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

into samba.org:/scratch/anton/linux-2.5_ppc64

arch/ppc64/Makefile

@@ -18,7 +18,8 @@ KERNELLOAD	=0xc000000000000000
 LINKFLAGS	= -T arch/ppc64/vmlinux.lds -Bstatic \
 		-e $(KERNELLOAD) -Ttext $(KERNELLOAD)
 CFLAGS		:= $(CFLAGS) -fsigned-char -msoft-float -pipe \
-		-Wno-uninitialized -mminimal-toc -mtraceback=full
+		-Wno-uninitialized -mminimal-toc -mtraceback=full \
+		-Wa,-mpower4 -finline-limit-2000
 CPP		= $(CC) -E $(CFLAGS)
 
 

arch/ppc64/boot/main.c

@@ -153,7 +153,7 @@ make_bi_recs(unsigned long addr)
 
 	rec = bi_rec_alloc(rec, 2);
 	rec->tag = BI_MACHTYPE;
-	rec->data[0] = _MACH_pSeries;
+	rec->data[0] = PLATFORM_PSERIES;
 	rec->data[1] = 1;
 
 	if ( initrd_size > 0 ) {

arch/ppc64/config.in

@@ -5,7 +5,6 @@
 define_bool CONFIG_UID16 n
 define_bool CONFIG_RWSEM_GENERIC_SPINLOCK n
 define_bool CONFIG_RWSEM_XCHGADD_ALGORITHM y
-define_bool CONFIG_GENERIC_BUST_SPINLOCK n
 define_bool CONFIG_GENERIC_ISA_DMA y
 define_bool CONFIG_HAVE_DEC_LOCK y
 
@@ -30,8 +29,6 @@ define_bool CONFIG_PREEMPT n
 
 if [ "$CONFIG_PPC_ISERIES" = "y" ]; then
    define_bool CONFIG_MSCHUNKS y
-else
-bool 'MsChunks Physical to Absolute address translation support' CONFIG_MSCHUNKS
 fi
 
 endmenu
@@ -103,8 +100,12 @@ if [ "$CONFIG_SCSI" != "n" ]; then
 fi
 endmenu
 
+source drivers/message/fusion/Config.in
+
 source drivers/ieee1394/Config.in
 
+source drivers/message/i2o/Config.in
+
 if [ "$CONFIG_NET" = "y" ]; then
    mainmenu_option next_comment
    comment 'Network device support'
@@ -181,6 +182,9 @@ if [ "$CONFIG_VIOCD" = "y" ]; then
 fi
 
 source drivers/char/Config.in
+
+source drivers/media/Config.in
+
 source fs/Config.in
 
 mainmenu_option next_comment
@@ -194,13 +198,21 @@ endmenu
 
 source drivers/usb/Config.in
 
+source net/bluetooth/Config.in
+
 mainmenu_option next_comment
 comment 'Kernel hacking'
 
-bool 'Magic SysRq key' CONFIG_MAGIC_SYSRQ
-bool 'Include kgdb kernel debugger' CONFIG_KGDB
-bool 'Include xmon kernel debugger' CONFIG_XMON
-bool 'Include PPCDBG realtime debugging' CONFIG_PPCDBG
+bool 'Kernel debugging' CONFIG_DEBUG_KERNEL
+if [ "$CONFIG_DEBUG_KERNEL" != "n" ]; then
+   bool '  Debug memory allocations' CONFIG_DEBUG_SLAB
+   bool '  Magic SysRq key' CONFIG_MAGIC_SYSRQ
+   bool '  Include xmon kernel debugger' CONFIG_XMON
+   if [ "$CONFIG_XMON" = "y" ]; then
+      bool '  Enable xmon by default' CONFIG_XMON_DEFAULT
+   fi
+   bool '  Include PPCDBG realtime debugging' CONFIG_PPCDBG
+fi
 endmenu
 
 source lib/Config.in

arch/ppc64/defconfig

@@ -4,7 +4,6 @@
 # CONFIG_UID16 is not set
 # CONFIG_RWSEM_GENERIC_SPINLOCK is not set
 CONFIG_RWSEM_XCHGADD_ALGORITHM=y
-# CONFIG_GENERIC_BUST_SPINLOCK is not set
 CONFIG_GENERIC_ISA_DMA=y
 CONFIG_HAVE_DEC_LOCK=y
 
@@ -38,9 +37,7 @@ CONFIG_PPC64=y
 CONFIG_SMP=y
 CONFIG_IRQ_ALL_CPUS=y
 # CONFIG_HMT is not set
-# CONFIG_PPC_EEH is not set
 # CONFIG_PREEMPT is not set
-# CONFIG_MSCHUNKS is not set
 
 #
 # General setup
@@ -133,6 +130,11 @@ CONFIG_IPV6=m
 #
 # CONFIG_IPX is not set
 # CONFIG_ATALK is not set
+
+#
+# Appletalk devices
+#
+# CONFIG_DEV_APPLETALK is not set
 # CONFIG_DECNET is not set
 # CONFIG_BRIDGE is not set
 # CONFIG_X25 is not set
@@ -153,7 +155,6 @@ CONFIG_IPV6=m
 # ATA/IDE/MFM/RLL support
 #
 # CONFIG_IDE is not set
-# CONFIG_BLK_DEV_IDE_MODES is not set
 # CONFIG_BLK_DEV_HD is not set
 
 #
@@ -177,6 +178,7 @@ CONFIG_CHR_DEV_SG=y
 # Some SCSI devices (e.g. CD jukebox) support multiple LUNs
 #
 # CONFIG_SCSI_MULTI_LUN is not set
+# CONFIG_SCSI_REPORT_LUNS is not set
 # CONFIG_SCSI_CONSTANTS is not set
 # CONFIG_SCSI_LOGGING is not set
 
@@ -232,11 +234,26 @@ CONFIG_SCSI_SYM53C8XX_MAX_TAGS=64
 # CONFIG_SCSI_MESH is not set
 # CONFIG_SCSI_MAC53C94 is not set
 
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+
 #
 # IEEE 1394 (FireWire) support (EXPERIMENTAL)
 #
 # CONFIG_IEEE1394 is not set
 
+#
+# I2O device support
+#
+# CONFIG_I2O is not set
+# CONFIG_I2O_PCI is not set
+# CONFIG_I2O_BLOCK is not set
+# CONFIG_I2O_LAN is not set
+# CONFIG_I2O_SCSI is not set
+# CONFIG_I2O_PROC is not set
+
 #
 # Network device support
 #
@@ -282,12 +299,9 @@ CONFIG_PCNET32=y
 # CONFIG_ADAPTEC_STARFIRE is not set
 # CONFIG_APRICOT is not set
 # CONFIG_CS89x0 is not set
-# CONFIG_DE2104X is not set
-# CONFIG_TULIP is not set
-# CONFIG_DE4X5 is not set
 # CONFIG_DGRS is not set
-# CONFIG_DM9102 is not set
 CONFIG_EEPRO100=y
+# CONFIG_E100 is not set
 # CONFIG_LNE390 is not set
 # CONFIG_FEALNX is not set
 # CONFIG_NATSEMI is not set
@@ -306,20 +320,21 @@ CONFIG_EEPRO100=y
 # CONFIG_TLAN is not set
 # CONFIG_VIA_RHINE is not set
 # CONFIG_VIA_RHINE_MMIO is not set
-# CONFIG_WINBOND_840 is not set
 # CONFIG_NET_POCKET is not set
 
 #
 # Ethernet (1000 Mbit)
 #
 CONFIG_ACENIC=y
-# CONFIG_ACENIC_OMIT_TIGON_I is not set
+CONFIG_ACENIC_OMIT_TIGON_I=y
 # CONFIG_DL2K is not set
+# CONFIG_E1000 is not set
 # CONFIG_MYRI_SBUS is not set
 # CONFIG_NS83820 is not set
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_SK98LIN is not set
+# CONFIG_TIGON3 is not set
 # CONFIG_FDDI is not set
 # CONFIG_HIPPI is not set
 # CONFIG_PLIP is not set
@@ -337,6 +352,7 @@ CONFIG_ACENIC=y
 CONFIG_TR=y
 CONFIG_IBMOL=y
 # CONFIG_IBMLS is not set
+# CONFIG_3C359 is not set
 # CONFIG_TMS380TR is not set
 # CONFIG_NET_FC is not set
 # CONFIG_RCPCI is not set
@@ -347,6 +363,11 @@ CONFIG_IBMOL=y
 #
 # CONFIG_WAN is not set
 
+#
+# "Tulip" family network device support
+#
+# CONFIG_NET_TULIP is not set
+
 #
 # Amateur Radio support
 #
@@ -360,7 +381,7 @@ CONFIG_IBMOL=y
 #
 # ISDN subsystem
 #
-# CONFIG_ISDN is not set
+# CONFIG_ISDN_BOOL is not set
 
 #
 # Old CD-ROM drivers (not SCSI, not IDE)
@@ -414,6 +435,25 @@ CONFIG_FONT_8x16=y
 # CONFIG_FONT_SUN8x16 is not set
 # CONFIG_FONT_PEARL_8x8 is not set
 # CONFIG_FONT_ACORN_8x8 is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+# CONFIG_INPUT_KEYBDEV is not set
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_GAMEPORT is not set
+CONFIG_SOUND_GAMEPORT=y
+# CONFIG_GAMEPORT_NS558 is not set
+# CONFIG_GAMEPORT_L4 is not set
+# CONFIG_INPUT_EMU10K1 is not set
+# CONFIG_GAMEPORT_PCIGAME is not set
+# CONFIG_GAMEPORT_FM801 is not set
+# CONFIG_GAMEPORT_CS461x is not set
+# CONFIG_SERIO is not set
+# CONFIG_SERIO_SERPORT is not set
 CONFIG_VIOPATH=y
 
 #
@@ -460,6 +500,11 @@ CONFIG_PSMOUSE=y
 # CONFIG_AGP is not set
 # CONFIG_DRM is not set
 
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+
 #
 # File systems
 #
@@ -490,6 +535,9 @@ CONFIG_RAMFS=y
 CONFIG_ISO9660_FS=y
 # CONFIG_JOLIET is not set
 # CONFIG_ZISOFS is not set
+CONFIG_JFS_FS=y
+# CONFIG_JFS_DEBUG is not set
+# CONFIG_JFS_STATISTICS is not set
 # CONFIG_MINIX_FS is not set
 # CONFIG_VXFS_FS is not set
 # CONFIG_NTFS_FS is not set
@@ -520,9 +568,11 @@ CONFIG_NFS_V3=y
 # CONFIG_ROOT_NFS is not set
 CONFIG_NFSD=y
 CONFIG_NFSD_V3=y
+# CONFIG_NFSD_TCP is not set
 CONFIG_SUNRPC=y
 CONFIG_LOCKD=y
 CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
 CONFIG_SMB_FS=y
 # CONFIG_SMB_NLS_DEFAULT is not set
 # CONFIG_NCP_FS is not set
@@ -569,6 +619,7 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_CODEPAGE_949 is not set
 # CONFIG_NLS_CODEPAGE_874 is not set
 # CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
 # CONFIG_NLS_CODEPAGE_1251 is not set
 # CONFIG_NLS_ISO8859_1 is not set
 # CONFIG_NLS_ISO8859_2 is not set
@@ -596,119 +647,18 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_USB is not set
 
