ia64: Manual merge.

arch/ia64/Makefile

@@ -26,7 +26,7 @@ CFLAGS_KERNEL := -mconstant-gp
 GCC_VERSION=$(shell $(CC) -v 2>&1 | fgrep 'gcc version' | cut -f3 -d' ' | cut -f1 -d'.')
 
 ifneq ($(GCC_VERSION),2)
-	CFLAGS += -frename-registers --param max-inline-insns=2000
+	CFLAGS += -frename-registers --param max-inline-insns=5000
 endif
 
 ifeq ($(CONFIG_ITANIUM_BSTEP_SPECIFIC),y)

arch/ia64/config.in

@@ -64,7 +64,8 @@ if [ "$CONFIG_MCKINLEY" = "y" ]; then
 	fi
 fi
 
-if [ "$CONFIG_IA64_GENERIC" = "y" -o  "$CONFIG_IA64_DIG" = "y" -o "$CONFIG_IA64_HP_ZX1" = "y" ]; then
+if [ "$CONFIG_IA64_GENERIC" = "y" -o "$CONFIG_IA64_DIG" = "y" -o "$CONFIG_IA64_HP_ZX1" = "y" ];
+then
 	bool '  Enable IA-64 Machine Check Abort' CONFIG_IA64_MCA
 	define_bool CONFIG_PM y
 fi
@@ -99,21 +100,21 @@ tristate 'Kernel support for ELF binaries' CONFIG_BINFMT_ELF
 tristate 'Kernel support for MISC binaries' CONFIG_BINFMT_MISC
 
 if [ "$CONFIG_IA64_HP_SIM" = "n" ]; then
+  source drivers/acpi/Config.in
 
-source drivers/acpi/Config.in
 
-bool 'PCI support' CONFIG_PCI
-source drivers/pci/Config.in
+  bool 'PCI support' CONFIG_PCI
+  source drivers/pci/Config.in
 
-bool 'Support for hot-pluggable devices' CONFIG_HOTPLUG
-if [ "$CONFIG_HOTPLUG" = "y" ]; then
+  bool 'Support for hot-pluggable devices' CONFIG_HOTPLUG
+  if [ "$CONFIG_HOTPLUG" = "y" ]; then
+     source drivers/hotplug/Config.in
      source drivers/pcmcia/Config.in
-else
+  else
      define_bool CONFIG_PCMCIA n
-fi
-
-source drivers/parport/Config.in
+  fi
 
+  source drivers/parport/Config.in
 fi # !HP_SIM
 
 endmenu
@@ -124,38 +125,17 @@ fi
 
 if [ "$CONFIG_IA64_HP_SIM" = "n" ]; then
 
-source drivers/mtd/Config.in
-source drivers/pnp/Config.in
-source drivers/block/Config.in
-source drivers/ieee1394/Config.in
-source drivers/message/i2o/Config.in
-source drivers/md/Config.in
-source drivers/message/fusion/Config.in
 
-mainmenu_option next_comment
-comment 'ATA/ATAPI/MFM/RLL support'
 
-tristate 'ATA/ATAPI/MFM/RLL support' CONFIG_IDE
 
-if [ "$CONFIG_IDE" != "n" ]; then
-  source drivers/ide/Config.in
-else
-  define_bool CONFIG_BLK_DEV_HD n
-fi
-endmenu
 
-else # ! HP_SIM
-mainmenu_option next_comment
-comment 'Block devices'
-tristate 'Loopback device support' CONFIG_BLK_DEV_LOOP
-dep_tristate 'Network block device support' CONFIG_BLK_DEV_NBD $CONFIG_NET
 
-tristate 'RAM disk support' CONFIG_BLK_DEV_RAM
-if [ "$CONFIG_BLK_DEV_RAM" = "y" -o "$CONFIG_BLK_DEV_RAM" = "m" ]; then
-   int '  Default RAM disk size' CONFIG_BLK_DEV_RAM_SIZE 4096
-fi
-endmenu
-fi # !HP_SIM
+
+
+
+
+
+
 
 mainmenu_option next_comment
 comment 'SCSI support'
@@ -168,8 +148,7 @@ fi
 endmenu
 
 if [ "$CONFIG_IA64_HP_SIM" = "n" ]; then
-
-if [ "$CONFIG_NET" = "y" ]; then
+  if [ "$CONFIG_NET" = "y" ]; then
     mainmenu_option next_comment
     comment 'Network device support'
 
