Merge http://lia64.bkbits.net/to-linus-2.5

into ppc970.osdl.org:/home/torvalds/v2.6/linux

arch/ia64/defconfig

@@ -16,8 +16,10 @@ CONFIG_BROKEN_ON_SMP=y
 #
 CONFIG_SWAP=y
 CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
 CONFIG_BSD_PROCESS_ACCT=y
 CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
 CONFIG_LOG_BUF_SHIFT=16
 CONFIG_HOTPLUG=y
 CONFIG_IKCONFIG=y
@@ -29,6 +31,7 @@ CONFIG_EPOLL=y
 CONFIG_IOSCHED_NOOP=y
 CONFIG_IOSCHED_AS=y
 CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
 # CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
 
 #
@@ -69,7 +72,6 @@ CONFIG_VIRTUAL_MEM_MAP=y
 # CONFIG_IA64_CYCLONE is not set
 CONFIG_IOSAPIC=y
 CONFIG_FORCE_MAX_ZONEORDER=18
-# CONFIG_IA64_PAL_IDLE is not set
 CONFIG_SMP=y
 CONFIG_NR_CPUS=16
 # CONFIG_PREEMPT is not set
@@ -78,6 +80,10 @@ CONFIG_IA32_SUPPORT=y
 CONFIG_COMPAT=y
 CONFIG_PERFMON=y
 CONFIG_IA64_PALINFO=y
+
+#
+# Firmware Drivers
+#
 CONFIG_EFI_VARS=y
 CONFIG_BINFMT_ELF=y
 CONFIG_BINFMT_MISC=y
@@ -87,13 +93,12 @@ CONFIG_BINFMT_MISC=y
 #
 CONFIG_PM=y
 CONFIG_ACPI=y
-CONFIG_ACPI_INTERPRETER=y
-CONFIG_ACPI_KERNEL_CONFIG=y
 
 #
 # ACPI (Advanced Configuration and Power Interface) Support
 #
 CONFIG_ACPI_BOOT=y
+CONFIG_ACPI_INTERPRETER=y
 CONFIG_ACPI_BUTTON=y
 CONFIG_ACPI_FAN=y
 CONFIG_ACPI_PROCESSOR=y
@@ -109,6 +114,7 @@ CONFIG_ACPI_SYSTEM=y
 #
 CONFIG_PCI=y
 CONFIG_PCI_DOMAINS=y
+# CONFIG_PCI_USE_VECTOR is not set
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
 
@@ -149,7 +155,6 @@ CONFIG_PCI_NAMES=y
 #
 # Block devices
 #
-# CONFIG_BLK_DEV_FD is not set
 # CONFIG_BLK_CPQ_DA is not set
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
@@ -345,7 +350,6 @@ CONFIG_IP_MULTICAST=y
 # CONFIG_NET_IPGRE is not set
 # CONFIG_IP_MROUTE is not set
 # CONFIG_ARPD is not set
-# CONFIG_INET_ECN is not set
 # CONFIG_SYN_COOKIES is not set
 # CONFIG_INET_AH is not set
 # CONFIG_INET_ESP is not set
@@ -356,8 +360,6 @@ CONFIG_IP_MULTICAST=y
 #
 # CONFIG_IP_VS is not set
 # CONFIG_IPV6 is not set
-# CONFIG_DECNET is not set
-# CONFIG_BRIDGE is not set
 CONFIG_NETFILTER=y
 # CONFIG_NETFILTER_DEBUG is not set
 
@@ -376,10 +378,11 @@ CONFIG_IP_NF_ARPTABLES=y
 #
 # SCTP Configuration (EXPERIMENTAL)
 #
-CONFIG_IPV6_SCTP__=y
 # CONFIG_IP_SCTP is not set
 # CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
 # CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
 # CONFIG_LLC2 is not set
 # CONFIG_IPX is not set
 # CONFIG_ATALK is not set
@@ -400,16 +403,21 @@ CONFIG_IPV6_SCTP__=y
 # Network testing
 #
 # CONFIG_NET_PKTGEN is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
 CONFIG_NETDEVICES=y
+CONFIG_DUMMY=y
+CONFIG_BONDING=y
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
 
 #
 # ARCnet devices
 #
 # CONFIG_ARCNET is not set
-CONFIG_DUMMY=y
-CONFIG_BONDING=y
-# CONFIG_EQUALIZER is not set
-# CONFIG_TUN is not set
 
 #
 # Ethernet (10 or 100Mbit)
@@ -456,7 +464,6 @@ CONFIG_E1000=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
-# CONFIG_SIS190 is not set
 # CONFIG_SK98LIN is not set
 CONFIG_TIGON3=y
 
@@ -464,45 +471,29 @@ CONFIG_TIGON3=y
 # Ethernet (10000 Mbit)
 #
 # CONFIG_IXGB is not set
-# CONFIG_FDDI is not set
-# CONFIG_HIPPI is not set
-# CONFIG_PPP is not set
-# CONFIG_SLIP is not set
-
-#
-# Wireless LAN (non-hamradio)
-#
-# CONFIG_NET_RADIO is not set
+# CONFIG_S2IO is not set
 
 #
 # Token Ring devices
 #
 # CONFIG_TR is not set
-# CONFIG_NET_FC is not set
-# CONFIG_SHAPER is not set
-# CONFIG_NETCONSOLE is not set
-
-#
-# Wan interfaces
-#
-# CONFIG_WAN is not set
 
 #
-# Amateur Radio support
-#
-# CONFIG_HAMRADIO is not set
-
-#
-# IrDA (infrared) support
+# Wireless LAN (non-hamradio)
 #
-# CONFIG_IRDA is not set
+# CONFIG_NET_RADIO is not set
 
 #
-# Bluetooth support
+# Wan interfaces
 #
-# CONFIG_BT is not set
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
 
 #
 # ISDN subsystem
@@ -577,11 +568,6 @@ CONFIG_SERIAL_CORE_CONSOLE=y
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
-
-#
-# Mice
-#
-# CONFIG_BUSMOUSE is not set
 # CONFIG_QIC02_TAPE is not set
 
 #
@@ -594,7 +580,6 @@ CONFIG_LEGACY_PTY_COUNT=256
 #
 # CONFIG_WATCHDOG is not set
 # CONFIG_HW_RANDOM is not set
-# CONFIG_GEN_RTC is not set
 CONFIG_EFI_RTC=y
 # CONFIG_DTLK is not set
 # CONFIG_R3964 is not set
@@ -631,6 +616,7 @@ CONFIG_I2C_ALGOBIT=y
 # I2C Hardware Bus support
 #
 # CONFIG_I2C_ALI1535 is not set
+# CONFIG_I2C_ALI1563 is not set
 # CONFIG_I2C_ALI15X3 is not set
 # CONFIG_I2C_AMD756 is not set
 # CONFIG_I2C_AMD8111 is not set
@@ -674,6 +660,8 @@ CONFIG_I2C_ALGOBIT=y
 # Other I2C Chip support
 #
 # CONFIG_SENSORS_EEPROM is not set
+# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -704,7 +692,7 @@ CONFIG_FB=y
 CONFIG_FB_RIVA=m
 # CONFIG_FB_MATROX is not set
 # CONFIG_FB_RADEON_OLD is not set
-CONFIG_FB_RADEON=y
+CONFIG_FB_RADEON=m
 CONFIG_FB_RADEON_I2C=y
 # CONFIG_FB_RADEON_DEBUG is not set
 # CONFIG_FB_ATY128 is not set
@@ -837,6 +825,8 @@ CONFIG_USB_BANDWIDTH=y
 # USB Host Controller Drivers
 #
 CONFIG_USB_EHCI_HCD=y
+# CONFIG_USB_EHCI_SPLIT_ISO is not set
+CONFIG_USB_EHCI_ROOT_HUB_TT=y
 CONFIG_USB_OHCI_HCD=y
 CONFIG_USB_UHCI_HCD=y
 
@@ -861,7 +851,9 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_WACOM is not set
 # CONFIG_USB_KBTAB is not set
 # CONFIG_USB_POWERMATE is not set
+# CONFIG_USB_MTOUCH is not set
 # CONFIG_USB_XPAD is not set
+# CONFIG_USB_ATI_REMOTE is not set
 
 #
 # USB Imaging devices
@@ -908,6 +900,7 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_LEGOTOWER is not set
 # CONFIG_USB_LCD is not set
 # CONFIG_USB_LED is not set
+# CONFIG_USB_CYTHERM is not set
 # CONFIG_USB_TEST is not set
 
 #
@@ -959,6 +952,7 @@ CONFIG_VFAT_FS=y
 #
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
 # CONFIG_DEVFS_FS is not set
 # CONFIG_DEVPTS_FS_XATTR is not set
 CONFIG_TMPFS=y
@@ -1075,6 +1069,7 @@ CONFIG_NLS_UTF8=y
 # Library routines
 #
 CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
 
 #
 # Profiling support
@@ -1090,6 +1085,7 @@ CONFIG_DEBUG_KERNEL=y
 CONFIG_IA64_PRINT_HAZARDS=y
 CONFIG_DISABLE_VHPT=y
 CONFIG_MAGIC_SYSRQ=y
+# CONFIG_IA64_EARLY_PRINTK_VGA is not set
 # CONFIG_DEBUG_SLAB is not set
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
@@ -1123,4 +1119,6 @@ CONFIG_CRYPTO_DES=y
 # CONFIG_CRYPTO_CAST6 is not set
 # CONFIG_CRYPTO_ARC4 is not set
 # CONFIG_CRYPTO_DEFLATE is not set
+# CONFIG_CRYPTO_MICHAEL_MIC is not set
+# CONFIG_CRYPTO_CRC32C is not set
 # CONFIG_CRYPTO_TEST is not set