 #
-# USB Host Controller Drivers
-#
-# CONFIG_USB_EHCI_HCD is not set
-# CONFIG_USB_OHCI_HCD is not set
-# CONFIG_USB_UHCI is not set
-# CONFIG_USB_UHCI_ALT is not set
-# CONFIG_USB_OHCI is not set
-
-#
-# USB Device Class drivers
-#
-# CONFIG_USB_AUDIO is not set
-# CONFIG_USB_BLUETOOTH is not set
-# CONFIG_USB_STORAGE is not set
-# CONFIG_USB_STORAGE_DEBUG is not set
-# CONFIG_USB_STORAGE_DATAFAB is not set
-# CONFIG_USB_STORAGE_FREECOM is not set
-# CONFIG_USB_STORAGE_ISD200 is not set
-# CONFIG_USB_STORAGE_DPCM is not set
-# CONFIG_USB_STORAGE_HP8200e is not set
-# CONFIG_USB_STORAGE_SDDR09 is not set
-# CONFIG_USB_STORAGE_JUMPSHOT is not set
-# CONFIG_USB_ACM is not set
-# CONFIG_USB_PRINTER is not set
-
-#
-# USB Human Interface Devices (HID)
-#
-# CONFIG_USB_HID is not set
-# CONFIG_USB_HIDDEV is not set
-# CONFIG_USB_KBD is not set
-# CONFIG_USB_MOUSE is not set
-# CONFIG_USB_WACOM is not set
-
-#
-# USB Imaging devices
-#
-# CONFIG_USB_DC2XX is not set
-# CONFIG_USB_MDC800 is not set
-# CONFIG_USB_SCANNER is not set
-# CONFIG_USB_MICROTEK is not set
-# CONFIG_USB_HPUSBSCSI is not set
-
-#
-# USB Multimedia devices
-#
-# CONFIG_USB_IBMCAM is not set
-# CONFIG_USB_OV511 is not set
-# CONFIG_USB_PWC is not set
-# CONFIG_USB_SE401 is not set
-# CONFIG_USB_STV680 is not set
-# CONFIG_USB_VICAM is not set
-# CONFIG_USB_DSBR is not set
-# CONFIG_USB_DABUSB is not set
-# CONFIG_USB_KONICAWC is not set
-
-#
-# USB Network adaptors
-#
-# CONFIG_USB_PEGASUS is not set
-# CONFIG_USB_KAWETH is not set
-# CONFIG_USB_CATC is not set
-# CONFIG_USB_CDCETHER is not set
-# CONFIG_USB_USBNET is not set
-
-#
-# USB port drivers
-#
-# CONFIG_USB_USS720 is not set
-
-#
-# USB Serial Converter support
-#
-# CONFIG_USB_SERIAL is not set
-# CONFIG_USB_SERIAL_GENERIC is not set
-# CONFIG_USB_SERIAL_BELKIN is not set
-# CONFIG_USB_SERIAL_WHITEHEAT is not set
-# CONFIG_USB_SERIAL_DIGI_ACCELEPORT is not set
-# CONFIG_USB_SERIAL_EMPEG is not set
-# CONFIG_USB_SERIAL_FTDI_SIO is not set
-# CONFIG_USB_SERIAL_VISOR is not set
-# CONFIG_USB_SERIAL_IPAQ is not set
-# CONFIG_USB_SERIAL_IR is not set
-# CONFIG_USB_SERIAL_EDGEPORT is not set
-# CONFIG_USB_SERIAL_KEYSPAN_PDA is not set
-# CONFIG_USB_SERIAL_KEYSPAN is not set
-# CONFIG_USB_SERIAL_KEYSPAN_USA28 is not set
-# CONFIG_USB_SERIAL_KEYSPAN_USA28X is not set
-# CONFIG_USB_SERIAL_KEYSPAN_USA28XA is not set
-# CONFIG_USB_SERIAL_KEYSPAN_USA28XB is not set
-# CONFIG_USB_SERIAL_KEYSPAN_USA19 is not set
-# CONFIG_USB_SERIAL_KEYSPAN_USA18X is not set
-# CONFIG_USB_SERIAL_KEYSPAN_USA19W is not set
-# CONFIG_USB_SERIAL_KEYSPAN_USA49W is not set
-# CONFIG_USB_SERIAL_MCT_U232 is not set
-# CONFIG_USB_SERIAL_KLSI is not set
-# CONFIG_USB_SERIAL_PL2303 is not set
-# CONFIG_USB_SERIAL_CYBERJACK is not set
-# CONFIG_USB_SERIAL_XIRCOM is not set
-# CONFIG_USB_SERIAL_OMNINET is not set
-
-#
-# USB Miscellaneous drivers
+# Bluetooth support
 #
-# CONFIG_USB_RIO500 is not set
-# CONFIG_USB_AUERSWALD is not set
+# CONFIG_BLUEZ is not set
 
 #
 # Kernel hacking
 #
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_DEBUG_SLAB is not set
 CONFIG_MAGIC_SYSRQ=y
-# CONFIG_KGDB is not set
 CONFIG_XMON=y
+CONFIG_XMON_DEFAULT=y
 # CONFIG_PPCDBG is not set
 
 #

arch/ppc64/kernel/Makefile

@@ -38,8 +38,6 @@ obj-$(CONFIG_PCI) += pSeries_pci.o pSeries_lpar.o pSeries_hvCall.o eeh.o
 obj-y += rtasd.o nvram.o
 endif
 
-obj-$(CONFIG_KGDB) += ppc-stub.o
-
 obj-$(CONFIG_SMP) += smp.o
 
 obj-y += prom.o lmb.o rtas.o rtas-proc.o chrp_setup.o i8259.o

arch/ppc64/kernel/entry.S

@@ -56,7 +56,6 @@ show_syscalls_task:
  * Handle a system call.
  */
 _GLOBAL(DoSyscall)
-	std	r0,THREAD+LAST_SYSCALL(r13)
 	ld	r11,_CCR(r1)	/* Clear SO bit in CR */
 	lis	r10,0x1000
 	andc	r11,r11,r10

arch/ppc64/kernel/idle.c

@@ -84,6 +84,8 @@ int cpu_idle(void)
 	lpaca = get_paca();
 
 	while (1) {
+		irq_stat[smp_processor_id()].idle_timestamp = jiffies;
+
 		if (lpaca->xLpPaca.xSharedProc) {
 			if (ItLpQueue_isLpIntPending(lpaca->lpQueuePtr))
 				process_iSeries_events();
@@ -123,6 +125,7 @@ int cpu_idle(void)
 	long oldval;
 
 	while (1) {
+		irq_stat[smp_processor_id()].idle_timestamp = jiffies;
 		oldval = test_and_clear_thread_flag(TIF_NEED_RESCHED);
 
 		if (!oldval) {
@@ -146,3 +149,8 @@ int cpu_idle(void)
 }
 
 #endif /* CONFIG_PPC_ISERIES */
+
+void default_idle(void)
+{
+	barrier();
+}

arch/ppc64/kernel/ioctl32.c

@@ -3756,6 +3756,7 @@ COMPATIBLE_IOCTL(TCSETSW),
 COMPATIBLE_IOCTL(TCSETSF),
 COMPATIBLE_IOCTL(TIOCLINUX),
 COMPATIBLE_IOCTL(TIOCSTART),
+COMPATIBLE_IOCTL(TIOCSTOP),
 /* Little t */
 COMPATIBLE_IOCTL(TIOCGETD),
 COMPATIBLE_IOCTL(TIOCSETD),
@@ -4336,8 +4337,6 @@ COMPATIBLE_IOCTL(RNDCLEARPOOL),
 COMPATIBLE_IOCTL(HCIDEVUP),
 COMPATIBLE_IOCTL(HCIDEVDOWN),
 COMPATIBLE_IOCTL(HCIDEVRESET),
-COMPATIBLE_IOCTL(HCIRESETSTAT),
-COMPATIBLE_IOCTL(HCIGETINFO),
 COMPATIBLE_IOCTL(HCIGETDEVLIST),
 COMPATIBLE_IOCTL(HCISETRAW),
 COMPATIBLE_IOCTL(HCISETSCAN),

arch/ppc64/kernel/irq.c

@@ -75,22 +75,6 @@ irq_desc_t irq_desc[NR_IRQS] __cacheline_aligned =
 int ppc_spurious_interrupts = 0;
 struct irqaction *ppc_irq_action[NR_IRQS];
 unsigned long lpEvent_count = 0;
-#ifdef CONFIG_XMON
-extern void xmon(struct pt_regs *regs);
-extern int xmon_bpt(struct pt_regs *regs);
-extern int xmon_sstep(struct pt_regs *regs);
-extern int xmon_iabr_match(struct pt_regs *regs);
-extern int xmon_dabr_match(struct pt_regs *regs);
-extern void (*xmon_fault_handler)(struct pt_regs *regs);
-#endif
-#ifdef CONFIG_XMON
-extern void (*debugger)(struct pt_regs *regs);
-extern int (*debugger_bpt)(struct pt_regs *regs);
-extern int (*debugger_sstep)(struct pt_regs *regs);
-extern int (*debugger_iabr_match)(struct pt_regs *regs);
-extern int (*debugger_dabr_match)(struct pt_regs *regs);
-extern void (*debugger_fault_handler)(struct pt_regs *regs);
-#endif
 
 /* nasty hack for shared irq's since we need to do kmalloc calls but
  * can't very early in the boot when we need to do a request irq.
@@ -410,6 +394,75 @@ handle_irq_event(int irq, struct pt_regs *regs, struct irqaction *action)
 	__cli();
 }
 
+#ifdef CONFIG_SMP
+extern unsigned int irq_affinity [NR_IRQS];
+
+typedef struct {
+	unsigned long cpu;
+	unsigned long timestamp;
+} ____cacheline_aligned irq_balance_t;
+
+static irq_balance_t irq_balance[NR_IRQS] __cacheline_aligned
+			= { [ 0 ... NR_IRQS-1 ] = { 1, 0 } };
+
+#define IDLE_ENOUGH(cpu,now) \
+		(idle_cpu(cpu) && ((now) - irq_stat[(cpu)].idle_timestamp > ((HZ/100)+1)))
+
+#define IRQ_ALLOWED(cpu,allowed_mask) \
+		((1 << cpu) & (allowed_mask))
+
+static unsigned long move(unsigned long curr_cpu, unsigned long allowed_mask,
+			  unsigned long now, int direction)
+{
+	int search_idle = 1;
+	int cpu = curr_cpu;
+
+	goto inside;
+
+	do {
+		if (unlikely(cpu == curr_cpu))
+			search_idle = 0;
+inside:
+		if (direction == 1) {
+			cpu++;
+			if (cpu >= smp_num_cpus)
+				cpu = 0;
+		} else {
+			cpu--;
+			if (cpu == -1)
+				cpu = smp_num_cpus-1;
+		}
+	} while (!IRQ_ALLOWED(cpu,allowed_mask) ||
+			(search_idle && !IDLE_ENOUGH(cpu,now)));
+
+	return cpu;
+}
+
+static inline void balance_irq(int irq)
+{
+	irq_balance_t *entry = irq_balance + irq;
+	unsigned long now = jiffies;
+
+	if (unlikely(entry->timestamp != now)) {
+		unsigned long allowed_mask;
+		unsigned long random_number;
+
+		if (!irq_desc[irq].handler->set_affinity)
+			return;
+
+		random_number = mftb();
+		random_number &= 1;
+
+		allowed_mask = cpu_online_map & irq_affinity[irq];
+		entry->timestamp = now;
+		entry->cpu = move(entry->cpu, allowed_mask, now, random_number);
+		irq_desc[irq].handler->set_affinity(irq, 1 << entry->cpu);
+	}
+}
+#else
+#define balance_irq(irq) do { } while (0)
+#endif
+
 /*
  * Eventually, this should take an array of interrupts and an array size
  * so it can dispatch multiple interrupts.
@@ -421,6 +474,8 @@ void ppc_irq_dispatch_handler(struct pt_regs *regs, int irq)
 	int cpu = smp_processor_id();
 	irq_desc_t *desc = irq_desc + irq;
 
+	balance_irq(irq);
+
 	kstat.irqs[cpu][irq]++;
 	spin_lock(&desc->lock);
 	ack_irq(irq);	

arch/ppc64/kernel/misc.S

@@ -265,24 +265,6 @@ _GLOBAL(__flush_dcache_icache)
 	isync
 	blr
 	
-/*
- * Copy a whole page.  Assumes a 4096B page size.
- */
-_GLOBAL(copy_page)
-	clrrdi	r3,r3,12           	    /* Page align */
-	clrrdi	r4,r4,12           	    /* Page align */
-	li	r5,256
-	mtctr	r5
-	addi	r3,r3,-8
-	addi	r4,r4,-8
-
-1:      ld 	r6,8(r4)
-        ldu	r7,16(r4)
-	std     r6,8(r3) 
-	stdu    r7,16(r3) 
-	bdnz+	1b
-	blr  
-
 /*
  * I/O string operations
  *
@@ -740,7 +722,11 @@ _GLOBAL(sys_call_table32)
 	.llong .sys_lremovexattr
 	.llong .sys_fremovexattr	/* 220 */
 	.llong .sys_futex
-	.rept NR_syscalls-221
+	.llong .sys_ni_syscall		/* reserved for tux */
+	.llong .sys32_sched_setaffinity
+	.llong .sys32_sched_getaffinity
+
+	.rept NR_syscalls-224
 		.llong .sys_ni_syscall
 	.endr
 #endif
@@ -969,6 +955,10 @@ _GLOBAL(sys_call_table)
 	.llong .sys_lremovexattr
 	.llong .sys_fremovexattr	/* 220 */
 	.llong .sys_futex
-	.rept NR_syscalls-221
+	.llong .sys_ni_syscall		/* reserved for tux */
+	.llong .sys_sched_setaffinity
+	.llong .sys_sched_getaffinity
+
+	.rept NR_syscalls-224
 	.llong .sys_ni_syscall
 	.endr

arch/ppc64/kernel/mk_defs.c

@@ -52,7 +52,6 @@ main(void)
 