@@ -178,21 +157,17 @@ if [ "$CONFIG_NET" = "y" ]; then
       source drivers/net/Config.in
     fi
     endmenu
-fi
 
-source net/ax25/Config.in
 
-source drivers/isdn/Config.in
 
-mainmenu_option next_comment
-comment 'CD-ROM drivers (not for SCSI or IDE/ATAPI drives)'
 
-bool 'Support non-SCSI/IDE/ATAPI CDROM drives' CONFIG_CD_NO_IDESCSI
-if [ "$CONFIG_CD_NO_IDESCSI" != "n" ]; then
-  source drivers/cdrom/Config.in
-fi
-endmenu
 
+
+
+
+
+  fi
+bool 'Support non-SCSI/IDE/ATAPI CDROM drives' CONFIG_CD_NO_IDESCSI
 fi # !HP_SIM
 
 #
@@ -220,21 +195,18 @@ fi
 
 if [ "$CONFIG_IA64_HP_SIM" = "n" ]; then
 
-mainmenu_option next_comment
-comment 'Sound'
+  mainmenu_option next_comment
+  comment 'Sound'
 
-tristate 'Sound card support' CONFIG_SOUND
-if [ "$CONFIG_SOUND" != "n" ]; then
+  tristate 'Sound card support' CONFIG_SOUND
+  if [ "$CONFIG_SOUND" != "n" ]; then
     source sound/Config.in
-fi
-endmenu
-
-source drivers/usb/Config.in
-
-source lib/Config.in
-
-source net/bluetooth/Config.in
+  fi
+  endmenu
 
+  source drivers/usb/Config.in
+  source lib/Config.in
+  source net/bluetooth/Config.in
 fi # !HP_SIM
 
 if [ "$CONFIG_IA64_HP_SIM" != "n" -o "$CONFIG_IA64_GENERIC" != "n" ]; then

arch/ia64/hp/sim/hpsim_console.c

@@ -30,12 +30,12 @@ static void simcons_write (struct console *, const char *, unsigned);
 static kdev_t simcons_console_device (struct console *);
 
 struct console hpsim_cons = {
-	name:		"simcons",
-	write:		simcons_write,
-	device:		simcons_console_device,
-	setup:		simcons_init,
-	flags:		CON_PRINTBUFFER,
-	index:		-1,
+	.name =		"simcons",
+	.write =	simcons_write,
+	.device =	simcons_console_device,
+	.setup =	simcons_init,
+	.flags =	CON_PRINTBUFFER,
+	.index =	-1,
 };
 
 static int

arch/ia64/hp/sim/hpsim_irq.c

@@ -22,14 +22,14 @@ hpsim_irq_noop (unsigned int irq)
 }
 
 static struct hw_interrupt_type irq_type_hp_sim = {
-	typename:	"hpsim",
-	startup:	hpsim_irq_startup,
-	shutdown:	hpsim_irq_noop,
-	enable:		hpsim_irq_noop,
-	disable:	hpsim_irq_noop,
-	ack:		hpsim_irq_noop,
-	end:		hpsim_irq_noop,
-	set_affinity:	(void (*)(unsigned int, unsigned long)) hpsim_irq_noop,
+	.typename =	"hpsim",
+	.startup =	hpsim_irq_startup,
+	.shutdown =	hpsim_irq_noop,
+	.enable =	hpsim_irq_noop,
+	.disable =	hpsim_irq_noop,
+	.ack =		hpsim_irq_noop,
+	.end =		hpsim_irq_noop,
+	.set_affinity =	(void (*)(unsigned int, unsigned long)) hpsim_irq_noop,
 };
 
 void __init

arch/ia64/hp/sim/simserial.c

@@ -31,6 +31,7 @@
 #include <linux/serialP.h>
 
 #include <asm/irq.h>
+#include <asm/hw_irq.h>
 #include <asm/uaccess.h>
 
 #ifdef CONFIG_KDB

arch/ia64/hp/zx1/hpzx1_machvec.c

 #define MACHVEC_PLATFORM_NAME	hpzx1
 #include <asm/machvec_init.h>
-#define MACHVEC_PLATFORM_NAME	hpzx1
-#include <asm/machvec_init.h>

arch/ia64/ia32/binfmt_elf32.c

@@ -67,7 +67,7 @@ ia32_install_shared_page (struct vm_area_struct *vma, unsigned long address, int
 }
 
 static struct vm_operations_struct ia32_shared_page_vm_ops = {
-	nopage:	ia32_install_shared_page
+	.nopage =ia32_install_shared_page
 };
 
 void

arch/ia64/kernel/acpi.c

@@ -56,6 +56,8 @@ asm (".weak iosapic_version");
 void (*pm_idle) (void);
 void (*pm_power_off) (void);
 