arch/ia64/hp/sim/boot/boot_head.S

@@ -105,37 +105,37 @@ static:	cmp.eq p6,p7=6,r28		/* PAL_PTCE_INFO */
 1:	cmp.eq p6,p7=15,r28		/* PAL_PERF_MON_INFO */
 (p7)	br.cond.sptk.few 1f
 	mov r8=0			/* status = 0 */
-	movl r9 =0x12082004		/* generic=4 width=32 retired=8 cycles=18 */
+	movl r9 =0x08122f04		/* generic=4 width=47 retired=8 cycles=18 */
 	mov r10=0			/* reserved */
 	mov r11=0			/* reserved */
 	mov r16=0xffff			/* implemented PMC */
-	mov r17=0xffff			/* implemented PMD */
+	mov r17=0x3ffff			/* implemented PMD */
 	add r18=8,r29			/* second index */
 	;;
 	st8 [r29]=r16,16		/* store implemented PMC */
 	st8 [r18]=r0,16			/* clear remaining bits  */
 	;;
-	st8 [r29]=r0,16			/* store implemented PMC */
+	st8 [r29]=r0,16			/* clear remaining bits  */
 	st8 [r18]=r0,16			/* clear remaining bits  */
 	;;
 	st8 [r29]=r17,16		/* store implemented PMD */
 	st8 [r18]=r0,16			/* clear remaining bits  */
 	mov r16=0xf0			/* cycles count capable PMC */
 	;;
-	st8 [r29]=r0,16			/* store implemented PMC */
+	st8 [r29]=r0,16			/* clear remaining bits  */
 	st8 [r18]=r0,16			/* clear remaining bits  */
-	mov r17=0x10			/* retired bundles capable PMC */
+	mov r17=0xf0			/* retired bundles capable PMC */
 	;;
 	st8 [r29]=r16,16		/* store cycles capable */
 	st8 [r18]=r0,16			/* clear remaining bits  */
 	;;
-	st8 [r29]=r0,16			/* store implemented PMC */
+	st8 [r29]=r0,16			/* clear remaining bits  */
 	st8 [r18]=r0,16			/* clear remaining bits  */
 	;;
 	st8 [r29]=r17,16		/* store retired bundle capable */
 	st8 [r18]=r0,16			/* clear remaining bits  */
 	;;
-	st8 [r29]=r0,16			/* store implemented PMC */
+	st8 [r29]=r0,16			/* clear remaining bits  */
 	st8 [r18]=r0,16			/* clear remaining bits  */
 	;;
 1:	br.cond.sptk.few rp

arch/ia64/ia32/Makefile

@@ -6,3 +6,7 @@ obj-y := ia32_entry.o sys_ia32.o ia32_ioctl.o ia32_signal.o \
 	 ia32_support.o ia32_traps.o binfmt_elf32.o ia32_ldt.o
 
 CFLAGS_ia32_ioctl.o += -Ifs/
+
+# Don't let GCC uses f16-f31 so that save_ia32_fpstate_live() and
+# restore_ia32_fpstate_live() can be sure the live register contain user-level state.
+CFLAGS_ia32_signal.o += -mfixed-range=f16-f31

arch/ia64/ia32/ia32_signal.c

@@ -43,22 +43,6 @@
 #define __IA32_NR_sigreturn            119
 #define __IA32_NR_rt_sigreturn         173
 
-#ifdef ASM_SUPPORTED
-/*
- * Don't let GCC uses f16-f31 so that save_ia32_fpstate_live() and
- * restore_ia32_fpstate_live() can be sure the live register contain user-level state.
- */
-register double f16 asm ("f16"); register double f17 asm ("f17");
-register double f18 asm ("f18"); register double f19 asm ("f19");
-register double f20 asm ("f20"); register double f21 asm ("f21");
-register double f22 asm ("f22"); register double f23 asm ("f23");
-
-register double f24 asm ("f24"); register double f25 asm ("f25");
-register double f26 asm ("f26"); register double f27 asm ("f27");
-register double f28 asm ("f28"); register double f29 asm ("f29");
-register double f30 asm ("f30"); register double f31 asm ("f31");
-#endif
-
 struct sigframe_ia32
 {
        int pretcode;
@@ -173,7 +157,8 @@ copy_siginfo_to_user32 (siginfo_t32 *to, siginfo_t *from)
 		case __SI_MESGQ >> 16:
 			err |= __put_user(from->si_uid, &to->si_uid);
 			err |= __put_user(from->si_pid, &to->si_pid);
-			err |= __put_user(from->si_ptr, &to->si_ptr);
+			addr = (unsigned long) from->si_ptr;
+			err |= __put_user(addr, &to->si_ptr);
 			break;
 		}
 	}

arch/ia64/kernel/Makefile

@@ -23,6 +23,9 @@ targets += gate.so gate-syms.o
 
 extra-y += gate.so gate-syms.o gate.lds.s gate.o
 
+# fp_emulate() expects f2-f5,f16-f31 to contain the user-level state.
+CFLAGS_traps.o  += -mfixed-range=f2-f5,f16-f31
+
 AFLAGS_gate.lds.o += -P -C -U$(ARCH)
 
 quiet_cmd_gate = GATE $@

arch/ia64/kernel/efi.c

@@ -39,7 +39,7 @@ extern efi_status_t efi_call_phys (void *, ...);
 struct efi efi;
 EXPORT_SYMBOL(efi);
 static efi_runtime_services_t *runtime;
-static unsigned long mem_limit = ~0UL;
+static unsigned long mem_limit = ~0UL, max_addr = ~0UL;
 
 #define efi_call_virt(f, args...)	(*(f))(args)
 
@@ -290,6 +290,7 @@ efi_memmap_walk (efi_freemem_callback_t callback, void *arg)
 	void *efi_map_start, *efi_map_end, *p, *q;
 	efi_memory_desc_t *md, *check_md;
 	u64 efi_desc_size, start, end, granule_addr, last_granule_addr, first_non_wb_addr = 0;
+	unsigned long total_mem = 0;
 
 	efi_map_start = __va(ia64_boot_param->efi_memmap);
 	efi_map_end   = efi_map_start + ia64_boot_param->efi_memmap_size;
@@ -331,12 +332,18 @@ efi_memmap_walk (efi_freemem_callback_t callback, void *arg)
 			trim_top(md, last_granule_addr);
 
 		if (is_available_memory(md)) {
-			if (md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) > mem_limit) {
-				if (md->phys_addr > mem_limit)
+			if (md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) > max_addr) {
+				if (md->phys_addr > max_addr)
 					continue;
-				md->num_pages = (mem_limit - md->phys_addr) >> EFI_PAGE_SHIFT;
+				md->num_pages = (max_addr - md->phys_addr) >> EFI_PAGE_SHIFT;
 			}
 
+			if (total_mem >= mem_limit)
+				continue;
+			total_mem += (md->num_pages << EFI_PAGE_SHIFT);
+			if (total_mem > mem_limit)
+				md->num_pages -= ((total_mem - mem_limit) >> EFI_PAGE_SHIFT);
+
 			if (md->num_pages == 0)
 				continue;
 
@@ -470,7 +477,13 @@ efi_init (void)
 	for (cp = saved_command_line; *cp; ) {
 		if (memcmp(cp, "mem=", 4) == 0) {
 			cp += 4;
-			mem_limit = memparse(cp, &end) - 1;
+			mem_limit = memparse(cp, &end) - 2;
+			if (end != cp)
+				break;
+			cp = end;
+		} else if (memcmp(cp, "max_addr=", 9) == 0) {
+			cp += 9;
+			max_addr = memparse(cp, &end) - 1;
 			if (end != cp)
 				break;
 			cp = end;
@@ -481,8 +494,8 @@ efi_init (void)
 				++cp;
 		}
 	}
-	if (mem_limit != ~0UL)
-		printk(KERN_INFO "Ignoring memory above %luMB\n", mem_limit >> 20);
+	if (max_addr != ~0UL)
+		printk(KERN_INFO "Ignoring memory above %luMB\n", max_addr >> 20);
 
 	efi.systab = __va(ia64_boot_param->efi_systab);
 

arch/ia64/kernel/fsys.S

@@ -345,40 +345,33 @@ ENTRY(fsys_rt_sigprocmask)
 	.altrp b6
 	.body
 
-	mf					// ensure reading of current->blocked is ordered
 	add r2=IA64_TASK_BLOCKED_OFFSET,r16
 	add r9=TI_FLAGS+IA64_TASK_SIZE,r16
+	cmp4.ltu p6,p0=SIG_SETMASK,r32
+
+	cmp.ne p15,p0=r0,r34			// oset != NULL?
+	tnat.nz p8,p0=r34
+	add r31=IA64_TASK_SIGHAND_OFFSET,r16
 	;;
-	/*
-	 * Since we're only reading a single word, we can do it
-	 * atomically without acquiring current->sighand->siglock.  To
-	 * be on the safe side, we need a fully-ordered load, though:
-	 */
-	ld8.acq r3=[r2]				// read/prefetch current->blocked
+	ld8 r3=[r2]				// read/prefetch current->blocked
 	ld4 r9=[r9]
-	add r31=IA64_TASK_SIGHAND_OFFSET,r16
+	tnat.nz.or p6,p0=r35
+
+	cmp.ne.or p6,p0=_NSIG_WORDS*8,r35
+	tnat.nz.or p6,p0=r32
+(p6)	br.spnt.few .fail_einval		// fail with EINVAL
 	;;
 #ifdef CONFIG_SMP
 	ld8 r31=[r31]				// r31 <- current->sighand
 #endif
 	and r9=TIF_ALLWORK_MASK,r9
-	tnat.nz p6,p0=r32
-	;;
-	cmp.ne p7,p0=0,r9
-	tnat.nz.or p6,p0=r35
-	tnat.nz p8,p0=r34
-	;;
-	cmp.ne p15,p0=r0,r34			// oset != NULL?
-	cmp.ne.or p6,p0=_NSIG_WORDS*8,r35
 	tnat.nz.or p8,p0=r33
-
-(p6)	br.spnt.few .fail_einval		// fail with EINVAL
-(p7)	br.spnt.many fsys_fallback_syscall	// got pending kernel work...
-(p8)	br.spnt.few .fail_efault		// fail with EFAULT
 	;;
-
-	cmp.eq p6,p7=r0,r33			// set == NULL?
+	cmp.ne p7,p0=0,r9
+	cmp.eq p6,p0=r0,r33			// set == NULL?
 	add r31=IA64_SIGHAND_SIGLOCK_OFFSET,r31	// r31 <- current->sighand->siglock
+(p8)	br.spnt.few .fail_efault		// fail with EFAULT
+(p7)	br.spnt.many fsys_fallback_syscall	// got pending kernel work...
 (p6)	br.dpnt.many .store_mask		// -> short-circuit to just reading the signal mask
 
 	/* Argh, we actually have to do some work and _update_ the signal mask: */
@@ -462,12 +455,10 @@ EX(.fail_efault, ld8 r14=[r33])			// r14 <- *set
 	st4.rel [r31]=r0			// release the lock
 #endif
 	ssm psr.i
-	cmp.ne p9,p0=r8,r0			// check for bad HOW value
 	;;
 
 	srlz.d					// ensure psr.i is set again
 	mov r18=0					// i must not leak kernel bits...
-(p9)	br.spnt.few .fail_einval		// bail out for bad HOW value
 