 	/* task_struct->thread */
 	DEFINE(THREAD, offsetof(struct task_struct, thread));
-	DEFINE(LAST_SYSCALL, offsetof(struct thread_struct, last_syscall));
 	DEFINE(PT_REGS, offsetof(struct thread_struct, regs));
 	DEFINE(THREAD_FPR0, offsetof(struct thread_struct, fpr[0]));
 	DEFINE(THREAD_FPSCR, offsetof(struct thread_struct, fpscr));

arch/ppc64/kernel/pSeries_htab.c

@@ -359,7 +359,7 @@ static void pSeries_hpte_invalidate(unsigned long slot, unsigned long va,
 
 	/* Invalidate the tlb */
 	if (!large && local && __is_processor(PV_POWER4)) {
-		_tlbiel(va, large);
+		_tlbiel(va);
 	} else {
 		spin_lock_irqsave(&pSeries_tlbie_lock, flags);
 		_tlbie(va, large);

arch/ppc64/kernel/pci.c

@@ -498,6 +498,8 @@ pcibios_init(void)
 
 }
 
+subsys_initcall(pcibios_init);
+
 int __init
 pcibios_assign_all_busses(void)
 {

arch/ppc64/kernel/ppc_ksyms.c

@@ -257,7 +257,7 @@ EXPORT_SYMBOL(console_drivers);
 EXPORT_SYMBOL(xmon);
 #endif
 
-#if defined(CONFIG_KGDB) || defined(CONFIG_XMON)
+#ifdef CONFIG_DEBUG_KERNEL
 extern void (*debugger)(struct pt_regs *regs);
 extern int (*debugger_bpt)(struct pt_regs *regs);
 extern int (*debugger_sstep)(struct pt_regs *regs);

arch/ppc64/kernel/process.c

@@ -117,7 +117,6 @@ void show_regs(struct pt_regs * regs)
 	       regs->msr&MSR_DR ? 1 : 0);
 	printk("TASK = %p[%d] '%s' ",
 	       current, current->pid, current->comm);
-	printk("Last syscall: %ld ", current->thread.last_syscall);
 	printk("\nlast math %p ", last_task_used_math);
 	
 #ifdef CONFIG_SMP
@@ -223,8 +222,6 @@ copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 	memcpy(&p->thread.fpr, &current->thread.fpr, sizeof(p->thread.fpr));
 	p->thread.fpscr = current->thread.fpscr;
 
-	p->thread.last_syscall = -1;
-
 	return 0;
 }
 

arch/ppc64/kernel/setup.c

@@ -130,6 +130,14 @@ void setup_system(unsigned long r3, unsigned long r4, unsigned long r5,
 	/* This should be fixed properly in kernel/resource.c */
 	iomem_resource.end = MEM_SPACE_LIMIT;
 
+#ifdef CONFIG_XMON_DEFAULT
+	debugger = xmon;
+	debugger_bpt = xmon_bpt;
+	debugger_sstep = xmon_sstep;
+	debugger_iabr_match = xmon_iabr_match;
+	debugger_dabr_match = xmon_dabr_match;
+#endif
+
 	/* pSeries systems are identified in prom.c via OF. */
 	if ( itLpNaca.xLparInstalled == 1 )
 		naca->platform = PLATFORM_ISERIES_LPAR;
@@ -507,11 +515,6 @@ void __init setup_arch(char **cmdline_p)
 
 	ppc_md.progress("setup_arch:enter", 0x3eab);
 
-#if defined(CONFIG_KGDB)
-	kgdb_map_scc();
-	set_debug_traps();
-	breakpoint();
-#endif
 	/*
 	 * Set cache line size based on type of cpu as a default.
 	 * Systems with OF can look in the properties on the cpu node(s)
@@ -543,26 +546,6 @@ void __init setup_arch(char **cmdline_p)
 	ppc_md.progress("setup_arch: exit", 0x3eab);
 }
 
-void exception_trace(unsigned long trap)
-{
-	unsigned long x, srr0, srr1, reg20, reg1, reg21;
-
-	asm("mflr %0" : "=r" (x) :);
-	asm("mfspr %0,0x1a" : "=r" (srr0) :);
-	asm("mfspr %0,0x1b" : "=r" (srr1) :);
-	asm("mr %0,1" : "=r" (reg1) :);
-	asm("mr %0,20" : "=r" (reg20) :);
-	asm("mr %0,21" : "=r" (reg21) :);
-
-	udbg_puts("\n");
-	udbg_puts("Took an exception : "); udbg_puthex(x); udbg_puts("\n");
-	udbg_puts("   "); udbg_puthex(reg1); udbg_puts("\n");
-	udbg_puts("   "); udbg_puthex(reg20); udbg_puts("\n");
-	udbg_puts("   "); udbg_puthex(reg21); udbg_puts("\n");
-	udbg_puts("   "); udbg_puthex(srr0); udbg_puts("\n");
-	udbg_puts("   "); udbg_puthex(srr1); udbg_puts("\n");
-}
-
 int set_spread_lpevents( char * str )
 {
 	/* The parameter is the number of processors to share in processing lp events */

arch/ppc64/kernel/smp.c

@@ -58,7 +58,6 @@ volatile int smp_commenced = 0;
 int smp_num_cpus = 1;
 int smp_tb_synchronized = 0;
 spinlock_t kernel_flag __cacheline_aligned = SPIN_LOCK_UNLOCKED;
-cycles_t cacheflush_time;
 unsigned long cache_decay_ticks;
 static int max_cpus __initdata = NR_CPUS;
 
@@ -595,12 +594,13 @@ void __init smp_boot_cpus(void)
 	}
 
 	/*
-	 * XXX very rough, assumes 20 bus cycles to read a cache line,
-	 * timebase increments every 4 bus cycles, 32kB L1 data cache.
+	 * XXX very rough. On POWER4 we optimise tlb flushes for
+	 * tasks that only run on one cpu so we increase decay ticks.
 	 */
-	cacheflush_time = 5 * 1024;
-	/* XXX - Fix - Anton */
-	cache_decay_ticks = 0;
+	if (__is_processor(PV_POWER4))
+		cache_decay_ticks = HZ/50;
+	else
+		cache_decay_ticks = HZ/100;
 
 	/* Probe arch for CPUs */
 	cpu_nr = ppc_md.smp_probe();

arch/ppc64/kernel/sys_ppc32.c

@@ -4671,3 +4671,53 @@ asmlinkage long sys32_time(__kernel_time_t32* tloc)
 
 	return secs;
 }
+
+extern asmlinkage int sys_sched_setaffinity(pid_t pid, unsigned int len,
+					    unsigned long *user_mask_ptr);
+
+asmlinkage int sys32_sched_setaffinity(__kernel_pid_t32 pid, unsigned int len,
+				       u32 *user_mask_ptr)
+{
+	unsigned long kernel_mask;
+	mm_segment_t old_fs;
+	int ret;
+
+	if (get_user(kernel_mask, user_mask_ptr))
+		return -EFAULT;
+
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+	ret = sys_sched_setaffinity(pid,
+				    /* XXX Nice api... */
+				    sizeof(kernel_mask),
+				    &kernel_mask);
+	set_fs(old_fs);
+
+	return ret;
+}
+
+extern asmlinkage int sys_sched_getaffinity(pid_t pid, unsigned int len,
+					    unsigned long *user_mask_ptr);
+
+asmlinkage int sys32_sched_getaffinity(__kernel_pid_t32 pid, unsigned int len,
+				       u32 *user_mask_ptr)
+{
+	unsigned long kernel_mask;
+	mm_segment_t old_fs;
+	int ret;
+
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+	ret = sys_sched_getaffinity(pid,
+				    /* XXX Nice api... */
+				    sizeof(kernel_mask),
+				    &kernel_mask);
+	set_fs(old_fs);
+
+	if (ret == 0) {
+		if (put_user(kernel_mask, user_mask_ptr))
+			ret = -EFAULT;
+	}
+
+	return ret;
+}

arch/ppc64/kernel/traps.c

@@ -38,29 +38,12 @@
 #include <asm/ppcdebug.h>
 
 extern int fix_alignment(struct pt_regs *);
-extern void bad_page_fault(struct pt_regs *, unsigned long);
+extern void bad_page_fault(struct pt_regs *, unsigned long, int);
 
 /* This is true if we are using the firmware NMI handler (typically LPAR) */
 extern int fwnmi_active;
 
-#ifdef CONFIG_XMON
-extern void xmon(struct pt_regs *regs);
-extern int xmon_bpt(struct pt_regs *regs);
-extern int xmon_sstep(struct pt_regs *regs);
-extern int xmon_iabr_match(struct pt_regs *regs);
-extern int xmon_dabr_match(struct pt_regs *regs);
-extern void (*xmon_fault_handler)(struct pt_regs *regs);
-#endif
-
-#ifdef CONFIG_XMON
-void (*debugger)(struct pt_regs *regs) = xmon;
-int (*debugger_bpt)(struct pt_regs *regs) = xmon_bpt;
-int (*debugger_sstep)(struct pt_regs *regs) = xmon_sstep;
-int (*debugger_iabr_match)(struct pt_regs *regs) = xmon_iabr_match;
-int (*debugger_dabr_match)(struct pt_regs *regs) = xmon_dabr_match;
-void (*debugger_fault_handler)(struct pt_regs *regs);
-#else
-#ifdef CONFIG_KGDB
+#ifdef CONFIG_DEBUG_KERNEL
 void (*debugger)(struct pt_regs *regs);
 int (*debugger_bpt)(struct pt_regs *regs);
 int (*debugger_sstep)(struct pt_regs *regs);
@@ -68,30 +51,44 @@ int (*debugger_iabr_match)(struct pt_regs *regs);
 int (*debugger_dabr_match)(struct pt_regs *regs);
 void (*debugger_fault_handler)(struct pt_regs *regs);
 #endif
-#endif
+
 /*
  * Trap & Exception support
  */
 
-void
-_exception(int signr, struct pt_regs *regs)
+/* Should we panic on bad kernel exceptions or try to recover */
+#undef PANIC_ON_ERROR
+
+static spinlock_t die_lock = SPIN_LOCK_UNLOCKED;
+
+void die(const char *str, struct pt_regs *regs, long err)
 {
-	if (!user_mode(regs))
-	{
+	console_verbose();
+	spin_lock_irq(&die_lock);
+	bust_spinlocks(1);
+	printk("Oops: %s, sig: %ld\n", str, err);
 	show_regs(regs);
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
-		debugger(regs);
-#endif
 	print_backtrace((unsigned long *)regs->gpr[1]);
-		panic("Exception in kernel pc %lx signal %d",regs->nip,signr);
-#if defined(CONFIG_PPCDBG) && (defined(CONFIG_XMON) || defined(CONFIG_KGDB))
-	/* Allow us to catch SIGILLs for 64-bit app/glibc debugging. -Peter */
-	} else if (signr == SIGILL) {
-		ifppcdebug(PPCDBG_SIGNALXMON)
-			debugger(regs);
+	bust_spinlocks(0);
+	spin_unlock_irq(&die_lock);
+
+#ifdef PANIC_ON_ERROR
+	panic(str);
+#else
+	do_exit(SIGSEGV);
 #endif
+}
+
+static void
+_exception(int signr, siginfo_t *info, struct pt_regs *regs)
+{
+	if (!user_mode(regs)) {
+		if (debugger)
+			debugger(regs);
+		die("Exception in kernel mode\n", regs, signr);
 	}
-	force_sig(signr, current);
+
+	force_sig_info(signr, info, current);
 }
 
 /* Get the error information for errors coming through the
@@ -130,9 +127,8 @@ static void FWNMI_release_errinfo(void)
 void
 SystemResetException(struct pt_regs *regs)
 {
-	char *msg = "System Reset in kernel mode.\n";
-	udbg_printf(msg); printk(msg);
 	if (fwnmi_active) {
+		char *msg;
 		unsigned long *r3 = __va(regs->gpr[3]); /* for FWNMI debug */
 		struct rtas_error_log *errlog;
 
