struct pt_regs *regs)
 {
@@ -330,6 +381,10 @@ ProgramCheckException(struct pt_regs *regs)
 		if (debugger_bpt && debugger_bpt(regs))
 			return;
 #endif
+		if (check_bug_trap(regs)) {
+			regs->nip += 4;
+			return;
+		}
 		info.si_signo = SIGTRAP;
 		info.si_errno = 0;
 		info.si_code = TRAP_BRKPT;

arch/ppc64/mm/init.c

@@ -59,6 +59,7 @@
 #include <asm/eeh.h>
 #include <asm/processor.h>
 #include <asm/mmzone.h>
+#include <asm/cputable.h>
 
 #include <asm/ppcdebug.h>
 
@@ -512,7 +513,7 @@ void __init paging_init(void)
 
 static struct kcore_list kcore_vmem;
 
-static void setup_kcore(void)
+static int __init setup_kcore(void)
 {
 	int i;
 
@@ -536,7 +537,10 @@ static void setup_kcore(void)
 	}
 
 	kclist_add(&kcore_vmem, (void *)VMALLOC_START, VMALLOC_END-VMALLOC_START);
+
+	return 0;
 }
+module_init(setup_kcore);
 
 void initialize_paca_hardware_interrupt_stack(void);
 
@@ -606,8 +610,6 @@ void __init mem_init(void)
 #endif
 	mem_init_done = 1;
 
-	setup_kcore();
-
 	/* set the last page of each hardware interrupt stack to be protected */
 	initialize_paca_hardware_interrupt_stack();
 
@@ -698,7 +700,7 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long ea,
 	int local = 0;
 
 	/* handle i-cache coherency */
-	if (!cpu_has_noexecute()) {
+	if (!(cur_cpu_spec->cpu_features & CPU_FTR_NOEXECUTE)) {
 		unsigned long pfn = pte_pfn(pte);
 		if (pfn_valid(pfn)) {
 			struct page *page = pfn_to_page(pfn);

arch/ppc64/mm/numa.c

@@ -154,8 +154,28 @@ static int __init parse_numa_properties(void)
 		if (max_domain < numa_domain)
 			max_domain = numa_domain;
 
-		node_data[numa_domain].node_start_pfn = start / PAGE_SIZE;
+		/* 
+		 * For backwards compatibility, OF splits the first node
+		 * into two regions (the first being 0-4GB). Check for
+		 * this simple case and complain if there is a gap in
+		 * memory
+		 */
+		if (node_data[numa_domain].node_size) {
+			unsigned long shouldstart =
+				node_data[numa_domain].node_start_pfn + 
+				node_data[numa_domain].node_size;
+			if (shouldstart != (start / PAGE_SIZE)) {
+				printk(KERN_ERR "Hole in node, disabling "
+						"region start %lx length %lx\n",
+						start, size);
+				continue;
+			}
+			node_data[numa_domain].node_size += size / PAGE_SIZE;
+		} else {
+			node_data[numa_domain].node_start_pfn =
+				start / PAGE_SIZE;
 			node_data[numa_domain].node_size = size / PAGE_SIZE;
+		}
 
 		for (i = start ; i < (start+size); i += MEMORY_INCREMENT)
 			numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] =
@@ -174,6 +194,20 @@ static int __init parse_numa_properties(void)
 	return 0;
 }
 
+void setup_nonnuma(void)
+{
+	unsigned long i;
+
+	for (i = 0; i < NR_CPUS; i++)
+		map_cpu_to_node(i, 0);
+
+	node_data[0].node_start_pfn = 0;
+	node_data[0].node_size = lmb_end_of_DRAM() / PAGE_SIZE;
+
+	for (i = 0 ; i < lmb_end_of_DRAM(); i += MEMORY_INCREMENT)
+		numa_memory_lookup_table[i >> MEMORY_INCREMENT_SHIFT] = 0;
+}
+
 void __init do_init_bootmem(void)
 {
 	int nid;
@@ -181,9 +215,8 @@ void __init do_init_bootmem(void)
 	min_low_pfn = 0;
 	max_low_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT;
 
-	/* XXX FIXME: support machines without associativity information */
 	if (parse_numa_properties())
-		BUG();
+		setup_nonnuma();
 
 	for (nid = 0; nid < numnodes; nid++) {
 		unsigned long start_paddr, end_paddr;
@@ -204,7 +237,7 @@ void __init do_init_bootmem(void)
 