 .store_mask:
 EX(.fail_efault, (p15) probe.w.fault r34, 3)	// verify user has write-access to *oset
@@ -574,6 +565,10 @@ GLOBAL_ENTRY(fsys_bubble_down)
 	or r29=r8,r29			// construct cr.ipsr value to save
 	addl r22=IA64_RBS_OFFSET,r2	// compute base of RBS
 	;;
+	// GAS reports a spurious RAW hazard on the read of ar.rnat because it thinks
+	// we may be reading ar.itc after writing to psr.l.  Avoid that message with
+	// this directive:
+	dv_serialize_data
 	mov.m r24=ar.rnat		// read ar.rnat (5 cyc lat)
 	lfetch.fault.excl.nt1 [r22]
 	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r2

arch/ia64/kernel/head.S

@@ -56,8 +56,7 @@ halt_msg:
 GLOBAL_ENTRY(_start)
 start_ap:
 	.prologue
-	.save rp, r4		// terminate unwind chain with a NULL rp
-	mov r4=r0
+	.save rp, r0		// terminate unwind chain with a NULL rp
 	.body
 
 	rsm psr.i | psr.ic

arch/ia64/kernel/ivt.S

@@ -181,6 +181,12 @@ ENTRY(vhpt_miss)
 (p7)	itc.d r24
 	;;
 #ifdef CONFIG_SMP
+	/*
+	 * Tell the assemblers dependency-violation checker that the above "itc" instructions
+	 * cannot possibly affect the following loads:
+	 */
+	dv_serialize_data
+
 	/*
 	 * Re-check L2 and L3 pagetable.  If they changed, we may have received a ptc.g
 	 * between reading the pagetable and the "itc".  If so, flush the entry we
@@ -229,6 +235,12 @@ ENTRY(itlb_miss)
 	itc.i r18
 	;;
 #ifdef CONFIG_SMP
+	/*
+	 * Tell the assemblers dependency-violation checker that the above "itc" instructions
+	 * cannot possibly affect the following loads:
+	 */
+	dv_serialize_data
+
 	ld8 r19=[r17]				// read L3 PTE again and see if same
 	mov r20=PAGE_SHIFT<<2			// setup page size for purge
 	;;
@@ -267,6 +279,12 @@ dtlb_fault:
 	itc.d r18
 	;;
 #ifdef CONFIG_SMP
+	/*
+	 * Tell the assemblers dependency-violation checker that the above "itc" instructions
+	 * cannot possibly affect the following loads:
+	 */
+	dv_serialize_data
+
 	ld8 r19=[r17]				// read L3 PTE again and see if same
 	mov r20=PAGE_SHIFT<<2			// setup page size for purge
 	;;
@@ -504,6 +522,12 @@ ENTRY(dirty_bit)
 	;;
 (p6)	itc.d r25				// install updated PTE
 	;;
+	/*
+	 * Tell the assemblers dependency-violation checker that the above "itc" instructions
+	 * cannot possibly affect the following loads:
+	 */
+	dv_serialize_data
+
 	ld8 r18=[r17]				// read PTE again
 	;;
 	cmp.eq p6,p7=r18,r25			// is it same as the newly installed
@@ -563,6 +587,12 @@ ENTRY(iaccess_bit)
 	;;
 (p6)	itc.i r25				// install updated PTE
 	;;
+	/*
+	 * Tell the assemblers dependency-violation checker that the above "itc" instructions
+	 * cannot possibly affect the following loads:
+	 */
+	dv_serialize_data
+
 	ld8 r18=[r17]				// read PTE again
 	;;
 	cmp.eq p6,p7=r18,r25			// is it same as the newly installed
@@ -610,6 +640,11 @@ ENTRY(daccess_bit)
 	cmp.eq p6,p7=r26,r18
 	;;
 (p6)	itc.d r25				// install updated PTE
+	/*
+	 * Tell the assemblers dependency-violation checker that the above "itc" instructions
+	 * cannot possibly affect the following loads:
+	 */
+	dv_serialize_data
 	;;
 	ld8 r18=[r17]				// read PTE again
 	;;

arch/ia64/kernel/machvec.c

 #include <linux/config.h>
 #include <linux/module.h>
 
+#include <asm/machvec.h>
 #include <asm/system.h>
 
 #ifdef CONFIG_IA64_GENERIC
@@ -8,7 +9,6 @@
 #include <linux/kernel.h>
 #include <linux/string.h>
 
-#include <asm/machvec.h>
 #include <asm/page.h>
 
 struct ia64_machine_vector ia64_mv;

arch/ia64/kernel/perfmon_generic.h

@@ -6,13 +6,6 @@
  *               Stephane Eranian <eranian@hpl.hp.com>
  */
 
-
-#define RDEP(x)	(1UL<<(x))
-
-#if defined(CONFIG_ITANIUM) || defined (CONFIG_MCKINLEY)
-#error "This file should not be used when CONFIG_ITANIUM or CONFIG_MCKINLEY is defined"
-#endif
-
 static pfm_reg_desc_t pfm_gen_pmc_desc[PMU_MAX_PMCS]={
 /* pmc0  */ { PFM_REG_CONTROL , 0, 0x1UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
 /* pmc1  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
@@ -40,10 +33,9 @@ static pfm_reg_desc_t pfm_gen_pmd_desc[PMU_MAX_PMDS]={
 /*
  * impl_pmcs, impl_pmds are computed at runtime to minimize errors!
  */
-static pmu_config_t pmu_conf={
+static pmu_config_t pmu_conf_gen={
 	.pmu_name   = "Generic",
 	.pmu_family = 0xff, /* any */
-	.enabled    = 0,
 	.ovfl_val   = (1UL << 32) - 1,
 	.num_ibrs   = 0, /* does not use */
 	.num_dbrs   = 0, /* does not use */

arch/ia64/kernel/perfmon_hpsim.h

-/*
- * This file contains the HP SKI Simulator PMU register description tables
- * and pmc checkers used by perfmon.c.
- *
- * Copyright (C) 2002-2003  Hewlett Packard Co
- *               Stephane Eranian <eranian@hpl.hp.com>
- *
- * File mostly contributed by Ian Wienand <ianw@gelato.unsw.edu.au>
- *
- * This file is included as a dummy template so the kernel does not
- * try to initalize registers the simulator can't handle.
- *
- * Note the simulator does not (currently) implement these registers, i.e.,
- * they do not count anything. But you can read/write them.
- */
-
-#define RDEP(x)	(1UL<<(x))
-
-#ifndef CONFIG_IA64_HP_SIM
-#error "This file should only be included for the HP Simulator"
-#endif
-
-static pfm_reg_desc_t pfm_hpsim_pmc_desc[PMU_MAX_PMCS]={
-/* pmc0  */ { PFM_REG_CONTROL , 0, 0x1UL, -1UL, NULL, NULL, {0UL, 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc1  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL, 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc2  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL, 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc3  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL, 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc4  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(4), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc5  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(5), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc6  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(6), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc7  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(7), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc8  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(8), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc9  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(9), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc10 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(10), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc11 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(11), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc12 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(12), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc13 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(13), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc14 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(14), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmc15 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(15), 0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-	    { PFM_REG_END     , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}}, /* end marker */
-};
-
-static pfm_reg_desc_t pfm_hpsim_pmd_desc[PMU_MAX_PMDS]={
-/* pmd0  */ { PFM_REG_BUFFER, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmd1  */ { PFM_REG_BUFFER, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmd2  */ { PFM_REG_BUFFER, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmd3  */ { PFM_REG_BUFFER, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
-/* pmd4  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(4),0UL, 0UL, 0UL}},
-/* pmd5  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(5),0UL, 0UL, 0UL}},
-/* pmd6  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(6),0UL, 0UL, 0UL}},
-/* pmd7  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(7),0UL, 0UL, 0UL}},
-/* pmd8  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(8),0UL, 0UL, 0UL}},
-/* pmd9  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(9),0UL, 0UL, 0UL}},
-/* pmd10 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(10),0UL, 0UL, 0UL}},
-/* pmd11 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
-/* pmd12 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
-/* pmd13 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(13),0UL, 0UL, 0UL}},
-/* pmd14 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(14),0UL, 0UL, 0UL}},
-/* pmd15 */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(15),0UL, 0UL, 0UL}},
-	    { PFM_REG_END     , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}}, /* end marker */
-};
-
-/*
- * impl_pmcs, impl_pmds are computed at runtime to minimize errors!
- */
-static pmu_config_t pmu_conf={
-	.pmu_name   = "hpsim",
-	.pmu_family = 0x7, /* ski emulator reports as Itanium */
-	.enabled    = 0,
-	.ovfl_val   = (1UL << 32) - 1,
-	.num_ibrs   = 0, /* does not use */
-	.num_dbrs   = 0, /* does not use */
-	.pmd_desc   = pfm_hpsim_pmd_desc,
-	.pmc_desc   = pfm_hpsim_pmc_desc
-};

arch/ia64/kernel/perfmon_itanium.h

@@ -5,15 +5,7 @@
  * Copyright (C) 2002-2003  Hewlett Packard Co
  *               Stephane Eranian <eranian@hpl.hp.com>
  */
-
-#define RDEP(x)	(1UL<<(x))
-
-#ifndef CONFIG_ITANIUM
-#error "This file is only valid when CONFIG_ITANIUM is defined"
-#endif
-
 static int pfm_ita_pmc_check(struct task_struct *task, pfm_context_t *ctx, unsigned int cnum, unsigned long *val, struct pt_regs *regs);
-static int pfm_write_ibr_dbr(int mode, pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs);
 
 static pfm_reg_desc_t pfm_ita_pmc_desc[PMU_MAX_PMCS]={
 /* pmc0  */ { PFM_REG_CONTROL , 0, 0x1UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
@@ -55,31 +47,22 @@ static pfm_reg_desc_t pfm_ita_pmd_desc[PMU_MAX_PMDS]={
 	    { PFM_REG_END     , 0, 0UL, -1UL, NULL, NULL, {0,}, {0,}}, /* end marker */
 };
 
-/*
- * impl_pmcs, impl_pmds are computed at runtime to minimize errors!
- */
-static pmu_config_t pmu_conf={
-	.pmu_name      = "Itanium",
-	.pmu_family    = 0x7,
-	.enabled       = 0,
-	.ovfl_val      = (1UL << 32) - 1,
-	.pmd_desc      = pfm_ita_pmd_desc,
-	.pmc_desc      = pfm_ita_pmc_desc,
-	.num_ibrs      = 8,
-	.num_dbrs      = 8,
-	.use_rr_dbregs = 1 /* debug register are use for range retrictions */
-};
-
 static int
 pfm_ita_pmc_check(struct task_struct *task, pfm_context_t *ctx, unsigned int cnum, unsigned long *val, struct pt_regs *regs)
 {
 	int ret;
+	int is_loaded;
+
+	/* sanitfy check */
+	if (ctx == NULL) return -EINVAL;
+
+	is_loaded = ctx->ctx_state == PFM_CTX_LOADED || ctx->ctx_state == PFM_CTX_MASKED;
 