@@ -140,17 +136,31 @@ SystemResetException(struct pt_regs *regs)
 		udbg_printf(msg, r3); printk(msg, r3);
 		errlog = FWNMI_get_errinfo(regs);
 	}
-#if defined(CONFIG_XMON)
-	xmon(regs);
-	udbg_printf("leaving xmon...\n");
+
+	if (debugger)
+		debugger(regs);
+
+#ifdef PANIC_ON_ERROR
+	panic("System Reset");
 #else
-	for(;;);
+	/* Must die if the interrupt is not recoverable */
+	if (!(regs->msr & MSR_RI))
+		panic("Unrecoverable System Reset");
 #endif
+
+	/* What should we do here? We could issue a shutdown or hard reset. */
+}
+
+static int power4_handle_mce(struct pt_regs *regs)
+{
+	return 0;
 }
 
 void
 MachineCheckException(struct pt_regs *regs)
 {
+	siginfo_t info;
+
 	if (fwnmi_active) {
 		struct rtas_error_log *errhdr = FWNMI_get_errinfo(regs);
 		if (errhdr) {
@@ -158,117 +168,221 @@ MachineCheckException(struct pt_regs *regs)
 		}
 		FWNMI_release_errinfo();
 	}
-	if ( !user_mode(regs) )
-	{
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
+
+	if (!user_mode(regs)) {
+		/* Attempt to recover if the interrupt is recoverable */
+		if (regs->msr & MSR_RI) {
+			if (__is_processor(PV_POWER4) &&
+			    power4_handle_mce(regs))
+				return;
+		}
+
 		if (debugger_fault_handler) {
 			debugger_fault_handler(regs);
 			return;
 		}
-#endif
+		if (debugger)
+			debugger(regs);
+
+		console_verbose();
+		spin_lock_irq(&die_lock);
+		bust_spinlocks(1);
 		printk("Machine check in kernel mode.\n");
 		printk("Caused by (from SRR1=%lx): ", regs->msr);
 		show_regs(regs);
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
-		debugger(regs);
-#endif
 		print_backtrace((unsigned long *)regs->gpr[1]);
-		panic("machine check");
+		bust_spinlocks(0);
+		spin_unlock_irq(&die_lock);
+		panic("Unrecoverable Machine Check");
 	}
-	_exception(SIGSEGV, regs);	
-}
 
-void
-SMIException(struct pt_regs *regs)
-{
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
-	{
-		debugger(regs);
-		return;
-	}
-#endif
-	show_regs(regs);
-	print_backtrace((unsigned long *)regs->gpr[1]);
-	panic("System Management Interrupt");
+	/*
+	 * XXX we should check RI bit on exception exit and kill the
+	 * task if it was cleared
+	 */
+	info.si_signo = SIGBUS;
+	info.si_errno = 0;
+	info.si_code = BUS_ADRERR;
+	info.si_addr = (void *)regs->nip;
+	_exception(SIGSEGV, &info, regs);
 }
 
 void
 UnknownException(struct pt_regs *regs)
 {
+	siginfo_t info;
+
 	printk("Bad trap at PC: %lx, SR: %lx, vector=%lx\n",
 	       regs->nip, regs->msr, regs->trap);
-	_exception(SIGTRAP, regs);	
+
+	info.si_signo = SIGTRAP;
+	info.si_errno = 0;
+	info.si_code = 0;
+	info.si_addr = 0;
+	_exception(SIGTRAP, &info, regs);	
 }
 
 void
 InstructionBreakpointException(struct pt_regs *regs)
 {
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
-	if (debugger_iabr_match(regs))
+	siginfo_t info;
+
+	if (debugger_iabr_match && debugger_iabr_match(regs))
 		return;
-#endif
-	_exception(SIGTRAP, regs);
+
+	info.si_signo = SIGTRAP;
+	info.si_errno = 0;
+	info.si_code = TRAP_BRKPT;
+	info.si_addr = (void *)regs->nip;
+	_exception(SIGTRAP, &info, regs);
+}
+
+static void parse_fpe(struct pt_regs *regs)
+{
+	siginfo_t info;
+	unsigned int *tmp;
+	unsigned int fpscr;
+
+	if (regs->msr & MSR_FP)
+		giveup_fpu(current);
+
+	tmp = &current->thread.fpscr;
+	fpscr = *tmp;
+
+	/* Invalid operation */
+	if ((fpscr & FPSCR_VE) && (fpscr & FPSCR_VX))
+		info.si_code = FPE_FLTINV;
+
+	/* Overflow */
+	else if ((fpscr & FPSCR_OE) && (fpscr & FPSCR_OX))
+		info.si_code = FPE_FLTOVF;
+
+	/* Underflow */
+	else if ((fpscr & FPSCR_UE) && (fpscr & FPSCR_UX))
+		info.si_code = FPE_FLTUND;
+
+	/* Divide by zero */
+	else if ((fpscr & FPSCR_ZE) && (fpscr & FPSCR_ZX))
+		info.si_code = FPE_FLTDIV;
+
+	/* Inexact result */
+	else if ((fpscr & FPSCR_XE) && (fpscr & FPSCR_XX))
+		info.si_code = FPE_FLTRES;
+
+	else
+		info.si_code = 0;
+
+	info.si_signo = SIGFPE;
+	info.si_errno = 0;
+	info.si_addr = (void *)regs->nip;
+	_exception(SIGFPE, &info, regs);
 }
 
 void
 ProgramCheckException(struct pt_regs *regs)
 {
+	siginfo_t info;
+
 	if (regs->msr & 0x100000) {
 		/* IEEE FP exception */
-		_exception(SIGFPE, regs);
+
+		parse_fpe(regs);
+	} else if (regs->msr & 0x40000) {
+		/* Privileged instruction */
+
+		info.si_signo = SIGILL;
+		info.si_errno = 0;
+		info.si_code = ILL_PRVOPC;
+		info.si_addr = (void *)regs->nip;
+		_exception(SIGILL, &info, regs);
 	} else if (regs->msr & 0x20000) {
 		/* trap exception */
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
-		if (debugger_bpt(regs))
+
+		if (debugger_bpt && debugger_bpt(regs))
 			return;
-#endif
-		_exception(SIGTRAP, regs);
+
+		info.si_signo = SIGTRAP;
+		info.si_errno = 0;
+		info.si_code = TRAP_BRKPT;
+		info.si_addr = (void *)regs->nip;
+		_exception(SIGTRAP, &info, regs);
 	} else {
-		_exception(SIGILL, regs);
+		/* Illegal instruction */
+
+		info.si_signo = SIGILL;
+		info.si_errno = 0;
+		info.si_code = ILL_ILLTRP;
+		info.si_addr = (void *)regs->nip;
+		_exception(SIGILL, &info, regs);
 	}
 }
 
 void
 SingleStepException(struct pt_regs *regs)
 {
+	siginfo_t info;
+
 	regs->msr &= ~MSR_SE;  /* Turn off 'trace' bit */
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
-	if (debugger_sstep(regs))
+
+	if (debugger_sstep && debugger_sstep(regs))
 		return;
-#endif
-	_exception(SIGTRAP, regs);	
-}
 
-/* Dummy handler for Performance Monitor */
+	info.si_signo = SIGTRAP;
+	info.si_errno = 0;
+	info.si_code = TRAP_TRACE;
+	info.si_addr = (void *)regs->nip;
+	_exception(SIGTRAP, &info, regs);	
+}
 
 void
 PerformanceMonitorException(struct pt_regs *regs)
 {
-	_exception(SIGTRAP, regs);
+	siginfo_t info;
+
+	info.si_signo = SIGTRAP;
+	info.si_errno = 0;
+	info.si_code = TRAP_BRKPT;
+	info.si_addr = 0;
+	_exception(SIGTRAP, &info, regs);
 }
 
 void
 AlignmentException(struct pt_regs *regs)
 {
 	int fixed;
+	siginfo_t info;
 
 	fixed = fix_alignment(regs);
+
 	if (fixed == 1) {
-		ifppcdebug(PPCDBG_ALIGNFIXUP)
 		if (!user_mode(regs))
-				PPCDBG(PPCDBG_ALIGNFIXUP, "fix alignment at %lx\n", regs->nip);
+			PPCDBG(PPCDBG_ALIGNFIXUP, "fix alignment at %lx\n",
+			       regs->nip);
 		regs->nip += 4;	/* skip over emulated instruction */
 		return;
 	}
+
+	/* Operand address was bad */	
 	if (fixed == -EFAULT) {
-		/* fixed == -EFAULT means the operand address was bad */
-		if (user_mode(regs))
-			force_sig(SIGSEGV, current);
-		else
-			bad_page_fault(regs, regs->dar);
+		if (user_mode(regs)) {
+			info.si_signo = SIGSEGV;
+			info.si_errno = 0;
+			info.si_code = SEGV_MAPERR;
+			info.si_addr = (void *)regs->dar;
+			force_sig_info(SIGSEGV, &info, current);
+		} else {
+			/* Search exception table */
+			bad_page_fault(regs, regs->dar, SIGSEGV);
+		}
+
 		return;
 	}
-	_exception(SIGBUS, regs);	
+
+	info.si_signo = SIGBUS;
+	info.si_errno = 0;
+	info.si_code = BUS_ADRALN;
+	info.si_addr = (void *)regs->nip;
+	_exception(SIGBUS, &info, regs);	
 }
 
 void __init trap_init(void)

arch/ppc64/lib/Makefile

@@ -4,6 +4,7 @@
 
 O_TARGET = lib.o
 
-obj-y           := checksum.o dec_and_lock.o string.o strcase.o
+obj-y           := checksum.o dec_and_lock.o string.o strcase.o copypage.o \
+		   memcpy.o copyuser.o
 
 include $(TOPDIR)/Rules.make

arch/ppc64/lib/copypage.S

+/*
+ * arch/ppc64/lib/copypage.S
+ *
+ * Copyright (C) 2002 Paul Mackerras, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <asm/processor.h>
+#include "../kernel/ppc_asm.h"
+
+_GLOBAL(copy_page)
+	std	r31,-8(1)
+	std	r30,-16(1)
+	std	r29,-24(1)
+	std	r28,-32(1)
+	std	r27,-40(1)
+	std	r26,-48(1)
+	std	r25,-56(1)
+	std	r24,-64(1)
+	std	r23,-72(1)
+	std	r22,-80(1)
+	std	r21,-88(1)
+	std	r20,-96(1)
+	li	r5,4096/32 - 1
+	addi	r3,r3,-8
+	li	r12,5
+0:	addi	r5,r5,-24
+	mtctr	r12
+	ld	r22,640(4)
+	ld	r21,512(4)
+	ld	r20,384(4)
+	ld	r11,256(4)
+	ld	r9,128(4)
+	ld	r7,0(4)
+	ld	r25,648(4)
+	ld	r24,520(4)
+	ld	r23,392(4)
+	ld	r10,264(4)
+	ld	r8,136(4)
+	ldu	r6,8(4)
+	cmpwi	r5,24
+1:	std	r22,648(3)
+	std	r21,520(3)
+	std	r20,392(3)
+	std	r11,264(3)
+	std	r9,136(3)
+	std	r7,8(3)
+	ld	r28,648(4)
+	ld	r27,520(4)
+	ld	r26,392(4)
+	ld	r31,264(4)
+	ld	r30,136(4)
+	ld	r29,8(4)
+	std	r25,656(3)
+	std	r24,528(3)
+	std	r23,400(3)
+	std	r10,272(3)
+	std	r8,144(3)
+	std	r6,16(3)
+	ld	r22,656(4)
+	ld	r21,528(4)
+	ld	r20,400(4)
+	ld	r11,272(4)
+	ld	r9,144(4)
+	ld	r7,16(4)
+	std	r28,664(3)
+	std	r27,536(3)
+	std	r26,408(3)
+	std	r31,280(3)
+	std	r30,152(3)
+	stdu	r29,24(3)
+	ld	r25,664(4)
+	ld	r24,536(4)
+	ld	r23,408(4)
+	ld	r10,280(4)
+	ld	r8,152(4)
+	ldu	r6,24(4)
+	bdnz	1b
+	std	r22,648(3)
+	std	r21,520(3)
+	std	r20,392(3)
+	std	r11,264(3)
+	std	r9,136(3)
+	std	r7,8(3)
+	addi	r4,r4,640
+	addi	r3,r3,648
+	bge	0b
+	mtctr	r5
+	ld	r7,0(4)
+	ld	r8,8(4)
+	ldu	r9,16(4)
+3:	ld	r10,8(4)
+	std	r7,8(3)
+	ld	r7,16(4)
+	std	r8,16(3)
+	ld	r8,24(4)
+	std	r9,24(3)
+	ldu	r9,32(4)
+	stdu	r10,32(3)
+	bdnz	3b
+4:	ld	r10,8(4)
+	std	r7,8(3)
+	std	r8,16(3)
+	std	r9,24(3)
+	std	r10,32(3)
+9:	ld	r20,-96(1)
+	ld	r21,-88(1)
+	ld	r22,-80(1)
+	ld	r23,-72(1)
+	ld	r24,-64(1)
+	ld	r25,-56(1)
+	ld	r26,-48(1)
+	ld	r27,-40(1)
+	ld	r28,-32(1)
+	ld	r29,-24(1)
+	ld	r30,-16(1)
+	ld	r31,-8(1)
+	blr