+unsigned char acpi_kbd_controller_present = 1;
+
 const char *
 acpi_get_sysname (void)
 {
@@ -206,7 +208,7 @@ struct acpi_table_madt *	acpi_madt __initdata;
 static int __init
 acpi_parse_lapic_addr_ovr (acpi_table_entry_header *header)
 {
-	struct acpi_table_lapic_addr_ovr *lapic = NULL;
+	struct acpi_table_lapic_addr_ovr *lapic;
 
 	lapic = (struct acpi_table_lapic_addr_ovr *) header;
 	if (!lapic)
@@ -226,7 +228,7 @@ acpi_parse_lapic_addr_ovr (acpi_table_entry_header *header)
 static int __init
 acpi_parse_lsapic (acpi_table_entry_header *header)
 {
-	struct acpi_table_lsapic *lsapic = NULL;
+	struct acpi_table_lsapic *lsapic;
 
 	lsapic = (struct acpi_table_lsapic *) header;
 	if (!lsapic)
@@ -262,7 +264,7 @@ acpi_parse_lsapic (acpi_table_entry_header *header)
 static int __init
 acpi_parse_lapic_nmi (acpi_table_entry_header *header)
 {
-	struct acpi_table_lapic_nmi *lacpi_nmi = NULL;
+	struct acpi_table_lapic_nmi *lacpi_nmi;
 
 	lacpi_nmi = (struct acpi_table_lapic_nmi*) header;
 	if (!lacpi_nmi)
@@ -279,7 +281,7 @@ acpi_parse_lapic_nmi (acpi_table_entry_header *header)
 static int __init
 acpi_find_iosapic (int global_vector, u32 *irq_base, char **iosapic_address)
 {
-	struct acpi_table_iosapic *iosapic = NULL;
+	struct acpi_table_iosapic *iosapic;
 	int ver = 0;
 	int max_pin = 0;
 	char *p = 0;
@@ -338,7 +340,7 @@ acpi_parse_iosapic (acpi_table_entry_header *header)
 static int __init
 acpi_parse_plat_int_src (acpi_table_entry_header *header)
 {
-	struct acpi_table_plat_int_src *plintsrc = NULL;
+	struct acpi_table_plat_int_src *plintsrc;
 	int vector = 0;
 	u32 irq_base = 0;
 	char *iosapic_address = NULL;
@@ -381,7 +383,7 @@ acpi_parse_plat_int_src (acpi_table_entry_header *header)
 static int __init
 acpi_parse_int_src_ovr (acpi_table_entry_header *header)
 {
-	struct acpi_table_int_src_ovr *p = NULL;
+	struct acpi_table_int_src_ovr *p;
 
 	p = (struct acpi_table_int_src_ovr *) header;
 	if (!p)
@@ -404,7 +406,7 @@ acpi_parse_int_src_ovr (acpi_table_entry_header *header)
 static int __init
 acpi_parse_nmi_src (acpi_table_entry_header *header)
 {
-	struct acpi_table_nmi_src *nmi_src = NULL;
+	struct acpi_table_nmi_src *nmi_src;
 
 	nmi_src = (struct acpi_table_nmi_src*) header;
 	if (!nmi_src)
@@ -425,10 +427,6 @@ acpi_parse_madt (unsigned long phys_addr, unsigned long size)
 		return -EINVAL;
 
 	acpi_madt = (struct acpi_table_madt *) __va(phys_addr);
-	if (!acpi_madt) {
-		printk(KERN_WARNING PREFIX "Unable to map MADT\n");
-		return -ENODEV;
-	}
 
 	/* Get base address of IPI Message Block */
 
@@ -442,6 +440,28 @@ acpi_parse_madt (unsigned long phys_addr, unsigned long size)
 }
 