 	/*
 	 * we must clear the (instruction) debug registers if pmc13.ta bit is cleared
 	 * before they are written (fl_using_dbreg==0) to avoid picking up stale information.
 	 */
-	if (cnum == 13 && ((*val & 0x1) == 0UL) && ctx->ctx_fl_using_dbreg == 0) {
+	if (cnum == 13 && is_loaded && ((*val & 0x1) == 0UL) && ctx->ctx_fl_using_dbreg == 0) {
 
 		DPRINT(("pmc[%d]=0x%lx has active pmc13.ta cleared, clearing ibr\n", cnum, *val));
 
@@ -98,7 +81,7 @@ pfm_ita_pmc_check(struct task_struct *task, pfm_context_t *ctx, unsigned int cnu
 	 * we must clear the (data) debug registers if pmc11.pt bit is cleared
 	 * before they are written (fl_using_dbreg==0) to avoid picking up stale information.
 	 */
-	if (cnum == 11 && ((*val >> 28)& 0x1) == 0 && ctx->ctx_fl_using_dbreg == 0) {
+	if (cnum == 11 && is_loaded && ((*val >> 28)& 0x1) == 0 && ctx->ctx_fl_using_dbreg == 0) {
 
 		DPRINT(("pmc[%d]=0x%lx has active pmc11.pt cleared, clearing dbr\n", cnum, *val));
 
@@ -115,3 +98,18 @@ pfm_ita_pmc_check(struct task_struct *task, pfm_context_t *ctx, unsigned int cnu
 	return 0;
 }
 
+/*
+ * impl_pmcs, impl_pmds are computed at runtime to minimize errors!
+ */
+static pmu_config_t pmu_conf_ita={
+	.pmu_name      = "Itanium",
+	.pmu_family    = 0x7,
+	.ovfl_val      = (1UL << 32) - 1,
+	.pmd_desc      = pfm_ita_pmd_desc,
+	.pmc_desc      = pfm_ita_pmc_desc,
+	.num_ibrs      = 8,
+	.num_dbrs      = 8,
+	.use_rr_dbregs = 1, /* debug register are use for range retrictions */
+};
+
+

arch/ia64/kernel/perfmon_mckinley.h

@@ -5,15 +5,7 @@
  * Copyright (C) 2002-2003  Hewlett Packard Co
  *               Stephane Eranian <eranian@hpl.hp.com>
  */
-
-#define RDEP(x)	(1UL<<(x))
-
-#ifndef CONFIG_MCKINLEY
-#error "This file is only valid when CONFIG_MCKINLEY is defined"
-#endif
-
 static int pfm_mck_pmc_check(struct task_struct *task, pfm_context_t *ctx, unsigned int cnum, unsigned long *val, struct pt_regs *regs);
-static int pfm_write_ibr_dbr(int mode, pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs);
 
 static pfm_reg_desc_t pfm_mck_pmc_desc[PMU_MAX_PMCS]={
 /* pmc0  */ { PFM_REG_CONTROL , 0, 0x1UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
@@ -57,21 +49,6 @@ static pfm_reg_desc_t pfm_mck_pmd_desc[PMU_MAX_PMDS]={
 	    { PFM_REG_END     , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}}, /* end marker */
 };
 
-/*
- * impl_pmcs, impl_pmds are computed at runtime to minimize errors!
- */
-static pmu_config_t pmu_conf={
-	.pmu_name      = "Itanium 2",
-	.pmu_family    = 0x1f,
-	.enabled       = 0,
-	.ovfl_val      = (1UL << 47) - 1,
-	.pmd_desc      = pfm_mck_pmd_desc,
-	.pmc_desc      = pfm_mck_pmc_desc,
-	.num_ibrs       = 8,
-	.num_dbrs       = 8,
-	.use_rr_dbregs = 1 /* debug register are use for range retrictions */
-};
-
 /*
  * PMC reserved fields must have their power-up values preserved
  */
@@ -120,12 +97,11 @@ pfm_mck_pmc_check(struct task_struct *task, pfm_context_t *ctx, unsigned int cnu
 	 * 	one of the pmc13.cfg_dbrpXX field is different from 0x3
 	 * AND
 	 * 	at the corresponding pmc13.ena_dbrpXX is set.
-	 *
-	 * For now, we just check on cfg_dbrXX != 0x3.
 	 */
 	DPRINT(("cnum=%u val=0x%lx, using_dbreg=%d loaded=%d\n", cnum, *val, ctx->ctx_fl_using_dbreg, is_loaded));
 
-	if (cnum == 13 && is_loaded && ((*val & 0x18181818UL) != 0x18181818UL) && ctx->ctx_fl_using_dbreg == 0) {
+	if (cnum == 13 && is_loaded
+	    && (*val & 0x1e00000000000UL) && (*val & 0x18181818UL) != 0x18181818UL && ctx->ctx_fl_using_dbreg == 0) {
 
 		DPRINT(("pmc[%d]=0x%lx has active pmc13 settings, clearing dbr\n", cnum, *val));
 
@@ -192,3 +168,20 @@ pfm_mck_pmc_check(struct task_struct *task, pfm_context_t *ctx, unsigned int cnu
 
 	return ret ? -EINVAL : 0;
 }
+
+/*
+ * impl_pmcs, impl_pmds are computed at runtime to minimize errors!
+ */
+static pmu_config_t pmu_conf_mck={
+	.pmu_name      = "Itanium 2",
+	.pmu_family    = 0x1f,
+	.flags	       = PFM_PMU_IRQ_RESEND,
+	.ovfl_val      = (1UL << 47) - 1,
+	.pmd_desc      = pfm_mck_pmd_desc,
+	.pmc_desc      = pfm_mck_pmc_desc,
+	.num_ibrs       = 8,
+	.num_dbrs       = 8,
+	.use_rr_dbregs = 1 /* debug register are use for range retrictions */
+};
+
+

arch/ia64/kernel/process.c

@@ -96,6 +96,7 @@ show_regs (struct pt_regs *regs)
 {
 	unsigned long ip = regs->cr_iip + ia64_psr(regs)->ri;
 
+	print_modules();
 	printk("\nPid: %d, CPU %d, comm: %20s\n", current->pid, smp_processor_id(), current->comm);
 	printk("psr : %016lx ifs : %016lx ip  : [<%016lx>]    %s\n",
 	       regs->cr_ipsr, regs->cr_ifs, ip, print_tainted());

arch/ia64/kernel/signal.c

@@ -42,23 +42,6 @@
 # define GET_SIGSET(k,u)	__get_user((k)->sig[0], &(u)->sig[0])
 #endif
 
-#ifdef ASM_SUPPORTED
-/*
- * Don't let GCC uses f16-f31 so that when we setup/restore the registers in the signal
- * context in __kernel_sigtramp(), we can be sure that registers f16-f31 contain user-level
- * values.
- */
-register double f16 asm ("f16"); register double f17 asm ("f17");
-register double f18 asm ("f18"); register double f19 asm ("f19");
-register double f20 asm ("f20"); register double f21 asm ("f21");
-register double f22 asm ("f22"); register double f23 asm ("f23");
-
-register double f24 asm ("f24"); register double f25 asm ("f25");
-register double f26 asm ("f26"); register double f27 asm ("f27");
-register double f28 asm ("f28"); register double f29 asm ("f29");
-register double f30 asm ("f30"); register double f31 asm ("f31");
-#endif
-
 long
 ia64_rt_sigsuspend (sigset_t *uset, size_t sigsetsize, struct sigscratch *scr)
 {

arch/ia64/kernel/smp.c

@@ -28,7 +28,6 @@
 #include <linux/mm.h>
 #include <linux/cache.h>
 #include <linux/delay.h>
-#include <linux/cache.h>
 #include <linux/efi.h>
 
 #include <asm/atomic.h>

arch/ia64/kernel/smpboot.c

@@ -212,6 +212,14 @@ ia64_sync_itc (unsigned int master)
 	} t[NUM_ROUNDS];
 #endif
 
+	/*
+	 * Make sure local timer ticks are disabled while we sync.  If
+	 * they were enabled, we'd have to worry about nasty issues
+	 * like setting the ITC ahead of (or a long time before) the
+	 * next scheduled tick.
+	 */
+	BUG_ON((ia64_get_itv() & (1 << 16)) == 0);
+
 	go[MASTER] = 1;
 
 	if (smp_call_function_single(master, sync_master, NULL, 1, 0) < 0) {
@@ -257,16 +265,6 @@ ia64_sync_itc (unsigned int master)
 
 	printk(KERN_INFO "CPU %d: synchronized ITC with CPU %u (last diff %ld cycles, "
 	       "maxerr %lu cycles)\n", smp_processor_id(), master, delta, rt);
-
-	/*
-	 * Check whether we sync'd the itc ahead of the next timer interrupt.  If so, just
-	 * reset it.
-	 */
-	if (time_after(ia64_get_itc(), local_cpu_data->itm_next)) {
-		Dprintk("CPU %d: oops, jumped a timer tick; resetting timer.\n",
-			smp_processor_id());
-		ia64_cpu_local_tick();
-	}
 }
 
 /*
@@ -302,16 +300,6 @@ smp_callin (void)
 
 	smp_setup_percpu_timer();
 
-	/*
-	 * Get our bogomips.
-	 */
-	ia64_init_itm();
-
-	/*
-	 * Set I/O port base per CPU
-	 */
-	ia64_set_kr(IA64_KR_IO_BASE, __pa(ia64_iobase));
-
 	ia64_mca_cmc_vector_setup();	/* Setup vector on AP & enable */
 
 #ifdef CONFIG_PERFMON
@@ -319,11 +307,6 @@ smp_callin (void)
 #endif
 
 	local_irq_enable();
-	calibrate_delay();
-	local_cpu_data->loops_per_jiffy = loops_per_jiffy;
-#ifdef CONFIG_IA32_SUPPORT
-	ia32_gdt_init();
-#endif
 
 	if (!(sal_platform_features & IA64_SAL_PLATFORM_FEATURE_ITC_DRIFT)) {
 		/*
@@ -336,6 +319,17 @@ smp_callin (void)
 		ia64_sync_itc(0);
 	}
 