arch/ppc64/lib/copyuser.S

+/*
+ * arch/ppc64/lib/copyuser.S
+ *
+ * Copyright (C) 2002 Paul Mackerras, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <asm/processor.h>
+#include "../kernel/ppc_asm.h"
+
+	.align	7
+_GLOBAL(__copy_tofrom_user)
+	/* first check for a whole page copy on a page boundary */
+	cmpldi	cr1,r5,16
+	cmpdi	cr6,r5,4096
+	or	r0,r3,r4
+	neg	r6,r3		/* LS 3 bits = # bytes to 8-byte dest bdry */
+	andi.	r0,r0,4095
+	std	r3,-24(r1)
+	crand	cr0*4+2,cr0*4+2,cr6*4+2
+	std	r4,-16(r1)
+	std	r5,-8(r1)
+	dcbt	0,r4
+	beq	.Lcopy_page
+	andi.	r6,r6,7
+	mtcrf	0x01,r5
+	blt	cr1,.Lshort_copy
+	bne	.Ldst_unaligned
+.Ldst_aligned:
+	andi.	r0,r4,7
+	addi	r3,r3,-16
+	bne	.Lsrc_unaligned
+	srdi	r7,r5,4
+20:	ld	r9,0(r4)
+	addi	r4,r4,-8
+	mtctr	r7
+	andi.	r5,r5,7
+	bf	cr7*4+0,22f
+	addi	r3,r3,8
+	addi	r4,r4,8
+	mr	r8,r9
+	blt	cr1,72f
+21:	ld	r9,8(r4)
+70:	std	r8,8(r3)
+22:	ldu	r8,16(r4)
+71:	stdu	r9,16(r3)
+	bdnz	21b
+72:	std	r8,8(r3)
+	beq+	3f
+	addi	r3,r3,16
+23:	ld	r9,8(r4)
+.Ldo_tail:
+	bf	cr7*4+1,1f
+	rotldi	r9,r9,32
+73:	stw	r9,0(r3)
+	addi	r3,r3,4
+1:	bf	cr7*4+2,2f
+	rotldi	r9,r9,16
+74:	sth	r9,0(r3)
+	addi	r3,r3,2
+2:	bf	cr7*4+3,3f
+	rotldi	r9,r9,8
+75:	stb	r9,0(r3)
+3:	li	r3,0
+	blr
+
+.Lsrc_unaligned:
+	srdi	r6,r5,3
+	addi	r5,r5,-16
+	subf	r4,r0,r4
+	srdi	r7,r5,4
+	sldi	r10,r0,3
+	cmpldi	cr6,r6,3
+	andi.	r5,r5,7
+	mtctr	r7
+	subfic	r11,r10,64
+	add	r5,r5,r0
+	bt	cr7*4+0,28f
+
+24:	ld	r9,0(r4)	/* 3+2n loads, 2+2n stores */
+25:	ld	r0,8(r4)
+	sld	r6,r9,r10
+26:	ldu	r9,16(r4)
+	srd	r7,r0,r11
+	sld	r8,r0,r10
+	or	r7,r7,r6
+	blt	cr6,79f
+27:	ld	r0,8(r4)
+	b	2f
+
+28:	ld	r0,0(r4)	/* 4+2n loads, 3+2n stores */
+29:	ldu	r9,8(r4)
+	sld	r8,r0,r10
+	addi	r3,r3,-8
+	blt	cr6,5f
+30:	ld	r0,8(r4)
+	srd	r12,r9,r11
+	sld	r6,r9,r10
+31:	ldu	r9,16(r4)
+	or	r12,r8,r12
+	srd	r7,r0,r11
+	sld	r8,r0,r10
+	addi	r3,r3,16
+	beq	cr6,78f
+
+1:	or	r7,r7,r6
+32:	ld	r0,8(r4)
+76:	std	r12,8(r3)
+2:	srd	r12,r9,r11
+	sld	r6,r9,r10
+33:	ldu	r9,16(r4)
+	or	r12,r8,r12
+77:	stdu	r7,16(r3)
+	srd	r7,r0,r11
+	sld	r8,r0,r10
+	bdnz	1b
+
+78:	std	r12,8(r3)
+	or	r7,r7,r6
+79:	std	r7,16(r3)
+5:	srd	r12,r9,r11
+	or	r12,r8,r12
+80:	std	r12,24(r3)
+	bne	6f
+	li	r3,0
+	blr
+6:	cmpwi	cr1,r5,8
+	addi	r3,r3,32
+	sld	r9,r9,r10
+	blt	cr1,.Ldo_tail
+34:	ld	r0,8(r4)
+	srd	r7,r0,r11
+	or	r9,r7,r9
+	b	.Ldo_tail
+
+.Ldst_unaligned:
+	mtcrf	0x01,r6		/* put #bytes to 8B bdry into cr7 */
+	subf	r5,r6,r5
+	li	r7,0
+	cmpldi	r1,r5,16
+	bf	cr7*4+3,1f
+35:	lbz	r0,0(r4)
+81:	stb	r0,0(r3)
+	addi	r7,r7,1
+1:	bf	cr7*4+2,2f
+36:	lhzx	r0,r7,r4
+82:	sthx	r0,r7,r3
+	addi	r7,r7,2
+2:	bf	cr7*4+1,3f
+37:	lwzx	r0,r7,r4
+83:	stwx	r0,r7,r3
+3:	mtcrf	0x01,r5
+	add	r4,r6,r4
+	add	r3,r6,r3
+	b	.Ldst_aligned
+
+.Lshort_copy:
+	bf	cr7*4+0,1f
+38:	lwz	r0,0(r4)
+39:	lwz	r9,4(r4)
+	addi	r4,r4,8
+84:	stw	r0,0(r3)
+85:	stw	r9,4(r3)
+	addi	r3,r3,8
+1:	bf	cr7*4+1,2f
+40:	lwz	r0,0(r4)
+	addi	r4,r4,4
+86:	stw	r0,0(r3)
+	addi	r3,r3,4
+2:	bf	cr7*4+2,3f
+41:	lhz	r0,0(r4)
+	addi	r4,r4,2
+87:	sth	r0,0(r3)
+	addi	r3,r3,2
+3:	bf	cr7*4+3,4f
+42:	lbz	r0,0(r4)
+88:	stb	r0,0(r3)
+4:	li	r3,0
+	blr
+
+/*
+ * exception handlers follow
+ * we have to return the number of bytes not copied
+ * for an exception on a load, we set the rest of the destination to 0
+ */
+
+136:
+137:
+	add	r3,r3,r7
+	b	1f
+130:
+131:
+	addi	r3,r3,8
+120:
+122:
+124:
+125:
+126:
+127:
+128:
+129:
+133:
+	addi	r3,r3,8
+121:
+132:
+	addi	r3,r3,8
+123:
+134:
+135:
+138:
+139:
+140:
+141:
+142:
+
+/*
+ * here we have had a fault on a load and r3 points to the first
+ * unmodified byte of the destination
+ */
+1:	ld	r6,-24(r1)
+	ld	r4,-16(r1)
+	ld	r5,-8(r1)
+	subf	r6,r6,r3
+	add	r4,r4,r6
+	subf	r5,r6,r5	/* #bytes left to go */
+
+/*
+ * first see if we can copy any more bytes before hitting another exception
+ */
+	mtctr	r5
+43:	lbz	r0,0(r4)
+	addi	r4,r4,1
+89:	stb	r0,0(r3)
+	addi	r3,r3,1
+	bdnz	43b
+	li	r3,0		/* huh? all copied successfully this time? */
+	blr
+
+/*
+ * here we have trapped again, need to clear ctr bytes starting at r3
+ */
+143:	mfctr	r5
+	li	r0,0
+	mr	r4,r3
+	mr	r3,r5		/* return the number of bytes not copied */
+1:	andi.	r9,r4,7
+	beq	3f
+90:	stb	r0,0(r4)
+	addic.	r5,r5,-1
+	addi	r4,r4,1
+	bne	1b
+	blr
+3:	srdi	r9,r5,3
+	andi.	r5,r5,7
+	mtctr	r9
+91:	std	r0,0(r4)
+	addi	r4,r4,8
+	bdnz	91b
+	beqlr
+	mtctr	r5
+92:	stb	r0,0(r4)
+	addi	r4,r4,1
+	bdnz	92b
+	blr
+
+/*
+ * exception handlers for stores: we just need to work
+ * out how many bytes weren't copied
+ */
+182:
+183:
+	add	r3,r3,r7
+	b	1f
+180:
+	addi	r3,r3,8
+171:
+177:
+	addi	r3,r3,8
+170:
+172:
+176:
+178:
+	addi	r3,r3,4
+185:
+	addi	r3,r3,4
+173:
+174:
+175:
+179:
+181:
+184:
+186:
+187:
+188:
+189:	
+1:
+	ld	r6,-24(r1)
+	ld	r5,-8(r1)
+	add	r6,r6,r5
+	subf	r3,r3,r6	/* #bytes not copied */
+190:
+191:
+192:
+	blr			/* #bytes not copied in r3 */
+
+	.section __ex_table,"a"
+	.align	3
+	.llong	20b,120b
+	.llong	21b,121b
+	.llong	70b,170b
+	.llong	22b,122b
+	.llong	71b,171b
+	.llong	72b,172b
+	.llong	23b,123b
+	.llong	73b,173b
+	.llong	74b,174b
+	.llong	75b,175b
+	.llong	24b,124b
+	.llong	25b,125b
+	.llong	26b,126b
+	.llong	27b,127b
+	.llong	28b,128b
+	.llong	29b,129b
+	.llong	30b,130b
+	.llong	31b,131b
+	.llong	32b,132b
+	.llong	76b,176b
+	.llong	33b,133b
+	.llong	77b,177b
+	.llong	78b,178b
+	.llong	79b,179b
+	.llong	80b,180b
+	.llong	34b,134b
+	.llong	35b,135b
+	.llong	81b,181b
+	.llong	36b,136b
+	.llong	82b,182b
+	.llong	37b,137b
+	.llong	83b,183b
+	.llong	38b,138b
+	.llong	39b,139b
+	.llong	84b,184b
+	.llong	85b,185b
+	.llong	40b,140b
+	.llong	86b,186b
+	.llong	41b,141b
+	.llong	87b,187b
+	.llong	42b,142b
+	.llong	88b,188b
+	.llong	43b,143b
+	.llong	89b,189b
+	.llong	90b,190b
+	.llong	91b,191b
+	.llong	92b,192b
+	
+	.text
+
+/*
+ * Routine to copy a whole page of data, optimized for POWER4.
+ * On POWER4 it is more than 50% faster than the simple loop
+ * above (following the .Ldst_aligned label) but it runs slightly
+ * slower on POWER3.
+ */
+.Lcopy_page:
+	std	r31,-32(1)
+	std	r30,-40(1)
+	std	r29,-48(1)
+	std	r28,-56(1)
+	std	r27,-64(1)
+	std	r26,-72(1)
+	std	r25,-80(1)
+	std	r24,-88(1)
+	std	r23,-96(1)
+	std	r22,-104(1)
+	std	r21,-112(1)
+	std	r20,-120(1)
+	li	r5,4096/32 - 1
+	addi	r3,r3,-8
+	li	r0,5
+0:	addi	r5,r5,-24
+	mtctr	r0
+20:	ld	r22,640(4)
+21:	ld	r21,512(4)
+22:	ld	r20,384(4)
+23:	ld	r11,256(4)
+24:	ld	r9,128(4)
+25:	ld	r7,0(4)
+26:	ld	r25,648(4)
+27:	ld	r24,520(4)
+28:	ld	r23,392(4)
+29:	ld	r10,264(4)
+30:	ld	r8,136(4)
+31:	ldu	r6,8(4)
+	cmpwi	r5,24
+1:
+32:	std	r22,648(3)
+33:	std	r21,520(3)
+34:	std	r20,392(3)
+35:	std	r11,264(3)
+36:	std	r9,136(3)
+37:	std	r7,8(3)
+38:	ld	r28,648(4)
+39:	ld	r27,520(4)
+40:	ld	r26,392(4)
+41:	ld	r31,264(4)
+42:	ld	r30,136(4)
+43:	ld	r29,8(4)
+44:	std	r25,656(3)
+45:	std	r24,528(3)
+46:	std	r23,400(3)
+47:	std	r10,272(3)
+48:	std	r8,144(3)
+49:	std	r6,16(3)
+50:	ld	r22,656(4)
+51:	ld	r21,528(4)
+52:	ld	r20,400(4)
+53:	ld	r11,272(4)
+54:	ld	r9,144(4)
+55:	ld	r7,16(4)
+56:	std	r28,664(3)
+57:	std	r27,536(3)
+58:	std	r26,408(3)
+59:	std	r31,280(3)
+60:	std	r30,152(3)
+61:	stdu	r29,24(3)
+62:	ld	r25,664(4)
+63:	ld	r24,536(4)
+64:	ld	r23,408(4)
+65:	ld	r10,280(4)
+66:	ld	r8,152(4)
+67:	ldu	r6,24(4)
+	bdnz	1b
+68:	std	r22,648(3)
+69:	std	r21,520(3)
+70:	std	r20,392(3)
+71:	std	r11,264(3)
+72:	std	r9,136(3)
+73:	std	r7,8(3)
+74:	addi	r4,r4,640
+75:	addi	r3,r3,648
+	bge	0b
+	mtctr	r5
+76:	ld	r7,0(4)
+77:	ld	r8,8(4)
+78:	ldu	r9,16(4)
+3:
+79:	ld	r10,8(4)
+80:	std	r7,8(3)
+81:	ld	r7,16(4)
+82:	std	r8,16(3)
+83:	ld	r8,24(4)
+84:	std	r9,24(3)
+85:	ldu	r9,32(4)
+86:	stdu	r10,32(3)
+	bdnz	3b
+4:
+87:	ld	r10,8(4)
+88:	std	r7,8(3)
+89:	std	r8,16(3)
+90:	std	r9,24(3)
+91:	std	r10,32(3)
+9:	ld	r20,-120(1)
+	ld	r21,-112(1)
+	ld	r22,-104(1)
+	ld	r23,-96(1)
+	ld	r24,-88(1)
+	ld	r25,-80(1)
+	ld	r26,-72(1)
+	ld	r27,-64(1)
+	ld	r28,-56(1)
+	ld	r29,-48(1)
+	ld	r30,-40(1)
+	ld	r31,-32(1)
+	li	r3,0
+	blr
+
+/*
+ * on an exception, reset to the beginning and jump back into the
+ * standard __copy_tofrom_user
+ */
+100:	ld	r3,-24(r1)
+	ld	r4,-24(r1)
+	li	r5,4096
+	b	.Ldst_aligned
+
+	.section __ex_table,"a"
+	.align	3
+	.llong	20b,100b
+	.llong	21b,100b
+	.llong	22b,100b
+	.llong	23b,100b
+	.llong	24b,100b
+	.llong	25b,100b
+	.llong	26b,100b
+	.llong	27b,100b
+	.llong	28b,100b
+	.llong	29b,100b
+	.llong	30b,100b
+	.llong	31b,100b
+	.llong	32b,100b
+	.llong	33b,100b
+	.llong	34b,100b
+	.llong	35b,100b
+	.llong	36b,100b
+	.llong	37b,100b
+	.llong	38b,100b
+	.llong	39b,100b
+	.llong	40b,100b
+	.llong	41b,100b
+	.llong	42b,100b
+	.llong	43b,100b
+	.llong	44b,100b
+	.llong	45b,100b
+	.llong	46b,100b
+	.llong	47b,100b
+	.llong	48b,100b
+	.llong	49b,100b
+	.llong	50b,100b
+	.llong	51b,100b
+	.llong	52b,100b
+	.llong	53b,100b
+	.llong	54b,100b
+	.llong	55b,100b
+	.llong	56b,100b
+	.llong	57b,100b
+	.llong	58b,100b
+	.llong	59b,100b
+	.llong	60b,100b
+	.llong	61b,100b
+	.llong	62b,100b
+	.llong	63b,100b
+	.llong	64b,100b
+	.llong	65b,100b
+	.llong	66b,100b
+	.llong	67b,100b
+	.llong	68b,100b
+	.llong	69b,100b
+	.llong	70b,100b
+	.llong	71b,100b
+	.llong	72b,100b
+	.llong	73b,100b
+	.llong	74b,100b
+	.llong	75b,100b
+	.llong	76b,100b
+	.llong	77b,100b
+	.llong	78b,100b
+	.llong	79b,100b
+	.llong	80b,100b
+	.llong	81b,100b
+	.llong	82b,100b
+	.llong	83b,100b
+	.llong	84b,100b
+	.llong	85b,100b
+	.llong	86b,100b
+	.llong	87b,100b
+	.llong	88b,100b
+	.llong	89b,100b
+	.llong	90b,100b
+	.llong	91b,100b