 
+static int __init
+acpi_parse_fadt (unsigned long phys_addr, unsigned long size)
+{
+	struct acpi_table_header *fadt_header;
+	fadt_descriptor_rev2 *fadt;
+
+	if (!phys_addr || !size)
+		return -EINVAL;
+
+	fadt_header = (struct acpi_table_header *) __va(phys_addr);
+	if (fadt_header->revision != 3)
+		return -ENODEV;		/* Only deal with ACPI 2.0 FADT */
+
+	fadt = (fadt_descriptor_rev2 *) fadt_header;
+
+	if (!(fadt->iapc_boot_arch & BAF_8042_KEYBOARD_CONTROLLER))
+		acpi_kbd_controller_present = 0;
+
+	return 0;
+}
+
+
 int __init
 acpi_find_rsdp (unsigned long *rsdp_phys)
 {
@@ -467,8 +487,8 @@ acpi_find_rsdp (unsigned long *rsdp_phys)
 static int __init
 acpi_parse_spcr (unsigned long phys_addr, unsigned long size)
 {
-	acpi_ser_t *spcr = NULL;
-	unsigned long global_int = 0;
+	acpi_ser_t *spcr;
+	unsigned long global_int;
 
 	if (!phys_addr || !size)
 		return -EINVAL;
@@ -486,11 +506,6 @@ acpi_parse_spcr (unsigned long phys_addr, unsigned long size)
 	 */
 
 	spcr = (acpi_ser_t *) __va(phys_addr);
-	if (!spcr) {
-		printk(KERN_WARNING PREFIX "Unable to map SPCR\n");
-		return -ENODEV;
-	}
-
 	setup_serial_acpi(spcr);
 
 	if (spcr->length < sizeof(acpi_ser_t))
@@ -527,11 +542,11 @@ acpi_parse_spcr (unsigned long phys_addr, unsigned long size)
 int __init
 acpi_boot_init (char *cmdline)
 {
-	int result = 0;
+	int result;
 
 	/* Initialize the ACPI boot-time table parser */
 	result = acpi_table_init(cmdline);
-	if (0 != result)
+	if (result)
 		return result;
 
 	/*
@@ -542,57 +557,49 @@ acpi_boot_init (char *cmdline)
 	 * information -- the successor to MPS tables.
 	 */
 
-	result = acpi_table_parse(ACPI_APIC, acpi_parse_madt);
-	if (1 > result)
-		return result;
+	if (acpi_table_parse(ACPI_APIC, acpi_parse_madt) < 1) {
+		printk(KERN_ERR PREFIX "Can't find MADT\n");
+		goto skip_madt;
+	}
 
 	/* Local APIC */
 
-	result = acpi_table_parse_madt(ACPI_MADT_LAPIC_ADDR_OVR, acpi_parse_lapic_addr_ovr);
-	if (0 > result) {
+	if (acpi_table_parse_madt(ACPI_MADT_LAPIC_ADDR_OVR,
+				  acpi_parse_lapic_addr_ovr) < 0)
 		printk(KERN_ERR PREFIX "Error parsing LAPIC address override entry\n");
-		return result;
-	}
 
-	result = acpi_table_parse_madt(ACPI_MADT_LSAPIC, acpi_parse_lsapic);
-	if (1 > result) {
-		printk(KERN_ERR PREFIX "Error parsing MADT - no LAPIC entries!\n");
-		return -ENODEV;
-	}
+	if (acpi_table_parse_madt(ACPI_MADT_LSAPIC,
+				  acpi_parse_lsapic) < 1)
+		printk(KERN_ERR PREFIX "Error parsing MADT - no LAPIC entries\n");
 
-	result = acpi_table_parse_madt(ACPI_MADT_LAPIC_NMI, acpi_parse_lapic_nmi);
-	if (0 > result) {
+	if (acpi_table_parse_madt(ACPI_MADT_LAPIC_NMI,
+				  acpi_parse_lapic_nmi) < 0)
 		printk(KERN_ERR PREFIX "Error parsing LAPIC NMI entry\n");
-		return result;
-	}
 
 	/* I/O APIC */
 
-	result = acpi_table_parse_madt(ACPI_MADT_IOSAPIC, acpi_parse_iosapic);
-	if (1 > result) {
-		printk(KERN_ERR PREFIX "Error parsing MADT - no IOAPIC entries!\n");
-		return ((result == 0) ? -ENODEV : result);
-	}
+	if (acpi_table_parse_madt(ACPI_MADT_IOSAPIC,
+				  acpi_parse_iosapic) < 1)
+		printk(KERN_ERR PREFIX "Error parsing MADT - no IOAPIC entries\n");
 