+	/*
+	 * Get our bogomips.
+	 */
+	ia64_init_itm();
+	calibrate_delay();
+	local_cpu_data->loops_per_jiffy = loops_per_jiffy;
+
+#ifdef CONFIG_IA32_SUPPORT
+	ia32_gdt_init();
+#endif
+
 	/*
 	 * Allow the master to continue.
 	 */
@@ -352,6 +346,9 @@ start_secondary (void *unused)
 {
 	extern int cpu_idle (void);
 
+	/* Early console may use I/O ports */
+	ia64_set_kr(IA64_KR_IO_BASE, __pa(ia64_iobase));
+
 	Dprintk("start_secondary: starting CPU 0x%x\n", hard_smp_processor_id());
 	efi_map_pal_code();
 	cpu_init();

arch/ia64/kernel/traps.c

@@ -22,29 +22,6 @@
 #include <asm/processor.h>
 #include <asm/uaccess.h>
 
-/*
- * fp_emulate() needs to be able to access and update all floating point registers.  Those
- * saved in pt_regs can be accessed through that structure, but those not saved, will be
- * accessed directly.  To make this work, we need to ensure that the compiler does not end
- * up using a preserved floating point register on its own.  The following achieves this
- * by declaring preserved registers that are not marked as "fixed" as global register
- * variables.
- */
-#ifdef ASM_SUPPORTED
-register double f2 asm ("f2"); register double f3 asm ("f3");
-register double f4 asm ("f4"); register double f5 asm ("f5");
-
-register long f16 asm ("f16"); register long f17 asm ("f17");
-register long f18 asm ("f18"); register long f19 asm ("f19");
-register long f20 asm ("f20"); register long f21 asm ("f21");
-register long f22 asm ("f22"); register long f23 asm ("f23");
-
-register double f24 asm ("f24"); register double f25 asm ("f25");
-register double f26 asm ("f26"); register double f27 asm ("f27");
-register double f28 asm ("f28"); register double f29 asm ("f29");
-register double f30 asm ("f30"); register double f31 asm ("f31");
-#endif
-
 extern spinlock_t timerlist_lock;
 
 fpswa_interface_t *fpswa_interface;

arch/ia64/kernel/unwind.c

 /*
- * Copyright (C) 1999-2003 Hewlett-Packard Co
+ * Copyright (C) 1999-2004 Hewlett-Packard Co
  *	David Mosberger-Tang <davidm@hpl.hp.com>
  * Copyright (C) 2003 Fenghua Yu <fenghua.yu@intel.com>
  * 	- Change pt_regs_off() to make it less dependant on pt_regs structure.
@@ -89,6 +89,8 @@ static struct {
 	/* list of unwind tables (one per load-module) */
 	struct unw_table *tables;
 
+	unsigned long r0;			/* constant 0 for r0 */
+
 	/* table of registers that prologues can save (and order in which they're saved): */
 	const unsigned char save_order[8];
 
@@ -239,7 +241,11 @@ static struct {
 #endif
 };
 
-/* Unwind accessors.  */
+static inline int
+read_only (void *addr)
+{
+	return (unsigned long) ((char *) addr - (char *) &unw.r0) < sizeof(unw.r0);
+}
 
 /*
  * Returns offset of rREG in struct pt_regs.
@@ -274,6 +280,8 @@ get_scratch_regs (struct unw_frame_info *info)
 	return (struct pt_regs *) info->pt;
 }
 
+/* Unwind accessors.  */
+
 int
 unw_access_gr (struct unw_frame_info *info, int regnum, unsigned long *val, char *nat, int write)
 {
@@ -377,11 +385,15 @@ unw_access_gr (struct unw_frame_info *info, int regnum, unsigned long *val, char
 	}
 
 	if (write) {
+		if (read_only(addr))
+			UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n");
+		else {
 			*addr = *val;
 			if (*nat)
 				*nat_addr |= nat_mask;
 			else
 				*nat_addr &= ~nat_mask;
+		}
 	} else {
 		if ((*nat_addr & nat_mask) == 0) {
 			*val = *addr;
@@ -420,6 +432,9 @@ unw_access_br (struct unw_frame_info *info, int regnum, unsigned long *val, int
 		return -1;
 	}
 	if (write)
+		if (read_only(addr))
+			UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n");
+		else
 			*addr = *val;
 	else
 		*val = *addr;
@@ -465,6 +480,9 @@ unw_access_fr (struct unw_frame_info *info, int regnum, struct ia64_fpreg *val,
 	}
 
 	if (write)
+		if (read_only(addr))
+			UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n");
+		else
 			*addr = *val;
 	else
 		*val = *addr;
@@ -557,9 +575,12 @@ unw_access_ar (struct unw_frame_info *info, int regnum, unsigned long *val, int
 		return -1;
 	}
 
-	if (write)
-		*addr = *val;
+	if (write) {
+		if (read_only(addr))
+			UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n");
 		else
+			*addr = *val;
+	} else
 		*val = *addr;
 	return 0;
 }
@@ -574,9 +595,12 @@ unw_access_pr (struct unw_frame_info *info, unsigned long *val, int write)
 	if (!addr)
 		addr = &info->sw->pr;
 
-	if (write)
-		*addr = *val;
+	if (write) {
+		if (read_only(addr))
+			UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n");
 		else
+			*addr = *val;
+	} else
 		*val = *addr;
 	return 0;
 }
@@ -1407,6 +1431,9 @@ compile_reg (struct unw_state_record *sr, int i, struct unw_script *script)
 				need_nat_info = 0;
 			}
 			val = unw.preg_index[UNW_REG_R4 + (rval - 4)];
+		} else if (rval == 0) {
+			opc = UNW_INSN_MOVE_CONST;
+			val = 0;
 		} else {
 			/* register got spilled to a scratch register */
 			opc = UNW_INSN_MOVE_SCRATCH;
@@ -1729,6 +1756,17 @@ run_script (struct unw_script *script, struct unw_frame_info *state)
 			}
 			break;
 
+		      case UNW_INSN_MOVE_CONST:
+			if (val == 0)
+				s[dst] = (unsigned long) &unw.r0;
+			else {
+				s[dst] = 0;
+				UNW_DPRINT(0, "unwind.%s: UNW_INSN_MOVE_CONST bad val=%ld\n",
+					   __FUNCTION__, val);
+			}
+			break;
+
+
 		      case UNW_INSN_MOVE_STACKED:
 			s[dst] = (unsigned long) ia64_rse_skip_regs((unsigned long *)state->bsp,
 								    val);

arch/ia64/kernel/unwind_i.h

@@ -133,6 +133,7 @@ enum unw_insn_opcode {
 	UNW_INSN_SETNAT_TYPE,		/* s[dst+1].nat.type = val */
 	UNW_INSN_LOAD,			/* s[dst] = *s[val] */
 	UNW_INSN_MOVE_SCRATCH,		/* s[dst] = scratch reg "val" */
+	UNW_INSN_MOVE_CONST,            /* s[dst] = constant reg "val" */
 };
 
 struct unw_insn {

arch/ia64/mm/init.c

@@ -343,6 +343,7 @@ ia64_mmu_init (void *my_cpu_data)
 
 #ifdef	CONFIG_HUGETLB_PAGE
 	ia64_set_rr(HPAGE_REGION_BASE, HPAGE_SHIFT << 2);
+	ia64_srlz_d();
 #endif
 
 	cpu = smp_processor_id();

arch/ia64/scripts/check-serialize.S

+	.serialize.data
+	.serialize.instruction

arch/ia64/scripts/toolchain-flags

@@ -40,4 +40,14 @@ then
     CPPFLAGS="$CPPFLAGS -DHAVE_MODEL_SMALL_ATTRIBUTE"
 fi
 rm -f $out
+
+# Check whether assembler supports .serialize.{data,instruction} directive.
+
+$CC -c $dir/check-serialize.S -o $out 2>/dev/null
+res=$?
+rm -f $out
+if [ $res -eq 0 ]; then
+    CPPFLAGS="$CPPFLAGS -DHAVE_SERIALIZE_DIRECTIVE"
+fi
+
 echo $CPPFLAGS

arch/ia64/sn/io/io.c

@@ -651,7 +651,7 @@ hub_set_piomode(nasid_t nasid, int conveyor)
 	hubii_wcr_t ii_wcr;
 	int prbnum;
 
-	ASSERT(NASID_TO_COMPACT_NODEID(nasid) != INVALID_CNODEID);
+	ASSERT(nasid_to_cnodeid(nasid) != INVALID_CNODEID);
 
 	ii_iowa = REMOTE_HUB_L(nasid, IIO_OUTWIDGET_ACCESS);
 	REMOTE_HUB_S(nasid, IIO_OUTWIDGET_ACCESS, 0);

arch/ia64/sn/io/machvec/pci_bus_cvlink.c

@@ -328,6 +328,34 @@ sn_pci_fixup_slot(struct pci_dev *dev)
 			cmd |= PCI_COMMAND_MEMORY;
 	}
 
+        /*
+	 * Assign addresses to the ROMs, but don't enable them yet
+	 * Also note that we only map display card ROMs due to PIO mapping
+	 * space scarcity.
+	 */
+        if ((dev->class >> 16) == PCI_BASE_CLASS_DISPLAY) {
+                unsigned long addr;
+                size = dev->resource[PCI_ROM_RESOURCE].end -
+                        dev->resource[PCI_ROM_RESOURCE].start;
+
+                if (size) {
+                        addr = (unsigned long) pciio_pio_addr(vhdl, 0,
+					      PCIIO_SPACE_ROM,
+					      0, size, 0, PIOMAP_FIXED);
+                        if (!addr) {
+                                dev->resource[PCI_ROM_RESOURCE].start = 0;
+                                dev->resource[PCI_ROM_RESOURCE].end = 0;
+                                printk("sn_pci_fixup(): ROM pio map failure "
+				       "for %s\n", dev->slot_name);
+                        }
+                        addr |= __IA64_UNCACHED_OFFSET;
+                        dev->resource[PCI_ROM_RESOURCE].start = addr;
+                        dev->resource[PCI_ROM_RESOURCE].end = addr + size;
+                        if (dev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_MEM)
+                                cmd |= PCI_COMMAND_MEMORY;
+                }
+        }
+
 	/*
 	 * Update the Command Word on the Card.
 	 */
@@ -712,7 +740,7 @@ pci_bus_to_hcl_cvlink(void)
 			
 			/* Is this PCI bus associated with this moduleid? */
 			moduleid = NODE_MODULEID(
-				NASID_TO_COMPACT_NODEID(pcibr_soft->bs_nasid));
+				nasid_to_cnodeid(pcibr_soft->bs_nasid));
 			if (modules[i]->id == moduleid) {
 				struct pcibr_list_s *new_element;
 