 		NODE_DATA(nid)->bdata = &plat_node_bdata[nid];
 
-		bootmap_pages = bootmem_bootmap_pages(end_paddr - start_paddr);
+		bootmap_pages = bootmem_bootmap_pages((end_paddr - start_paddr) >> PAGE_SHIFT);
 		dbg("bootmap_pages = %lx\n", bootmap_pages);
 
 		bootmem_paddr = lmb_alloc_base(bootmap_pages << PAGE_SHIFT,

arch/ppc64/vmlinux.lds.S

@@ -65,6 +65,14 @@ SECTIONS
   __ex_table : { *(__ex_table) }
   __stop___ex_table = .;
 
+  __start___bug_table = .;
+  __bug_table : { *(__bug_table) }
+  __stop___bug_table = .;
+
+  __start___ftr_fixup = .;
+  __ftr_fixup : { *(__ftr_fixup) }
+  __stop___ftr_fixup = .;
+
   . = ALIGN(16384);		/* init_task */
   .data.init_task : { *(.data.init_task) }
 

include/asm-ppc64/bug.h

 #ifndef _PPC64_BUG_H
 #define _PPC64_BUG_H
 
-#include <linux/config.h>
-
 /*
  * Define an illegal instr to trap on the bug.
  * We don't use 0 because that marks the end of a function
@@ -13,29 +11,48 @@
 
 #ifndef __ASSEMBLY__
 
-#ifdef CONFIG_XMON
-struct pt_regs;
-extern void xmon(struct pt_regs *excp);
-#define BUG() do { \
-	printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \
-	xmon(0); \
-} while (0)
-#else
+struct bug_entry {
+	unsigned long	bug_addr;
+	long		line;
+	const char	*file;
+	const char	*function;
+};
+
+/*
+ * If this bit is set in the line number it means that the trap
+ * is for WARN_ON rather than BUG or BUG_ON.
+ */
+#define BUG_WARNING_TRAP	0x1000000
+
 #define BUG() do {							 \
-	printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \
-	__asm__ __volatile__(".long " BUG_ILLEGAL_INSTR); \
+	__asm__ __volatile__(						 \
+		"1:	twi 31,0,0\n"					 \
+		".section __bug_table,\"a\"\n\t"			 \
+		"	.llong 1b,%0,%1,%2\n"				 \
+		".previous"						 \
+		: : "i" (__LINE__), "i" (__FILE__), "i" (__FUNCTION__)); \
 } while (0)
-#endif
 
-#define BUG_ON(condition) do { if (unlikely((condition)!=0)) BUG(); } while(0)
+#define BUG_ON(x) do {						\
+	__asm__ __volatile__(					\
+		"1:	tdnei %0,0\n"				\
+		".section __bug_table,\"a\"\n\t"		\
+		"	.llong 1b,%1,%2,%3\n"			\
+		".previous"					\
+		: : "r" (x), "i" (__LINE__), "i" (__FILE__),	\
+		    "i" (__FUNCTION__));			\
+} while (0)
 
 #define PAGE_BUG(page) do { BUG(); } while (0)
 
-#define WARN_ON(condition) do { \
-	if (unlikely((condition)!=0)) { \
-		printk("Badness in %s at %s:%d\n", __FUNCTION__, __FILE__, __LINE__); \
-		dump_stack(); \
-	} \
+#define WARN_ON(x) do {						\
+	__asm__ __volatile__(					\
+		"1:	tdnei %0,0\n"				\
+		".section __bug_table,\"a\"\n\t"		\
+		"	.llong 1b,%1,%2,%3\n"			\
+		".previous"					\
+		: : "r" (x), "i" (__LINE__ + BUG_WARNING_TRAP),	\
+		    "i" (__FILE__), "i" (__FUNCTION__));	\
 } while (0)
 