arch/ppc64/lib/memcpy.S

+/*
+ * arch/ppc64/lib/memcpy.S
+ *
+ * Copyright (C) 2002 Paul Mackerras, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <asm/processor.h>
+#include "../kernel/ppc_asm.h"
+
+	.align	7
+_GLOBAL(memcpy)
+	mtcrf	0x01,r5
+	cmpldi	cr1,r5,16
+	neg	r6,r3		# LS 3 bits = # bytes to 8-byte dest bdry
+	andi.	r6,r6,7
+	dcbt	0,r4
+	blt	cr1,.Lshort_copy
+	bne	.Ldst_unaligned
+.Ldst_aligned:
+	andi.	r0,r4,7
+	addi	r3,r3,-16
+	bne	.Lsrc_unaligned
+	srdi	r7,r5,4
+	ld	r9,0(r4)
+	addi	r4,r4,-8
+	mtctr	r7
+	andi.	r5,r5,7
+	bf	cr7*4+0,2f
+	addi	r3,r3,8
+	addi	r4,r4,8
+	mr	r8,r9
+	blt	cr1,3f
+1:	ld	r9,8(r4)
+	std	r8,8(r3)
+2:	ldu	r8,16(r4)
+	stdu	r9,16(r3)
+	bdnz	1b
+3:	std	r8,8(r3)
+	beqlr
+	addi	r3,r3,16
+	ld	r9,8(r4)
+.Ldo_tail:
+	bf	cr7*4+1,1f
+	rotldi	r9,r9,32
+	stw	r9,0(r3)
+	addi	r3,r3,4
+1:	bf	cr7*4+2,2f
+	rotldi	r9,r9,16
+	sth	r9,0(r3)
+	addi	r3,r3,2
+2:	bf	cr7*4+3,3f
+	rotldi	r9,r9,8
+	stb	r9,0(r3)
+3:	blr
+
+.Lsrc_unaligned:
+	srdi	r6,r5,3
+	addi	r5,r5,-16
+	subf	r4,r0,r4
+	srdi	r7,r5,4
+	sldi	r10,r0,3
+	cmpdi	cr6,r6,3
+	andi.	r5,r5,7
+	mtctr	r7
+	subfic	r11,r10,64
+	add	r5,r5,r0
+
+	bt	cr7*4+0,0f
+
+	ld	r9,0(r4)	# 3+2n loads, 2+2n stores
+	ld	r0,8(r4)
+	sld	r6,r9,r10
+	ldu	r9,16(r4)
+	srd	r7,r0,r11
+	sld	r8,r0,r10
+	or	r7,r7,r6
+	blt	cr6,4f
+	ld	r0,8(r4)
+	# s1<< in r8, d0=(s0<<|s1>>) in r7, s3 in r0, s2 in r9, nix in r6 & r12
+	b	2f
+
+0:	ld	r0,0(r4)	# 4+2n loads, 3+2n stores
+	ldu	r9,8(r4)
+	sld	r8,r0,r10
+	addi	r3,r3,-8
+	blt	cr6,5f
+	ld	r0,8(r4)
+	srd	r12,r9,r11
+	sld	r6,r9,r10
+	ldu	r9,16(r4)
+	or	r12,r8,r12
+	srd	r7,r0,r11
+	sld	r8,r0,r10
+	addi	r3,r3,16
+	beq	cr6,3f
+
+	# d0=(s0<<|s1>>) in r12, s1<< in r6, s2>> in r7, s2<< in r8, s3 in r9
+1:	or	r7,r7,r6
+	ld	r0,8(r4)
+	std	r12,8(r3)
+2:	srd	r12,r9,r11
+	sld	r6,r9,r10
+	ldu	r9,16(r4)
+	or	r12,r8,r12
+	stdu	r7,16(r3)
+	srd	r7,r0,r11
+	sld	r8,r0,r10
+	bdnz	1b
+
+3:	std	r12,8(r3)
+	or	r7,r7,r6
+4:	std	r7,16(r3)
+5:	srd	r12,r9,r11
+	or	r12,r8,r12
+	std	r12,24(r3)
+	beqlr
+	cmpwi	cr1,r5,8
+	addi	r3,r3,32
+	sld	r9,r9,r10
+	blt	cr1,.Ldo_tail
+	ld	r0,8(r4)
+	srd	r7,r0,r11
+	or	r9,r7,r9
+	b	.Ldo_tail
+
+.Ldst_unaligned:
+	mtcrf	0x01,r6		# put #bytes to 8B bdry into cr7
+	subf	r5,r6,r5
+	li	r7,0
+	cmpldi	r1,r5,16
+	bf	cr7*4+3,1f
+	lbz	r0,0(r4)
+	stb	r0,0(r3)
+	addi	r7,r7,1
+1:	bf	cr7*4+2,2f
+	lhzx	r0,r7,r4
+	sthx	r0,r7,r3
+	addi	r7,r7,2
+2:	bf	cr7*4+1,3f
+	lwzx	r0,r7,r4
+	stwx	r0,r7,r3
+3:	mtcrf	0x01,r5
+	add	r4,r6,r4
+	add	r3,r6,r3
+	b	.Ldst_aligned
+
+.Lshort_copy:
+	bf	cr7*4+0,1f
+	lwz	r0,0(r4)
+	lwz	r9,4(r4)
+	addi	r4,r4,8
+	stw	r0,0(r3)
+	stw	r9,4(r3)
+	addi	r3,r3,8
+1:	bf	cr7*4+1,2f
+	lwz	r0,0(r4)
+	addi	r4,r4,4
+	stw	r0,0(r3)
+	addi	r3,r3,4
+2:	bf	cr7*4+2,3f
+	lhz	r0,0(r4)
+	addi	r4,r4,2
+	sth	r0,0(r3)
+	addi	r3,r3,2
+3:	bf	cr7*4+3,4f
+	lbz	r0,0(r4)
+	stb	r0,0(r3)
+4:	blr