@@ -791,8 +819,7 @@ sn_pci_init (void)
 	struct list_head *ln;
 	struct pci_bus *pci_bus = NULL;
 	struct pci_dev *pci_dev = NULL;
-	extern int numnodes;
-	int cnode, ret;
+	int ret;
 #ifdef CONFIG_PROC_FS
 	extern void register_sn_procfs(void);
 #endif

arch/ia64/sn/io/platform_init/sgi_io_init.c

@@ -26,7 +26,6 @@ extern vertex_hdl_t hwgraph_root;
 extern void io_module_init(void);
 extern int pci_bus_to_hcl_cvlink(void);
 
-cpuid_t master_procid;
 nasid_t console_nasid = (nasid_t) - 1;
 char master_baseio_wid;
 
@@ -47,10 +46,10 @@ per_hub_init(cnodeid_t cnode)
 	ii_ibcr_u_t ii_ibcr;
 	ii_ilcsr_u_t ii_ilcsr;
 
-	nasid = COMPACT_TO_NASID_NODEID(cnode);
+	nasid = cnodeid_to_nasid(cnode);
 
 	ASSERT(nasid != INVALID_NASID);
-	ASSERT(NASID_TO_COMPACT_NODEID(nasid) == cnode);
+	ASSERT(nasid_to_cnodeid(nasid) == cnode);
 
 	npdap = NODEPDA(cnode);
 
@@ -150,7 +149,7 @@ inline int
 check_nasid_equiv(nasid_t nasida, nasid_t nasidb)
 {
 	if ((nasida == nasidb)
-	    || (nasida == NODEPDA(NASID_TO_COMPACT_NODEID(nasidb))->xbow_peer))
+	    || (nasida == NODEPDA(nasid_to_cnodeid(nasidb))->xbow_peer))
 		return 1;
 	else
 		return 0;

arch/ia64/sn/io/sn2/klgraph.c

@@ -65,7 +65,7 @@ klhwg_add_disabled_cpu(vertex_hdl_t node_vertex, cnodeid_t cnode, klcpu_t *cpu,
         cpuid_t cpu_id;
 	nasid_t nasid;
 
-	nasid = COMPACT_TO_NASID_NODEID(cnode);
+	nasid = cnodeid_to_nasid(cnode);
         cpu_id = nasid_slice_to_cpuid(nasid, cpu->cpu_info.physid);
         if(cpu_id != -1){
 		snprintf(name, 120, "%s/%s/%c", EDGE_LBL_DISABLED, EDGE_LBL_CPU, 'a' + cpu->cpu_info.physid);
@@ -88,7 +88,7 @@ klhwg_add_cpu(vertex_hdl_t node_vertex, cnodeid_t cnode, klcpu_t *cpu)
         cpuid_t cpu_id;
 	nasid_t nasid;
 
-	nasid = COMPACT_TO_NASID_NODEID(cnode);
+	nasid = cnodeid_to_nasid(cnode);
         cpu_id = nasid_slice_to_cpuid(nasid, cpu->cpu_info.physid);
 
         snprintf(name, 120, "%s/%d/%c",
@@ -145,7 +145,7 @@ klhwg_add_xbow(cnodeid_t cnode, nasid_t nasid)
 			continue;
 		}
 
-		hub_cnode = NASID_TO_COMPACT_NODEID(hub_nasid);
+		hub_cnode = nasid_to_cnodeid(hub_nasid);
 
 		if (hub_cnode == INVALID_CNODEID) {
 			continue;
@@ -178,7 +178,7 @@ klhwg_add_xbow(cnodeid_t cnode, nasid_t nasid)
 		 */
 		if (hub_nasid != nasid) {
 			NODEPDA(hub_cnode)->xbow_peer = nasid;
-			NODEPDA(NASID_TO_COMPACT_NODEID(nasid))->xbow_peer =
+			NODEPDA(nasid_to_cnodeid(nasid))->xbow_peer =
 				hub_nasid;
 		}
 	}
@@ -200,7 +200,7 @@ klhwg_add_node(vertex_hdl_t hwgraph_root, cnodeid_t cnode)
 	klcpu_t *cpu;
 	vertex_hdl_t cpu_dir;
 
-	nasid = COMPACT_TO_NASID_NODEID(cnode);
+	nasid = cnodeid_to_nasid(cnode);
 	brd = find_lboard_any((lboard_t *)KL_CONFIG_INFO(nasid), KLTYPE_SNIA);
 	ASSERT(brd);
 
@@ -280,7 +280,7 @@ klhwg_add_all_routers(vertex_hdl_t hwgraph_root)
 	int rv;
 
 	for (cnode = 0; cnode < numnodes; cnode++) {
-		nasid = COMPACT_TO_NASID_NODEID(cnode);
+		nasid = cnodeid_to_nasid(cnode);
 		brd = find_lboard_class_any((lboard_t *)KL_CONFIG_INFO(nasid),
 				KLTYPE_ROUTER);
 
@@ -363,7 +363,7 @@ klhwg_connect_one_router(vertex_hdl_t hwgraph_root, lboard_t *brd,
 				 port));
 			continue;
 		}
-		if (NASID_TO_COMPACT_NODEID(router->rou_port[port].port_nasid) 
+		if (nasid_to_cnodeid(router->rou_port[port].port_nasid) 
 		    == INVALID_CNODEID) {
 			continue;
 		}
@@ -414,7 +414,7 @@ klhwg_connect_routers(vertex_hdl_t hwgraph_root)
 	lboard_t *brd;
 
 	for (cnode = 0; cnode < numnodes; cnode++) {
-		nasid = COMPACT_TO_NASID_NODEID(cnode);
+		nasid = cnodeid_to_nasid(cnode);
 		brd = find_lboard_class_any((lboard_t *)KL_CONFIG_INFO(nasid),
 				KLTYPE_ROUTER);
 
@@ -423,7 +423,7 @@ klhwg_connect_routers(vertex_hdl_t hwgraph_root)
 
 		do {
 
-			nasid = COMPACT_TO_NASID_NODEID(cnode);
+			nasid = cnodeid_to_nasid(cnode);
 
 			klhwg_connect_one_router(hwgraph_root, brd,
 						 cnode, nasid);
@@ -451,7 +451,7 @@ klhwg_connect_hubs(vertex_hdl_t hwgraph_root)
 	int port;
 
 	for (cnode = 0; cnode < numionodes; cnode++) {
-		nasid = COMPACT_TO_NASID_NODEID(cnode);
+		nasid = cnodeid_to_nasid(cnode);
 
 		brd = find_lboard_any((lboard_t *)KL_CONFIG_INFO(nasid), KLTYPE_SNIA);
 
@@ -463,7 +463,7 @@ klhwg_connect_hubs(vertex_hdl_t hwgraph_root)
 				continue; /* Port not active */
 			}
 
-			if (NASID_TO_COMPACT_NODEID(hub->hub_port[port].port_nasid) == INVALID_CNODEID)
+			if (nasid_to_cnodeid(hub->hub_port[port].port_nasid) == INVALID_CNODEID)
 				continue;
 
 			/* Generate a hardware graph path for this board. */

arch/ia64/sn/io/sn2/ml_SN_init.c

@@ -31,13 +31,9 @@ void init_platform_nodepda(nodepda_t *npda, cnodeid_t node)
 	hubinfo_t hubinfo;
 	nasid_t nasid;
 
-	extern void router_map_init(nodepda_t *);
-	extern void router_queue_init(nodepda_t *,cnodeid_t);
-	extern void intr_init_vecblk(nodepda_t *, cnodeid_t, int);
-
 	/* Allocate per-node platform-dependent data */
 	
-	nasid = COMPACT_TO_NASID_NODEID(node);
+	nasid = cnodeid_to_nasid(node);
 	if (node >= numnodes) /* Headless/memless IO nodes */
 		hubinfo = (hubinfo_t)alloc_bootmem_node(NODE_DATA(0), sizeof(struct hubinfo_s));
 	else
@@ -81,7 +77,7 @@ init_platform_hubinfo(nodepda_t **nodepdaindr)
 	for (cnode = 0; cnode < numionodes; cnode++) {
 		npda = nodepdaindr[cnode];
 		hubinfo = (hubinfo_t)npda->pdinfo;
-		hubinfo->h_nasid = COMPACT_TO_NASID_NODEID(cnode);
+		hubinfo->h_nasid = cnodeid_to_nasid(cnode);
 		hubinfo->h_widgetid = hub_widget_id(hubinfo->h_nasid);
 	}
 }

arch/ia64/sn/io/sn2/ml_SN_intr.c

@@ -12,6 +12,7 @@
 #include <asm/smp.h>
 #include <asm/irq.h>
 #include <asm/hw_irq.h>
+#include <asm/topology.h>
 #include <asm/sn/sgi.h>
 #include <asm/sn/iograph.h>
 #include <asm/sn/hcl.h>
@@ -43,7 +44,6 @@ void intr_init_vecblk(cnodeid_t node)
 	sh_ii_int0_config_u_t		ii_int_config;
 	cpuid_t				cpu;
 	cpuid_t				cpu0, cpu1;
-	nodepda_t			*lnodepda;
 	sh_ii_int0_enable_u_t		ii_int_enable;
 	sh_int_node_id_config_u_t	node_id_config;
 	sh_local_int5_config_u_t	local5_config;
@@ -60,15 +60,13 @@ void intr_init_vecblk(cnodeid_t node)
 		HUB_S((unsigned long *)GLOBAL_MMR_ADDR(nasid, SH_INT_NODE_ID_CONFIG),
 			node_id_config.sh_int_node_id_config_regval);
 		cnode = nasid_to_cnodeid(master_nasid);
-		lnodepda = NODEPDA(cnode);
-		cpu = lnodepda->node_first_cpu;
+		cpu = first_cpu(node_to_cpumask(cnode));
 		cpu = cpu_physical_id(cpu);
 		SAL_CALL(ret_stuff, SN_SAL_REGISTER_CE, nasid, cpu, master_nasid,0,0,0,0);
 		if (ret_stuff.status < 0)
 			printk("%s: SN_SAL_REGISTER_CE SAL_CALL failed\n",__FUNCTION__);
 	} else {
-		lnodepda = NODEPDA(node);
-		cpu = lnodepda->node_first_cpu;
+		cpu = first_cpu(node_to_cpumask(node));
 		cpu = cpu_physical_id(cpu);
 	}
 

arch/ia64/sn/io/sn2/ml_iograph.c

@@ -313,7 +313,7 @@ io_xswitch_widget_init(vertex_hdl_t  	xswitchv,
 
 	hubinfo_get(hubv, &hubinfo);
 	nasid = hubinfo->h_nasid;
-	cnode = NASID_TO_COMPACT_NODEID(nasid);
+	cnode = nasid_to_cnodeid(nasid);
 	hub_widgetid = hubinfo->h_widgetid;
 