 #endif

include/asm-ppc64/cputable.h

+/*
+ *  include/asm-ppc64/cputable.h
+ *
+ *  Copyright (C) 2001 Ben. Herrenschmidt (benh@kernel.crashing.org)
+ *
+ *  Modifications for ppc64:
+ *      Copyright (C) 2003 Dave Engebretsen <engebret@us.ibm.com>
+ * 
+ *  This program is free software; you can redistribute it and/or
+ *  modify it under the terms of the GNU General Public License
+ *  as published by the Free Software Foundation; either version
+ *  2 of the License, or (at your option) any later version.
+ */
+
+#ifndef __ASM_PPC_CPUTABLE_H
+#define __ASM_PPC_CPUTABLE_H
+
+/* Exposed to userland CPU features - Must match ppc32 definitions */
+#define PPC_FEATURE_32			0x80000000
+#define PPC_FEATURE_64			0x40000000
+#define PPC_FEATURE_601_INSTR		0x20000000
+#define PPC_FEATURE_HAS_ALTIVEC		0x10000000
+#define PPC_FEATURE_HAS_FPU		0x08000000
+#define PPC_FEATURE_HAS_MMU		0x04000000
+#define PPC_FEATURE_HAS_4xxMAC		0x02000000
+#define PPC_FEATURE_UNIFIED_CACHE	0x01000000
+
+#ifdef __KERNEL__
+
+#ifndef __ASSEMBLY__
+
+/* This structure can grow, it's real size is used by head.S code
+ * via the mkdefs mecanism.
+ */
+struct cpu_spec;
+
+typedef	void (*cpu_setup_t)(unsigned long offset, struct cpu_spec* spec);
+
+struct cpu_spec {
+	/* CPU is matched via (PVR & pvr_mask) == pvr_value */
+	unsigned int	pvr_mask;
+	unsigned int	pvr_value;
+
+	char		*cpu_name;
+	unsigned long	cpu_features;		/* Kernel features */
+	unsigned int	cpu_user_features;	/* Userland features */
+
+	/* cache line sizes */
+	unsigned int	icache_bsize;
+	unsigned int	dcache_bsize;
+
+	/* this is called to initialize various CPU bits like L1 cache,
+	 * BHT, SPD, etc... from head.S before branching to identify_machine
+	 */
+	cpu_setup_t	cpu_setup;
+
+	/* This is used to identify firmware features which are available
+	 * to the kernel.
+	 */
+	unsigned long   firmware_features;
+};
+
+extern struct cpu_spec		cpu_specs[];
+extern struct cpu_spec		*cur_cpu_spec;
+
+
+/* firmware feature bitmask values */
+#define FIRMWARE_MAX_FEATURES 63
+
+#define FW_FEATURE_PFT		(1UL<<0)
+#define FW_FEATURE_TCE		(1UL<<1)	
+#define FW_FEATURE_SPRG0	(1UL<<2)	
+#define FW_FEATURE_DABR		(1UL<<3)	
+#define FW_FEATURE_COPY		(1UL<<4)	
+#define FW_FEATURE_ASR		(1UL<<5)	
+#define FW_FEATURE_DEBUG	(1UL<<6)	
+#define FW_FEATURE_PERF		(1UL<<7)	
+#define FW_FEATURE_DUMP		(1UL<<8)	
+#define FW_FEATURE_INTERRUPT	(1UL<<9)	
+#define FW_FEATURE_MIGRATE	(1UL<<10)	
+
+typedef struct {
+    unsigned long val;
+    char * name;
+} firmware_feature_t;
+
+extern firmware_feature_t firmware_features_table[];
+
+#endif /* __ASSEMBLY__ */
+
+/* CPU kernel features */
+
+/* Retain the 32b definitions for the time being - use bottom half of word */
+#define CPU_FTR_SPLIT_ID_CACHE		0x0000000000000001
+#define CPU_FTR_L2CR			0x0000000000000002
+#define CPU_FTR_SPEC7450		0x0000000000000004
+#define CPU_FTR_ALTIVEC			0x0000000000000008
+#define CPU_FTR_TAU			0x0000000000000010
+#define CPU_FTR_CAN_DOZE		0x0000000000000020
+#define CPU_FTR_USE_TB			0x0000000000000040
+#define CPU_FTR_604_PERF_MON		0x0000000000000080
+#define CPU_FTR_601			0x0000000000000100
+#define CPU_FTR_HPTE_TABLE		0x0000000000000200
+#define CPU_FTR_CAN_NAP			0x0000000000000400
+#define CPU_FTR_L3CR			0x0000000000000800
+#define CPU_FTR_L3_DISABLE_NAP		0x0000000000001000
+#define CPU_FTR_NAP_DISABLE_L2_PR	0x0000000000002000
+#define CPU_FTR_DUAL_PLL_750FX		0x0000000000004000
+
+/* Add the 64b processor unique features in the top half of the word */
+#define CPU_FTR_SLB           		0x0000000100000000
+#define CPU_FTR_16M_PAGE      		0x0000000200000000
+#define CPU_FTR_TLBIEL         		0x0000000400000000
+#define CPU_FTR_NOEXECUTE     		0x0000000800000000
+#define CPU_FTR_NODSISRALIGN  		0x0000001000000000
+
+/* Platform firmware features */
+#define FW_FTR_                         0x0000000000000001
+
+#ifndef __ASSEMBLY__
+#define COMMON_USER_PPC64	(PPC_FEATURE_32 | PPC_FEATURE_64 | \
+			         PPC_FEATURE_HAS_FPU | PPC_FEATURE_HAS_MMU)
+
+#define CPU_FTR_PPCAS_ARCH_V2   (CPU_FTR_SLB | CPU_FTR_16M_PAGE | \
+                                 CPU_FTR_TLBIEL | CPU_FTR_NOEXECUTE | \
+                                 CPU_FTR_NODSISRALIGN)
+
+#define COMMON_PPC64_FW	(0)
+#endif
+
+#ifdef __ASSEMBLY__
+
+#define BEGIN_FTR_SECTION		98:
+
+#define END_FTR_SECTION(msk, val)		\
+99:						\
+	.section __ftr_fixup,"a";		\
+	.align 3;				\
+	.llong msk;			        \
+	.llong val;			        \
+	.llong 98b;			        \
+	.llong 99b;	 		        \
+	.previous
+
+#define END_FTR_SECTION_IFSET(msk)	END_FTR_SECTION((msk), (msk))
+#define END_FTR_SECTION_IFCLR(msk)	END_FTR_SECTION((msk), 0)
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __ASM_PPC_CPUTABLE_H */
+#endif /* __KERNEL__ */
+