 	/* System-Level Interrupt Routing */
 
-	result = acpi_table_parse_madt(ACPI_MADT_PLAT_INT_SRC, acpi_parse_plat_int_src);
-	if (0 > result) {
+	if (acpi_table_parse_madt(ACPI_MADT_PLAT_INT_SRC,
+				  acpi_parse_plat_int_src) < 0)
 		printk(KERN_ERR PREFIX "Error parsing platform interrupt source entry\n");
-		return result;
-	}
 
-	result = acpi_table_parse_madt(ACPI_MADT_INT_SRC_OVR, acpi_parse_int_src_ovr);
-	if (0 > result) {
+	if (acpi_table_parse_madt(ACPI_MADT_INT_SRC_OVR,
+				  acpi_parse_int_src_ovr) < 0)
 		printk(KERN_ERR PREFIX "Error parsing interrupt source overrides entry\n");
-		return result;
-	}
 
-	result = acpi_table_parse_madt(ACPI_MADT_NMI_SRC, acpi_parse_nmi_src);
-	if (0 > result) {
+	if (acpi_table_parse_madt(ACPI_MADT_NMI_SRC,
+				  acpi_parse_nmi_src) < 0)
 		printk(KERN_ERR PREFIX "Error parsing NMI SRC entry\n");
-		return result;
-	}
+skip_madt:
+
+	/* FADT says whether a legacy keyboard controller is present. */
+	if (acpi_table_parse(ACPI_FACP, acpi_parse_fadt) < 1)
+		printk(KERN_ERR PREFIX "Can't find FADT\n");
 
 #ifdef CONFIG_SERIAL_ACPI
 	/*
@@ -602,7 +609,7 @@ acpi_boot_init (char *cmdline)
 	 *      serial ports, EC, SMBus, etc.
 	 */
 	acpi_table_parse(ACPI_SPCR, acpi_parse_spcr);
-#endif /*CONFIG_SERIAL_ACPI*/
+#endif
 
 #ifdef CONFIG_SMP
 	if (available_cpus == 0) {
@@ -625,9 +632,9 @@ acpi_boot_init (char *cmdline)
 int __init
 acpi_get_prt (struct pci_vector_struct **vectors, int *count)
 {
-	struct pci_vector_struct *vector = NULL;
-	struct list_head *node = NULL;
-	struct acpi_prt_entry *entry = NULL;
+	struct pci_vector_struct *vector;
+	struct list_head *node;
+	struct acpi_prt_entry *entry;
 	int i = 0;
 
 	if (!vectors || !count)

arch/ia64/kernel/efi.c

@@ -125,9 +125,79 @@ efi_gettimeofday (struct timeval *tv)
 	tv->tv_usec = tm.nanosecond / 1000;
 }
 