 	/*
@@ -567,7 +567,7 @@ io_init_node(cnodeid_t cnodeid)
 	 * and hence widget id is Not 0.
 	 */
 	widget_partnum = (((*(volatile int32_t *)(NODE_SWIN_BASE
-			(COMPACT_TO_NASID_NODEID(cnodeid), 0) + 
+			(cnodeid_to_nasid(cnodeid), 0) + 
 			WIDGET_ID))) & WIDGET_PART_NUM) 
 			>> WIDGET_PART_NUM_SHFT;
 
@@ -584,7 +584,7 @@ io_init_node(cnodeid_t cnodeid)
 	} else {
 		void	*bridge;
 
-		bridge = (void *)NODE_SWIN_BASE(COMPACT_TO_NASID_NODEID(cnodeid), 0);
+		bridge = (void *)NODE_SWIN_BASE(cnodeid_to_nasid(cnodeid), 0);
 		npdap->basew_id = pcireg_bridge_control_get(bridge) & WIDGET_WIDGET_ID;
 
 		printk(" ****io_init_node: Unknown Widget Part Number 0x%x Widget ID 0x%x attached to Hubv 0x%p ****\n", widget_partnum, npdap->basew_id, (void *)hubv);
@@ -637,7 +637,7 @@ io_init_node(cnodeid_t cnodeid)
 
 		/* If there's someone else on this crossbow, recognize him */
 		if (npdap->xbow_peer != INVALID_NASID) {
-			nodepda_t *peer_npdap = NODEPDA(NASID_TO_COMPACT_NODEID(npdap->xbow_peer));
+			nodepda_t *peer_npdap = NODEPDA(nasid_to_cnodeid(npdap->xbow_peer));
 			peer_sema = &peer_npdap->xbow_sema;
 			volunteer_for_widgets(switchv, peer_npdap->node_vertex);
 		}

arch/ia64/sn/io/sn2/module.c

@@ -196,7 +196,7 @@ io_module_init(void)
      * We do not support memoryless compute nodes.
      */
     for (node = 0; node < numnodes; node++) {
-	nasid = COMPACT_TO_NASID_NODEID(node);
+	nasid = cnodeid_to_nasid(node);
 	board = find_lboard_nasid((lboard_t *) KL_CONFIG_INFO(nasid), nasid, KLTYPE_SNIA);
 	ASSERT(board);
 
@@ -214,7 +214,7 @@ io_module_init(void)
 	nasid_t		nasid;
 	char		serial_number[16];
 
-        nasid = COMPACT_TO_NASID_NODEID(node);
+        nasid = cnodeid_to_nasid(node);
 	board = find_lboard_nasid((lboard_t *) KL_CONFIG_INFO(nasid), 
 				nasid, KLTYPE_SNIA);
 	ASSERT(board);

arch/ia64/sn/io/sn2/pcibr/pcibr_dvr.c

@@ -758,7 +758,7 @@ pcibr_directmap_init(pcibr_soft_t pcibr_soft)
     cnodeid_t		cnodeid = 0;	/* We need api for diroff api */
     nasid_t		nasid;
 
-    nasid = COMPACT_TO_NASID_NODEID(cnodeid);
+    nasid = cnodeid_to_nasid(cnodeid);
     paddr = NODE_OFFSET(nasid) + 0;
 
     /* Assume that if we ask for a DMA mapping to zero the XIO host will
@@ -2092,7 +2092,7 @@ pcibr_get_dmatrans_node(vertex_hdl_t pconn_vhdl)
 	pciio_info_t	pciio_info = pciio_info_get(pconn_vhdl);
 	pcibr_soft_t	pcibr_soft = (pcibr_soft_t) pciio_info_mfast_get(pciio_info);
 
-	return NASID_TO_COMPACT_NODEID(NASID_GET(pcibr_soft->bs_dir_xbase));
+	return nasid_to_cnodeid(NASID_GET(pcibr_soft->bs_dir_xbase));
 }
 
 /*ARGSUSED */
@@ -2645,7 +2645,7 @@ isIO9(nasid_t nasid)
                         brd = KLCF_NEXT(brd);
 	}
 	/* if it's dual ported, check the peer also */
-	nasid = NODEPDA(NASID_TO_COMPACT_NODEID(nasid))->xbow_peer;
+	nasid = NODEPDA(nasid_to_cnodeid(nasid))->xbow_peer;
 	if (nasid < 0) return 0;
 	brd = (lboard_t *)KL_CONFIG_INFO(nasid);
 	while (brd) {

arch/ia64/sn/io/sn2/pic.c

@@ -124,7 +124,7 @@ pic_bus1_widget_info_dup(vertex_hdl_t conn_v, vertex_hdl_t peer_conn_v,
 static vertex_hdl_t
 pic_bus1_redist(nasid_t nasid, vertex_hdl_t conn_v)
 {
-	cnodeid_t cnode = NASID_TO_COMPACT_NODEID(nasid);
+	cnodeid_t cnode = nasid_to_cnodeid(nasid);
 	cnodeid_t xbow_peer = -1;
 	char pathname[256], peer_path[256], tmpbuf[256];
 	char *p;
@@ -137,7 +137,7 @@ pic_bus1_redist(nasid_t nasid, vertex_hdl_t conn_v)
 		/* create a path for this widget on the peer Cbrick */
 		/* pcibr widget hw/module/001c11/slab/0/Pbrick/xtalk/12 */
 		/* sprintf(pathname, "%v", conn_v); */
-		xbow_peer = NASID_TO_COMPACT_NODEID(NODEPDA(cnode)->xbow_peer);
+		xbow_peer = nasid_to_cnodeid(NODEPDA(cnode)->xbow_peer);
 		pos = hwgfs_generate_path(conn_v, tmpbuf, 256);
 		strcpy(pathname, &tmpbuf[pos]);
 		p = pathname + strlen("hw/module/001c01/slab/0/");

arch/ia64/sn/io/sn2/shuberror.c

@@ -157,7 +157,7 @@ hubii_eint_init(cnodeid_t cnode)
     hubio_eint.ii_iidsr_regval = 0;
     hubio_eint.ii_iidsr_fld_s.i_enable = 1;
     hubio_eint.ii_iidsr_fld_s.i_level = bit;/* Take the least significant bits*/
-    hubio_eint.ii_iidsr_fld_s.i_node = COMPACT_TO_NASID_NODEID(cnode);
+    hubio_eint.ii_iidsr_fld_s.i_node = cnodeid_to_nasid(cnode);
     hubio_eint.ii_iidsr_fld_s.i_pi_id = cpuid_to_subnode(intr_cpu);
     REMOTE_HUB_S(hinfo->h_nasid, IIO_IIDSR, hubio_eint.ii_iidsr_regval);
 
@@ -450,7 +450,7 @@ hubiio_crb_error_handler(vertex_hdl_t hub_v, hubinfo_t hinfo)
 	int		rc;
 
 	nasid = hinfo->h_nasid;
-	cnode = NASID_TO_COMPACT_NODEID(nasid);
+	cnode = nasid_to_cnodeid(nasid);
 
 	/*
 	 * XXX - Add locking for any recovery actions

arch/ia64/sn/io/sn2/shubio.c

@@ -76,7 +76,7 @@ hub_xp_error_handler(
 	/* Get the error state of the hub */
 	e_state = error_state_get(hub_v);
 
-	cnode = NASID_TO_COMPACT_NODEID(nasid);
+	cnode = nasid_to_cnodeid(nasid);
 
 	xswitch = NODEPDA(cnode)->basew_xc;
 
@@ -119,7 +119,7 @@ is_widget_pio_enabled(ioerror_t *ioerror)
 		return(0);
 
 	/* Get the nasid for the cnode */
-	src_nasid = COMPACT_TO_NASID_NODEID(src_node);
+	src_nasid = cnodeid_to_nasid(src_node);
 	if (src_nasid == INVALID_NASID)
 		return(0);
 

arch/ia64/sn/kernel/setup.c

@@ -451,10 +451,6 @@ sn_cpu_init(void)
 	}
 	pda->shub_1_1_found = shub_1_1_found;
 	
-	if (local_node_data->active_cpu_count == 1)
-		nodepda->node_first_cpu = cpuid;
-
-
 
 	/*
 	 * We must use different memory allocators for first cpu (bootmem 
@@ -474,7 +470,7 @@ sn_cpu_init(void)
 	pda->mem_write_status_addr = (volatile u64 *)
 			LOCAL_MMR_ADDR((slice < 2 ? SH_MEMORY_WRITE_STATUS_0 : SH_MEMORY_WRITE_STATUS_1 ) );
 
-	if (nodepda->node_first_cpu == cpuid) {
+	if (local_node_data->active_cpu_count++ == 0) {
 		int	buddy_nasid;
 		buddy_nasid = cnodeid_to_nasid(numa_node_id() == numnodes-1 ? 0 : numa_node_id()+ 1);
 		pda->pio_shub_war_cam_addr = (volatile unsigned long*)GLOBAL_MMR_ADDR(nasid, SH_PI_CAM_CONTROL);

arch/ia64/sn/kernel/sn2/prominfo_proc.c

@@ -228,11 +228,9 @@ read_version_entry(char *page, char **start, off_t off, int count, int *eof,
 {
 	int len = 0;
 
-	MOD_INC_USE_COUNT;
 	/* data holds the pointer to this node's FIT */
 	len = dump_version(page, (unsigned long *)data);
 	len = proc_calc_metrics(page, start, off, count, eof, len);
-	MOD_DEC_USE_COUNT;
 	return len;
 }
 
@@ -242,11 +240,9 @@ read_fit_entry(char *page, char **start, off_t off, int count, int *eof,
 {
 	int len = 0;
 