include/asm-ppc64/machdep.h

@@ -67,7 +67,6 @@ struct machdep_calls {
 	void		(*get_cpuinfo)(struct seq_file *m);
 
 	void		(*init_IRQ)(void);
-	void		(*init_ras_IRQ)(void);
 	int		(*get_irq)(struct pt_regs *);
 
 	/* Optional, may be NULL. */

include/asm-ppc64/module.h

 #ifndef _ASM_PPC64_MODULE_H
 #define _ASM_PPC64_MODULE_H
 
+#include <linux/list.h>
+#include <asm/bug.h>
+
 struct mod_arch_specific
 {
 	/* Index of stubs section within module. */
@@ -8,8 +11,15 @@ struct mod_arch_specific
 
 	/* What section is the TOC? */
 	unsigned int toc_section;
+
+	/* List of BUG addresses, source line numbers and filenames */
+	struct list_head bug_list;
+	struct bug_entry *bug_table;
+	unsigned int num_bugs;
 };
 
+extern struct bug_entry *module_find_bug(unsigned long bugaddr);
+
 #define Elf_Shdr Elf64_Shdr
 #define Elf_Sym Elf64_Sym
 #define Elf_Ehdr Elf64_Ehdr

include/asm-ppc64/processor.h

@@ -730,18 +730,6 @@ static inline void prefetchw(const void *x)
 
 #define spin_lock_prefetch(x)	prefetchw(x)
 
-#define cpu_has_largepage()	(processor_type() == PV_POWER4 || \
-				 processor_type() == PV_POWER4p)
-
-#define cpu_has_slb()		(processor_type() == PV_POWER4 || \
-				 processor_type() == PV_POWER4p)
-
-#define cpu_has_tlbiel()	(processor_type() == PV_POWER4 || \
-				 processor_type() == PV_POWER4p)
-
-#define cpu_has_noexecute()	(processor_type() == PV_POWER4 || \
-				 processor_type() == PV_POWER4p)
-
 /* XXX we have to call HV to set when in LPAR */
 #define cpu_has_dabr()		(1)
 

include/asm-ppc64/uaccess.h

@@ -226,10 +226,22 @@ copy_to_user(void *to, const void *from, unsigned long n)
 	return n;
 }
 
+static inline unsigned long
+copy_in_user(void *to, const void *from, unsigned long n)
+{
+        if (!access_ok(VERIFY_READ, from, n) ||
+	     !access_ok(VERIFY_WRITE, to, n))
+		return n;
+
+	return __copy_tofrom_user(to, from, n);
+}
+
 #define __copy_from_user(to, from, size) \
 	__copy_tofrom_user((to), (from), (size))
 #define __copy_to_user(to, from, size) \
 	__copy_tofrom_user((to), (from), (size))
+#define __copy_in_user(to, from, size) \
+	__copy_tofrom_user((to), (from), (size))
 
 extern unsigned long __clear_user(void *addr, unsigned long size);