arch/ppc64/lib/string.S

@@ -12,61 +12,6 @@
 #include <asm/processor.h>
 #include <asm/errno.h>
 
-#define CACHE_LINE_SIZE		128
-#define LG_CACHE_LINE_SIZE	7
-#define MAX_COPY_PREFETCH	1
-
-#define COPY_16_BYTES		\
-	lwz	r7,4(r4);	\
-	lwz	r8,8(r4);	\
-	lwz	r9,12(r4);	\
-	lwzu	r10,16(r4);	\
-	stw	r7,4(r6);	\
-	stw	r8,8(r6);	\
-	stw	r9,12(r6);	\
-	stwu	r10,16(r6)
-
-#define COPY_16_BYTES_WITHEX(n)	\
-8 ## n ## 0:			\
-	lwz	r7,4(r4);	\
-8 ## n ## 1:			\
-	lwz	r8,8(r4);	\
-8 ## n ## 2:			\
-	lwz	r9,12(r4);	\
-8 ## n ## 3:			\
-	lwzu	r10,16(r4);	\
-8 ## n ## 4:			\
-	stw	r7,4(r6);	\
-8 ## n ## 5:			\
-	stw	r8,8(r6);	\
-8 ## n ## 6:			\
-	stw	r9,12(r6);	\
-8 ## n ## 7:			\
-	stwu	r10,16(r6)
-
-#define COPY_16_BYTES_EXCODE(n)			\
-9 ## n ## 0:					\
-	addi	r5,r5,-(16 * n);		\
-	b	104f;				\
-9 ## n ## 1:					\
-	addi	r5,r5,-(16 * n);		\
-	b	105f;				\
-.section __ex_table,"a";			\
-	.align	3;				\
-	.llong	8 ## n ## 0b,9 ## n ## 0b;	\
-	.llong	8 ## n ## 1b,9 ## n ## 0b;	\
-	.llong	8 ## n ## 2b,9 ## n ## 0b;	\
-	.llong	8 ## n ## 3b,9 ## n ## 0b;	\
-	.llong	8 ## n ## 4b,9 ## n ## 1b;	\
-	.llong	8 ## n ## 5b,9 ## n ## 1b;	\
-	.llong	8 ## n ## 6b,9 ## n ## 1b;	\
-	.llong	8 ## n ## 7b,9 ## n ## 1b;	\
-.text
-
-CACHELINE_BYTES = CACHE_LINE_SIZE
-LG_CACHELINE_BYTES = LG_CACHE_LINE_SIZE
-CACHELINE_MASK = (CACHE_LINE_SIZE-1)
-
 _GLOBAL(strcpy)
 	addi	r5,r3,-1
 	addi	r4,r4,-1
@@ -148,48 +93,7 @@ _GLOBAL(memset)
 _GLOBAL(memmove)
 	cmplw	0,r3,r4
 	bgt	.backwards_memcpy
-	/* fall through */
-
-_GLOBAL(memcpy)
-	srwi.	r7,r5,3
-	addi	r6,r3,-4
-	addi	r4,r4,-4
-	beq	2f			/* if less than 8 bytes to do */
-	andi.	r0,r6,3			/* get dest word aligned */
-	mtctr	r7
-	bne	5f
-1:	lwz	r7,4(r4)
-	lwzu	r8,8(r4)
-	stw	r7,4(r6)
-	stwu	r8,8(r6)
-	bdnz	1b
-	andi.	r5,r5,7
-2:	cmplwi	0,r5,4
-	blt	3f
-	lwzu	r0,4(r4)
-	addi	r5,r5,-4
-	stwu	r0,4(r6)
-3:	cmpwi	0,r5,0
-	beqlr
-	mtctr	r5
-	addi	r4,r4,3
-	addi	r6,r6,3
-4:	lbzu	r0,1(r4)
-	stbu	r0,1(r6)
-	bdnz	4b
-	blr
-5:	subfic	r0,r0,4
-	mtctr	r0
-6:	lbz	r7,4(r4)
-	addi	r4,r4,1
-	stb	r7,4(r6)
-	addi	r6,r6,1
-	bdnz	6b
-	subf	r5,r0,r5
-	rlwinm.	r7,r5,32-3,3,31
-	beq	2b
-	mtctr	r7
-	b	1b
+	b	.memcpy
 
 _GLOBAL(backwards_memcpy)
 	rlwinm.	r7,r5,32-3,3,31		/* r0 = r5 >> 3 */
@@ -253,195 +157,6 @@ _GLOBAL(memchr)
 2:	li	r3,0
 	blr
 
-_GLOBAL(__copy_tofrom_user)
-	addi	r4,r4,-4
-	addi	r6,r3,-4
-	neg	r0,r3
-	andi.	r0,r0,CACHELINE_MASK	/* # bytes to start of cache line */
-	beq	58f
-
-	cmplw	0,r5,r0			/* is this more than total to do? */
-	blt	63f			/* if not much to do */
-	andi.	r8,r0,3			/* get it word-aligned first */
-	mtctr	r8
-	beq+	61f
-70:	lbz	r9,4(r4)		/* do some bytes */
-71:	stb	r9,4(r6)
-	addi	r4,r4,1
-	addi	r6,r6,1
-	bdnz	70b
-61:	subf	r5,r0,r5
-	srwi.	r0,r0,2
-	mtctr	r0
-	beq	58f
-72:	lwzu	r9,4(r4)		/* do some words */
-73:	stwu	r9,4(r6)
-	bdnz	72b
-
-58:	srwi.	r0,r5,LG_CACHELINE_BYTES /* # complete cachelines */
-	clrlwi	r5,r5,32-LG_CACHELINE_BYTES
-	li	r11,4
-	beq	63f
-
-	/* Here we decide how far ahead to prefetch the source */
-#if MAX_COPY_PREFETCH > 1
-	/* Heuristically, for large transfers we prefetch
-	   MAX_COPY_PREFETCH cachelines ahead.  For small transfers
-	   we prefetch 1 cacheline ahead. */
-	cmpwi	r0,MAX_COPY_PREFETCH
-	li	r7,1
-	li	r3,4
-	ble	111f
-	li	r7,MAX_COPY_PREFETCH
-111:	mtctr	r7
-112:	dcbt	r3,r4
-	addi	r3,r3,CACHELINE_BYTES
-	bdnz	112b
-#else /* MAX_COPY_PREFETCH == 1 */
-	li	r3,CACHELINE_BYTES + 4
-	dcbt	r11,r4
-#endif /* MAX_COPY_PREFETCH */
-
-	mtctr	r0
-53:
-	dcbt	r3,r4
-	dcbz	r11,r6
-/* had to move these to keep extable in order */
-	.section __ex_table,"a"
-	.align	3
-	.llong	70b,100f
-	.llong	71b,101f
-	.llong	72b,102f
-	.llong	73b,103f
-	.llong	53b,105f
-	.text
-/* the main body of the cacheline loop */
-	COPY_16_BYTES_WITHEX(0)
-#if CACHE_LINE_SIZE >= 32
-	COPY_16_BYTES_WITHEX(1)
-#if CACHE_LINE_SIZE >= 64
-	COPY_16_BYTES_WITHEX(2)
-	COPY_16_BYTES_WITHEX(3)
-#if CACHE_LINE_SIZE >= 128
-	COPY_16_BYTES_WITHEX(4)
-	COPY_16_BYTES_WITHEX(5)
-	COPY_16_BYTES_WITHEX(6)
-	COPY_16_BYTES_WITHEX(7)
-#endif
-#endif
-#endif
-	bdnz	53b
-
-63:	srwi.	r0,r5,2
-	mtctr	r0
-	beq	64f
-30:	lwzu	r0,4(r4)
-31:	stwu	r0,4(r6)
-	bdnz	30b
-
-64:	andi.	r0,r5,3
-	mtctr	r0
-	beq+	65f
-40:	lbz	r0,4(r4)
-41:	stb	r0,4(r6)
-	addi	r4,r4,1
-	addi	r6,r6,1
-	bdnz	40b
-65:	li	r3,0
-	blr
-
-/* read fault, initial single-byte copy */
-100:	li	r4,0
-	b	90f
-/* write fault, initial single-byte copy */
-101:	li	r4,1
-90:	subf	r5,r8,r5
-	li	r3,0
-	b	99f
-/* read fault, initial word copy */
-102:	li	r4,0
-	b	91f
-/* write fault, initial word copy */
-103:	li	r4,1
-91:	li	r3,2
-	b	99f
-
-/*
- * this stuff handles faults in the cacheline loop and branches to either
- * 104f (if in read part) or 105f (if in write part), after updating r5
- */
-	COPY_16_BYTES_EXCODE(0)
-#if CACHE_LINE_SIZE >= 32
-	COPY_16_BYTES_EXCODE(1)
-#if CACHE_LINE_SIZE >= 64
-	COPY_16_BYTES_EXCODE(2)
-	COPY_16_BYTES_EXCODE(3)
-#if CACHE_LINE_SIZE >= 128
-	COPY_16_BYTES_EXCODE(4)
-	COPY_16_BYTES_EXCODE(5)
-	COPY_16_BYTES_EXCODE(6)
-	COPY_16_BYTES_EXCODE(7)
-#endif
-#endif
-#endif
-
-/* read fault in cacheline loop */
-104:	li	r4,0
-	b	92f
-/* fault on dcbz (effectively a write fault) */
-/* or write fault in cacheline loop */
-105:	li	r4,1
-92:	li	r3,LG_CACHELINE_BYTES
-	b	99f
-/* read fault in final word loop */
-108:	li	r4,0
-	b	93f
-/* write fault in final word loop */
-109:	li	r4,1
-93:	andi.	r5,r5,3
-	li	r3,2
-	b	99f
-/* read fault in final byte loop */
-110:	li	r4,0
-	b	94f
-/* write fault in final byte loop */
-111:	li	r4,1
-94:	li	r5,0
-	li	r3,0
-/*
- * At this stage the number of bytes not copied is
- * r5 + (ctr << r3), and r4 is 0 for read or 1 for write.
- */
-99:	mfctr	r0
-	slw	r3,r0,r3
-	add	r3,r3,r5
-	cmpwi	0,r4,0
-	bne	120f
-/* for read fault, clear out the destination: r3 bytes starting at 4(r6) */
-	srwi.	r0,r3,2
-	li	r9,0
-	mtctr	r0
-	beq	113f
-112:	stwu	r9,4(r6)
-	bdnz	112b
-113:	andi.	r0,r3,3
-	mtctr	r0
-	beq	120f
-114:	stb	r9,4(r6)
-	addi	r6,r6,1
-	bdnz	114b
-120:	blr
-
-	.section __ex_table,"a"
-	.align	3
-	.llong	30b,108b
-	.llong	31b,109b
-	.llong	40b,110b
-	.llong	41b,111b
-	.llong	112b,120b
-	.llong	114b,120b
-	.text
-
 _GLOBAL(__clear_user)
 	addi	r6,r3,-4
 	li	r3,0

arch/ppc64/mm/fault.c

@@ -38,16 +38,11 @@
 
 #include <asm/ppcdebug.h>
 
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
-extern void (*debugger)(struct pt_regs *);
-extern void (*debugger_fault_handler)(struct pt_regs *);
-extern int (*debugger_dabr_match)(struct pt_regs *);
+#ifdef CONFIG_DEBUG_KERNEL
 int debugger_kernel_faults = 1;
 #endif
 
-extern void die_if_kernel(char *, struct pt_regs *, long);
-void bad_page_fault(struct pt_regs *, unsigned long);
-void do_page_fault(struct pt_regs *, unsigned long, unsigned long);
+void bad_page_fault(struct pt_regs *, unsigned long, int);
 
 /*
  * For 600- and 800-family processors, the error_code parameter is DSISR
@@ -71,7 +66,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
 	if (regs->trap == 0x400)
 		error_code &= 0x48200000;
 
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
+#ifdef CONFIG_DEBUG_KERNEL
 	if (debugger_fault_handler && (regs->trap == 0x300 ||
 				regs->trap == 0x380)) {
 		debugger_fault_handler(regs);
@@ -83,10 +78,10 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
 		if (debugger_dabr_match(regs))
 			return;
 	}
-#endif /* CONFIG_XMON || CONFIG_KGDB */
+#endif
 
 	if (in_interrupt() || mm == NULL) {
-		bad_page_fault(regs, address);
+		bad_page_fault(regs, address, SIGSEGV);
 		return;
 	}
 	down_read(&mm->mmap_sem);
@@ -159,7 +154,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
 		return;
 	}
 
-	bad_page_fault(regs, address);
+	bad_page_fault(regs, address, SIGSEGV);
 	return;
 
 /*
@@ -176,7 +171,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
 	printk("VM: killing process %s\n", current->comm);
 	if (user_mode(regs))
 		do_exit(SIGKILL);
-	bad_page_fault(regs, address);
+	bad_page_fault(regs, address, SIGKILL);
 	return;
 
 do_sigbus:
@@ -187,7 +182,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
 	info.si_addr = (void *)address;
 	force_sig_info (SIGBUS, &info, current);
 	if (!user_mode(regs))
-		bad_page_fault(regs, address);
+		bad_page_fault(regs, address, SIGBUS);
 }
 
 /*
@@ -196,8 +191,10 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
  * in traps.c.
  */
 void
-bad_page_fault(struct pt_regs *regs, unsigned long address)
+bad_page_fault(struct pt_regs *regs, unsigned long address, int sig)
 {
+	extern void die(const char *, struct pt_regs *, long);
+
 	unsigned long fixup;
 
 	/* Are we prepared to handle this fault?  */
@@ -207,13 +204,9 @@ bad_page_fault(struct pt_regs *regs, unsigned long address)
 	}
 
 	/* kernel has accessed a bad area */
-	show_regs(regs);
-#if defined(CONFIG_XMON) || defined(CONFIG_KGDB)
+#ifdef CONFIG_DEBUG_KERNEL
 	if (debugger_kernel_faults)
 		debugger(regs);
 #endif
-	print_backtrace( (unsigned long *)regs->gpr[1] );
-	panic("kernel access of bad area pc %lx lr %lx address %lX tsk %s/%d",
-	      regs->nip,regs->link,address,current->comm,current->pid);
+	die("Kernel access of bad area", regs, sig);
 }
-

arch/ppc64/mm/init.c

@@ -85,14 +85,10 @@ extern struct task_struct *current_set[NR_CPUS];
 
 void mm_init_ppc64(void);
 