-	MOD_INC_USE_COUNT;
 	/* data holds the pointer to this node's FIT */
 	len = dump_fit(page, (unsigned long *)data);
 	len = proc_calc_metrics(page, start, off, count, eof, len);
-	MOD_DEC_USE_COUNT;
 
 	return len;
 }
@@ -310,6 +306,7 @@ int __init
 prominfo_init(void)
 {
 	struct proc_dir_entry **entp;
+	struct proc_dir_entry *p;
 	cnodeid_t cnodeid;
 	nasid_t nasid;
 	char name[NODE_NAME_LEN];
@@ -333,12 +330,16 @@ prominfo_init(void)
 		sprintf(name, "node%d", cnodeid);
 		*entp = proc_mkdir(name, sgi_prominfo_entry);
 		nasid = cnodeid_to_nasid(cnodeid);
-		create_proc_read_entry(
+		p = create_proc_read_entry(
 			"fit", 0, *entp, read_fit_entry,
 			lookup_fit(nasid));
-		create_proc_read_entry(
+		if (p)
+			p->owner = THIS_MODULE;
+		p = create_proc_read_entry(
 			"version", 0, *entp, read_version_entry,
 			lookup_fit(nasid));
+		if (p)
+			p->owner = THIS_MODULE;
 	}
 
 	return 0;

drivers/char/sn_serial.c

@@ -502,7 +502,7 @@ sn_sal_connect_interrupt(void)
 	int result;
 
 	console_nasid = ia64_sn_get_console_nasid();
-	intr_cpuid = NODEPDA(NASID_TO_COMPACT_NODEID(console_nasid))->node_first_cpu;
+	intr_cpuid = first_cpu(node_to_cpumask(nasid_to_cnodeid(console_nasid)));
 	intr_cpuloc = cpu_physical_id(intr_cpuid);
 	console_irq = CPU_VECTOR_TO_IRQ(intr_cpuloc, SGI_UART_VECTOR);
 

include/asm-ia64/asmmacro.h

@@ -100,4 +100,12 @@
 # define TEXT_ALIGN(n)
 #endif
 
+#ifdef HAVE_SERIALIZE_DIRECTIVE
+# define dv_serialize_data		.serialize.data
+# define dv_serialize_instruction	.serialize.instruction
+#else
+# define dv_serialize_data
+# define dv_serialize_instruction
+#endif
+
 #endif /* _ASM_IA64_ASMMACRO_H */

include/asm-ia64/delay.h

@@ -44,6 +44,12 @@ ia64_set_itv (unsigned long val)
 	ia64_srlz_d();
 }
 
+static __inline__ unsigned long
+ia64_get_itv (void)
+{
+	return ia64_getreg(_IA64_REG_CR_ITV);
+}
+
 static __inline__ void
 ia64_set_itc (unsigned long val)
 {

include/asm-ia64/gcc_intrin.h

@@ -4,9 +4,10 @@
  *
  * Copyright (C) 2002,2003 Jun Nakajima <jun.nakajima@intel.com>
  * Copyright (C) 2002,2003 Suresh Siddha <suresh.b.siddha@intel.com>
- *
  */
 
+#include <linux/compiler.h>
+
 /* define this macro to get some asm stmts included in 'c' files */
 #define ASM_SUPPORTED
 
@@ -23,7 +24,7 @@
 extern void ia64_bad_param_for_setreg (void);
 extern void ia64_bad_param_for_getreg (void);
 
-register unsigned long ia64_r13 asm ("r13");
+register unsigned long ia64_r13 asm ("r13") __attribute_used__;
 
 #define ia64_setreg(regnum, val)						\
 ({										\
@@ -377,9 +378,16 @@ register unsigned long ia64_r13 asm ("r13");
 })
 
 #define ia64_srlz_i()	asm volatile (";; srlz.i ;;" ::: "memory")
-
 #define ia64_srlz_d()	asm volatile (";; srlz.d" ::: "memory");
 
+#ifdef HAVE_SERIALIZE_DIRECTIVE
+# define ia64_dv_serialize_data()		asm volatile (".serialize.data");
+# define ia64_dv_serialize_instruction()	asm volatile (".serialize.instruction");
+#else
+# define ia64_dv_serialize_data()
+# define ia64_dv_serialize_instruction()
+#endif
+
 #define ia64_nop(x)	asm volatile ("nop %0"::"i"(x));
 
 #define ia64_itci(addr)	asm volatile ("itc.i %0;;" :: "r"(addr) : "memory")

include/asm-ia64/intel_intrin.h

@@ -204,6 +204,9 @@ __s64 _m64_popcnt(__s64 a);
 #define ia64_srlz_d		__dsrlz
 #define ia64_srlz_i		__isrlz
 
+#define ia64_dv_serialize_data()
+#define ia64_dv_serialize_instruction()
+
 #define ia64_st1_rel		__st1_rel
 #define ia64_st2_rel		__st2_rel
 #define ia64_st4_rel		__st4_rel

include/asm-ia64/sn/arch.h

@@ -20,12 +20,6 @@ typedef u64	hubreg_t;
 typedef u64	mmr_t;
 typedef u64	nic_t;
 
-#define CNODE_TO_CPU_BASE(_cnode)	(NODEPDA(_cnode)->node_first_cpu)
-
-#define NASID_TO_COMPACT_NODEID(nasid)  (nasid_to_cnodeid(nasid))
-#define COMPACT_TO_NASID_NODEID(cnode)  (cnodeid_to_nasid(cnode))
-
-
 #define INVALID_NASID		((nasid_t)-1)
 #define INVALID_CNODEID		((cnodeid_t)-1)
 #define INVALID_PNODEID		((pnodeid_t)-1)
@@ -34,9 +28,7 @@ typedef u64	nic_t;
 #define	INVALID_PARTID		((partid_t)-1)
 
 extern cpuid_t cnodetocpu(cnodeid_t);
-void   sn_flush_all_caches(long addr, long bytes);
-
+extern void sn_flush_all_caches(long addr, long bytes);
 extern int is_fine_dirmode(void);
 
-
 #endif /* _ASM_IA64_SN_ARCH_H */

include/asm-ia64/sn/intr.h

@@ -8,9 +8,11 @@
 #ifndef _ASM_IA64_SN_INTR_H
 #define _ASM_IA64_SN_INTR_H
 
+#include <asm/sn/types.h>
 #include <asm/sn/sn2/intr.h>
 
 extern void sn_send_IPI_phys(long, int, int);
+extern void intr_init_vecblk(cnodeid_t node);
 
 #define CPU_VECTOR_TO_IRQ(cpuid,vector) (vector)
 #define SN_CPU_FROM_IRQ(irq)	(0)

include/asm-ia64/sn/module.h

@@ -184,10 +184,6 @@ extern module_t	       *modules[MODULE_MAX];	/* Indexed by cmoduleid_t   */
 extern int		nummodules;
 
 extern module_t	       *module_lookup(moduleid_t id);
-
-extern int		get_kmod_sys_snum(cmoduleid_t cmod,
-					  char *snum);
-
 extern void		format_module_id(char *buffer, moduleid_t m, int fmt);
 extern int		parse_module_id(char *buffer);
 

include/asm-ia64/sn/nodepda.h

@@ -36,13 +36,6 @@
 
 
 struct nodepda_s {
-
-
-	cpuid_t         node_first_cpu; /* Starting cpu number for node */
-					/* WARNING: no guarantee that   */
-					/*  the second cpu on a node is */
-					/*  node_first_cpu+1.           */
-
 	vertex_hdl_t 	xbow_vhdl;
 	nasid_t		xbow_peer;	/* NASID of our peer hub on xbow */
 	struct semaphore xbow_sema;	/* Sema for xbow synchronization */

include/asm-ia64/sn/sn2/sn_private.h

@@ -233,7 +233,7 @@ extern void sysctlr_keepalive(void);
  * address.
  */
 #define paddr_dimm(_pa)		((_pa & MD_BANK_MASK) >> MD_BANK_SHFT)
-#define paddr_cnode(_pa)	(NASID_TO_COMPACT_NODEID(NASID_GET(_pa)))
+#define paddr_cnode(_pa)	(nasid_to_cnodeid(NASID_GET(_pa)))
 extern void membank_pathname_get(paddr_t, char *);
 
 extern void crbx(nasid_t nasid, void (*pf) (char *, ...));

include/asm-ia64/sn/types.h

@@ -15,11 +15,9 @@ typedef signed char	partid_t;	/* partition ID type */
 typedef unsigned int    moduleid_t;     /* user-visible module number type */
 typedef unsigned int    cmoduleid_t;    /* kernel compact module id type */
 typedef signed char     slabid_t;
-typedef unsigned char	clusterid_t;	/* Clusterid of the cell */
 
 typedef unsigned long iopaddr_t;
 typedef unsigned long paddr_t;
-typedef unsigned long pfn_t;
 typedef short cnodeid_t;
 
 #endif /* _ASM_IA64_SN_TYPES_H */

include/asm-ia64/system.h

@@ -114,10 +114,16 @@ extern struct ia64_boot_param {
  */
 /* For spinlocks etc */
 
-/* clearing psr.i is implicitly serialized (visible by next insn) */
-/* setting psr.i requires data serialization */
+/*
+ * - clearing psr.i is implicitly serialized (visible by next insn)
+ * - setting psr.i requires data serialization
+ * - we need a stop-bit before reading PSR because we sometimes
+ *   write a floating-point register right before reading the PSR
+ *   and that writes to PSR.mfl
+ */
 #define __local_irq_save(x)			\
 do {						\
+	ia64_stop();				\
 	(x) = ia64_getreg(_IA64_REG_PSR);	\
 	ia64_stop();				\
 	ia64_rsm(IA64_PSR_I);			\
@@ -166,7 +172,7 @@ do {								\
 #endif /* !CONFIG_IA64_DEBUG_IRQ */
 
 #define local_irq_enable()	({ ia64_ssm(IA64_PSR_I); ia64_srlz_d(); })
-#define local_save_flags(flags)	((flags) = ia64_getreg(_IA64_REG_PSR))
+#define local_save_flags(flags)	({ ia64_stop(); (flags) = ia64_getreg(_IA64_REG_PSR); })
 
 #define irqs_disabled()				\
 ({						\

include/asm-ia64/unistd.h

@@ -4,7 +4,7 @@
 /*
  * IA-64 Linux syscall numbers and inline-functions.
  *
- * Copyright (C) 1998-2003 Hewlett-Packard Co
+ * Copyright (C) 1998-2004 Hewlett-Packard Co
  *	David Mosberger-Tang <davidm@hpl.hp.com>
  */
 
@@ -304,10 +304,10 @@ lseek (int fd, off_t off, int whence)
 	return sys_lseek(fd, off, whence);
 }
 
-static inline long
+static inline void
 _exit (int value)
 {
-	return sys_exit(value);
+	sys_exit(value);
 }
 
 #define exit(x) _exit(x)
@@ -373,7 +373,7 @@ asmlinkage long sys_rt_sigaction(int sig,
  * proper prototype, but we can't use __typeof__ either, because not all cond_syscall()
  * declarations have prototypes at the moment.
  */
-#define cond_syscall(x) asmlinkage long x() __attribute__((weak,alias("sys_ni_syscall")));
+#define cond_syscall(x) asmlinkage long x (void) __attribute__((weak,alias("sys_ni_syscall")));
 
 #endif /* !__ASSEMBLY__ */
 #endif /* __KERNEL__ */