+static int
+is_available_memory (efi_memory_desc_t *md)
+{
+	if (!(md->attribute & EFI_MEMORY_WB))
+		return 0;
+
+	switch (md->type) {
+	      case EFI_LOADER_CODE:
+	      case EFI_LOADER_DATA:
+	      case EFI_BOOT_SERVICES_CODE:
+	      case EFI_BOOT_SERVICES_DATA:
+	      case EFI_CONVENTIONAL_MEMORY:
+		return 1;
+	}
+	return 0;
+}
+
+/*
+ * Trim descriptor MD so its starts at address START_ADDR.  If the descriptor covers
+ * memory that is normally available to the kernel, issue a warning that some memory
+ * is being ignored.
+ */
+static void
+trim_bottom (efi_memory_desc_t *md, u64 start_addr)
+{
+	u64 num_skipped_pages;
+
+	if (md->phys_addr >= start_addr || !md->num_pages)
+		return;
+
+	num_skipped_pages = (start_addr - md->phys_addr) >> EFI_PAGE_SHIFT;
+	if (num_skipped_pages > md->num_pages)
+		num_skipped_pages = md->num_pages;
+
+	if (is_available_memory(md))
+		printk(KERN_NOTICE "efi.%s: ignoring %luKB of memory at 0x%lx due to granule hole "
+		       "at 0x%lx\n", __FUNCTION__,
+		       (num_skipped_pages << EFI_PAGE_SHIFT) >> 10,
+		       md->phys_addr, start_addr - IA64_GRANULE_SIZE);
+	/*
+	 * NOTE: Don't set md->phys_addr to START_ADDR because that could cause the memory
+	 * descriptor list to become unsorted.  In such a case, md->num_pages will be
+	 * zero, so the Right Thing will happen.
+	 */
+	md->phys_addr += num_skipped_pages << EFI_PAGE_SHIFT;
+	md->num_pages -= num_skipped_pages;
+}
+
+static void
+trim_top (efi_memory_desc_t *md, u64 end_addr)
+{
+	u64 num_dropped_pages, md_end_addr;
+
+	md_end_addr = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
+
+	if (md_end_addr <= end_addr || !md->num_pages)
+		return;
+
+	num_dropped_pages = (md_end_addr - end_addr) >> EFI_PAGE_SHIFT;
+	if (num_dropped_pages > md->num_pages)
+		num_dropped_pages = md->num_pages;
+
+	if (is_available_memory(md))
+		printk(KERN_NOTICE "efi.%s: ignoring %luKB of memory at 0x%lx due to granule hole "
+		       "at 0x%lx\n", __FUNCTION__,
+		       (num_dropped_pages << EFI_PAGE_SHIFT) >> 10,
+		       md->phys_addr, end_addr);
+	md->num_pages -= num_dropped_pages;
+}
+
 /*
- * Walks the EFI memory map and calls CALLBACK once for each EFI
- * memory descriptor that has memory that is available for OS use.
+ * Walks the EFI memory map and calls CALLBACK once for each EFI memory descriptor that
+ * has memory that is available for OS use.
  */
 void
 efi_memmap_walk (efi_freemem_callback_t callback, void *arg)
@@ -137,9 +207,9 @@ efi_memmap_walk (efi_freemem_callback_t callback, void *arg)
 		u64 start;
 		u64 end;
 	} prev, curr;
-	void *efi_map_start, *efi_map_end, *p;
-	efi_memory_desc_t *md;
-	u64 efi_desc_size, start, end;
+	void *efi_map_start, *efi_map_end, *p, *q;
+	efi_memory_desc_t *md, *check_md;
+	u64 efi_desc_size, start, end, granule_addr, first_non_wb_addr = 0;
 
 	efi_map_start = __va(ia64_boot_param->efi_memmap);
 	efi_map_end   = efi_map_start + ia64_boot_param->efi_memmap_size;
@@ -147,24 +217,56 @@ efi_memmap_walk (efi_freemem_callback_t callback, void *arg)
 
 	for (p = efi_map_start; p < efi_map_end; p += efi_desc_size) {
 		md = p;
-		switch (md->type) {
-		      case EFI_LOADER_CODE:
-		      case EFI_LOADER_DATA:
-		      case EFI_BOOT_SERVICES_CODE:
-		      case EFI_BOOT_SERVICES_DATA:
-		      case EFI_CONVENTIONAL_MEMORY:
+
+		/* skip over non-WB memory descriptors; that's all we're interested in... */
 		if (!(md->attribute & EFI_MEMORY_WB))
 			continue;
+
+		if (md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) > first_non_wb_addr) {
+			/*
+			 * Search for the next run of contiguous WB memory.  Start search
+			 * at first granule boundary covered by md.
+			 */
+			granule_addr = ((md->phys_addr + IA64_GRANULE_SIZE - 1)
+					& -IA64_GRANULE_SIZE);
+			first_non_wb_addr = granule_addr;
+			for (q = p; q < efi_map_end; q += efi_desc_size) {
+				check_md = q;
+
+				if (check_md->attribute & EFI_MEMORY_WB)
+					trim_bottom(md, granule_addr);
+
+				if (check_md->phys_addr < granule_addr)
+					continue;
+
+				if (!(check_md->attribute & EFI_MEMORY_WB))
+					break;	/* hit a non-WB region; stop search */
+
+				if (check_md->phys_addr != first_non_wb_addr)
+					break;	/* hit a memory hole; stop search */
+
+				first_non_wb_addr += check_md->num_pages << EFI_PAGE_SHIFT;
+			}
+			/* round it down to the previous granule-boundary: */
+			first_non_wb_addr &= -IA64_GRANULE_SIZE;
+
+			if (!(first_non_wb_addr > granule_addr))
+				continue;	/* couldn't find enough contiguous memory */
+		}
+
+		/* BUG_ON((md->phys_addr >> IA64_GRANULE_SHIFT) < first_non_wb_addr); */
+
+		trim_top(md, first_non_wb_addr);
+
+		if (is_available_memory(md)) {
 			if (md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) > mem_limit) {
 				if (md->phys_addr > mem_limit)
 					continue;
 				md->num_pages = (mem_limit - md->phys_addr) >> EFI_PAGE_SHIFT;
 			}
-			if (md->num_pages == 0) {
-				printk("efi_memmap_walk: ignoring empty region at 0x%lx",
-				       md->phys_addr);
+
+			if (md->num_pages == 0)
 				continue;
-			}
 