-unsigned long *pmac_find_end_of_memory(void);
-extern unsigned long *find_end_of_memory(void);
-
 extern pgd_t ioremap_dir[];
 pgd_t * ioremap_pgd = (pgd_t *)&ioremap_dir;
 
 static void map_io_page(unsigned long va, unsigned long pa, int flags);
-extern void die_if_kernel(char *,struct pt_regs *,long);
 
 unsigned long klimit = (unsigned long)_end;
 
@@ -246,20 +242,17 @@ static void map_io_page(unsigned long ea, unsigned long pa, int flags)
 void
 flush_tlb_mm(struct mm_struct *mm)
 {
-	if (mm->map_count) {
 	struct vm_area_struct *mp;
+
+	spin_lock(&mm->page_table_lock);
+
 	for (mp = mm->mmap; mp != NULL; mp = mp->vm_next)
 		__flush_tlb_range(mm, mp->vm_start, mp->vm_end);
-	} else {
-		/* MIKEC: It is not clear why this is needed */
-		/* paulus: it is needed to clear out stale HPTEs
-		 * when an address space (represented by an mm_struct)
-		 * is being destroyed. */
-		__flush_tlb_range(mm, USER_START, USER_END);
-	}
 
 	/* XXX are there races with checking cpu_vm_mask? - Anton */
 	mm->cpu_vm_mask = 0;
+
+	spin_unlock(&mm->page_table_lock);
 }
 
 /*
@@ -434,12 +427,11 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 }
 #endif
 
-
-
 /*
  * Do very early mm setup.
  */
-void __init mm_init_ppc64(void) {
+void __init mm_init_ppc64(void)
+{
 	struct paca_struct *lpaca;
 	unsigned long guard_page, index;
 
@@ -467,8 +459,6 @@ void __init mm_init_ppc64(void) {
 	ppc_md.progress("MM:exit", 0x211);
 }
 
-
-
 /*
  * Initialize the bootmem system and give it all the memory we
  * have available.
@@ -665,6 +655,8 @@ int __hash_page(unsigned long ea, unsigned long access, unsigned long vsid,
  * fault has been handled by updating a PTE in the linux page tables.
  * We use it to preload an HPTE into the hash table corresponding to
  * the updated linux PTE.
+ * 
+ * This must always be called with the mm->page_table_lock held
  */
 void update_mmu_cache(struct vm_area_struct *vma, unsigned long ea,
 		      pte_t pte)

arch/ppc64/xmon/xmon.c

@@ -309,7 +309,6 @@ xmon(struct pt_regs *excp)
 			std	29,232(%0)\n\
 			std	30,240(%0)\n\
 			std	31,248(%0)" : : "b" (&regs));
-		printf("xmon called\n");
 		/* Fetch the link reg for this stack frame.
 		 NOTE: the prev printf fills in the lr. */
 		regs.nip = regs.link = ((unsigned long *)(regs.gpr[1]))[2];

include/asm-ppc64/fcntl.h

@@ -26,7 +26,7 @@
 #define O_DIRECTORY      040000	/* must be a directory */
 #define O_NOFOLLOW      0100000	/* don't follow links */
 #define O_LARGEFILE     0200000
-#define O_DIRECT	0400000	/* direct disk access hint - currently ignored */
+#define O_DIRECT	0400000	/* direct disk access hint */
 
 #define F_DUPFD		0	/* dup */
 #define F_GETFD		1	/* get close_on_exec */

include/asm-ppc64/hardirq.h

@@ -26,7 +26,7 @@ typedef struct {
 #endif
 	unsigned int __local_bh_count;
 	unsigned int __syscall_count;
-	unsigned long __unused;
+	unsigned long idle_timestamp;
 	struct task_struct * __ksoftirqd_task;
 } ____cacheline_aligned irq_cpustat_t;
 

include/asm-ppc64/memory.h

@@ -50,7 +50,7 @@ static inline void isync(void)
 
 #define HMT_LOW		"\tor	1,1,1		# low priority\n"
 #define HMT_MEDIUM	"\tor	2,2,2		# medium priority\n"
-#define HMT_MEDIUM	"\tor	3,3,3		# high priority\n"
+#define HMT_HIGH	"\tor	3,3,3		# high priority\n"
 #else
 #define HMT_low()	do { } while(0)
 #define HMT_medium()	do { } while(0)

include/asm-ppc64/mmu.h

@@ -211,18 +211,11 @@ static inline void _tlbie(unsigned long va, int large)
 	asm volatile("eieio; tlbsync; ptesync": : :"memory");
 }
 
-static inline void _tlbiel(unsigned long va, int large)
+static inline void _tlbiel(unsigned long va)
 {
 	asm volatile("ptesync": : :"memory");
-
-	if (large) {
-		asm volatile("clrldi	%0,%0,16\n\
-			      tlbiel	%0,1" : : "r"(va) : "memory");
-	} else {
 	asm volatile("clrldi	%0,%0,16\n\
-			      tlbiel	%0,0" : : "r"(va) : "memory");
-	}
-
+		      tlbiel	%0" : : "r"(va) : "memory");
 	asm volatile("ptesync": : :"memory");
 }
 

include/asm-ppc64/paca.h

@@ -85,7 +85,8 @@ struct paca_struct {
 	u8 xProcEnabled;		/* 1=soft enabled			0x78 */
 	u8 xHrdIntCount;		/* Count of active hardware interrupts  0x79  */
 	u8 prof_enabled;		/* 1=iSeries profiling enabled          0x7A */
-	u8 resv1[5];			/*					0x7B-0x7F */
+	u8 stab_cache_pointer;	
+	u8 resv1[4];			/*					0x7B-0x7F */
 
 /*=====================================================================================
  * CACHE_LINE_2 0x0080 - 0x00FF

include/asm-ppc64/page.h

@@ -215,9 +215,12 @@ static inline int get_order(unsigned long size)
 #define __a2p(x) ((void *) absolute_to_phys(x))
 #define __a2v(x) ((void *) __va(absolute_to_phys(x)))
 
-#define virt_to_page(kaddr) (mem_map+(__pa((unsigned long)kaddr) >> PAGE_SHIFT))
+#define pfn_to_page(pfn)	(mem_map + (pfn))
+#define page_to_pfn(pfn)	((unsigned long)((pfn) - mem_map))
+#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
 
-#define VALID_PAGE(page)    ((page - mem_map) < max_mapnr)
+#define pfn_valid(pfn)		((pfn) < max_mapnr)
+#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
 
 #define MAP_NR(addr)		(__pa(addr) >> PAGE_SHIFT)
 

include/asm-ppc64/pgtable.h

@@ -167,22 +167,14 @@ extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
  * Conversion functions: convert a page and protection to a page entry,
  * and a page entry and page directory to the page they refer to.
  *
- * mk_pte_phys takes a physical address as input 
- *
  * mk_pte takes a (struct page *) as input
  */
+#define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
 
-#define mk_pte_phys(physpage,pgprot)                                      \
-({									  \
-	pte_t pte;							  \
-	pte_val(pte) = (((physpage)<<(PTE_SHIFT-PAGE_SHIFT)) | pgprot_val(pgprot)); \
-	pte;							          \
-})
-
-#define mk_pte(page,pgprot)                                               \
+#define pfn_pte(pfn,pgprot)						\
 ({									\
 	pte_t pte;							\
-	pte_val(pte) = ((unsigned long)((page) - mem_map) << PTE_SHIFT) |   \
+	pte_val(pte) = ((unsigned long)(pfn) << PTE_SHIFT) |   		\
                         pgprot_val(pgprot);				\
 	pte;								\
 })
@@ -195,8 +187,8 @@ extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
 
 /* pte_clear moved to later in this file */
 
-#define pte_pagenr(x)		((unsigned long)((pte_val(x) >> PTE_SHIFT)))
-#define pte_page(x)		(mem_map+pte_pagenr(x))
+#define pte_pfn(x)		((unsigned long)((pte_val(x) >> PTE_SHIFT)))
+#define pte_page(x)		pfn_to_page(pte_pfn(x))
 
 #define pmd_set(pmdp, ptep) 	(pmd_val(*(pmdp)) = (__ba_to_bpn(ptep)))
 #define pmd_none(pmd)		(!pmd_val(pmd))

include/asm-ppc64/processor.h

@@ -97,7 +97,7 @@
 #define FPSCR_VX	0x20000000	/* Invalid operation summary */
 #define FPSCR_OX	0x10000000	/* Overflow exception summary */
 #define FPSCR_UX	0x08000000	/* Underflow exception summary */
-#define FPSCR_ZX	0x04000000	/* Zero-devide exception summary */
+#define FPSCR_ZX	0x04000000	/* Zero-divide exception summary */
 #define FPSCR_XX	0x02000000	/* Inexact exception summary */
 #define FPSCR_VXSNAN	0x01000000	/* Invalid op for SNaN */
 #define FPSCR_VXISI	0x00800000	/* Invalid op for Inv - Inv */
@@ -651,9 +651,7 @@ struct thread_struct {
 	unsigned long	ksp;		/* Kernel stack pointer */
 	struct pt_regs	*regs;		/* Pointer to saved register state */
 	mm_segment_t	fs;		/* for get_fs() validation */
-	signed long     last_syscall;
 	double		fpr[32];	/* Complete floating point set */
-	unsigned long	fpscr_pad;	/* fpr ... fpscr must be contiguous */
 	unsigned long	fpscr;		/* Floating point status */
 };
 
@@ -663,8 +661,8 @@ struct thread_struct {
 	INIT_SP, /* ksp */ \
 	(struct pt_regs *)INIT_SP - 1, /* regs */ \
 	KERNEL_DS, /*fs*/ \
-	0, /* last_syscall */ \
-	{0}, 0, 0 \
+	{0}, /* fpr */ \
+	0 /* fpscr */ \
 }
 
 /*

include/asm-ppc64/system.h

@@ -55,9 +55,31 @@
 #define smp_wmb()	__asm__ __volatile__("": : :"memory")
 #endif /* CONFIG_SMP */
 
+#ifdef CONFIG_DEBUG_KERNEL
+extern void (*debugger)(struct pt_regs *regs);
+extern int (*debugger_bpt)(struct pt_regs *regs);
+extern int (*debugger_sstep)(struct pt_regs *regs);
+extern int (*debugger_iabr_match)(struct pt_regs *regs);
+extern int (*debugger_dabr_match)(struct pt_regs *regs);
+extern void (*debugger_fault_handler)(struct pt_regs *regs);
+#else
+#define debugger(regs)			do { } while (0)
+#define debugger_bpt(regs)		0
+#define debugger_sstep(regs)		0
+#define debugger_iabr_match(regs)	0
+#define debugger_dabr_match(regs)	0
+#define debugger_fault_handler		((void (*)(struct pt_regs *))0)
+#endif
+
 #ifdef CONFIG_XMON
 extern void xmon_irq(int, void *, struct pt_regs *);
-extern void xmon(struct pt_regs *excp);
+
+extern void xmon(struct pt_regs *regs);
+extern int xmon_bpt(struct pt_regs *regs);
+extern int xmon_sstep(struct pt_regs *regs);
+extern int xmon_iabr_match(struct pt_regs *regs);
+extern int xmon_dabr_match(struct pt_regs *regs);
+extern void (*xmon_fault_handler)(struct pt_regs *regs);
 #endif
 
 extern void print_backtrace(unsigned long *);

include/asm-ppc64/thread_info.h

@@ -65,6 +65,8 @@ static inline struct thread_info *current_thread_info(void)
 
 #endif /* __ASSEMBLY__ */
 
+#define PREEMPT_ACTIVE		0x4000000
+
 /*
  * thread information flag bit numbers
  */

include/asm-ppc64/timex.h

@@ -18,7 +18,6 @@
 		<< (SHIFT_SCALE-SHIFT_HZ)) / HZ)
 
 typedef unsigned long cycles_t;
-extern cycles_t cacheflush_time;
 
 static inline cycles_t get_cycles(void)
 {

include/asm-ppc64/unistd.h

@@ -230,6 +230,9 @@
 #define __NR_lremovexattr	219
 #define __NR_fremovexattr	220
 #define __NR_futex		221
+#define __NR_tux		222
+#define __NR_sched_setaffinity	223     
+#define __NR_sched_getaffinity	224
 
 #if 0
 /* Remind paulus to add these into ppc32 */