 			curr.start = PAGE_OFFSET + md->phys_addr;
 			curr.end   = curr.start + (md->num_pages << EFI_PAGE_SHIFT);
@@ -187,10 +289,6 @@ efi_memmap_walk (efi_freemem_callback_t callback, void *arg)
 					prev = curr;
 				}
 			}
-			break;
-
-		      default:
-			continue;
 		}
 	}
 	if (prev_valid) {
@@ -268,8 +366,9 @@ efi_map_pal_code (void)
 		 */
 		psr = ia64_clear_ic();
 		ia64_itr(0x1, IA64_TR_PALCODE, vaddr & mask,
-			 pte_val(pfn_pte(md->phys_addr >> PAGE_SHIFT, PAGE_KERNEL)), IA64_GRANULE_SHIFT);
-		ia64_set_psr(psr);
+			 pte_val(pfn_pte(md->phys_addr >> PAGE_SHIFT, PAGE_KERNEL)),
+			 IA64_GRANULE_SHIFT);
+		ia64_set_psr(psr);		/* restore psr */
 		ia64_srlz_i();
 	}
 }
@@ -376,7 +475,7 @@ efi_init (void)
 			md = p;
 			printk("mem%02u: type=%u, attr=0x%lx, range=[0x%016lx-0x%016lx) (%luMB)\n",
 			       i, md->type, md->attribute, md->phys_addr,
-			       md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) - 1,
+			       md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT),
 			       md->num_pages >> (20 - EFI_PAGE_SHIFT));
 		}
 	}

arch/ia64/kernel/init_task.c

@@ -34,8 +34,8 @@ union init_thread {
 	} s;
 	unsigned long stack[KERNEL_STACK_SIZE/sizeof (unsigned long)];
 } init_thread_union __attribute__((section(".data.init_task"))) = {{
-	task: INIT_TASK(init_thread_union.s.task),
-	thread_info: INIT_THREAD_INFO(init_thread_union.s.thread_info)
+	.task =		INIT_TASK(init_thread_union.s.task),
+	.thread_info =	INIT_THREAD_INFO(init_thread_union.s.thread_info)
 }};
 
 asm (".global init_task; init_task = init_thread_union");

arch/ia64/kernel/iosapic.c

@@ -88,7 +88,7 @@ static struct {
 
 static struct iosapic_irq {
 	char *addr;			/* base address of IOSAPIC */
-	unsigned char base_irq;		/* first irq assigned to this IOSAPIC */
+	unsigned int base_irq;		/* first irq assigned to this IOSAPIC */
 	char pin;			/* IOSAPIC pin (-1 => not an IOSAPIC irq) */
 	unsigned char dmode	: 3;	/* delivery mode (see iosapic.h) */
 	unsigned char polarity	: 1;	/* interrupt polarity (see iosapic.h) */
@@ -97,9 +97,9 @@ static struct iosapic_irq {
 
 static struct iosapic {
 	char *addr;			/* base address of IOSAPIC */
-	unsigned char	pcat_compat;	/* 8259 compatibility flag */
-	unsigned char 	base_irq;	/* first irq assigned to this IOSAPIC */
+	unsigned int 	base_irq;	/* first irq assigned to this IOSAPIC */
 	unsigned short 	max_pin;	/* max input pin supported in this IOSAPIC */
+	unsigned char	pcat_compat;	/* 8259 compatibility flag */
 } iosapic_lists[256] __initdata;
 
 static int num_iosapic = 0;
@@ -322,14 +322,14 @@ iosapic_end_level_irq (unsigned int irq)
 #define iosapic_ack_level_irq		nop
 
 struct hw_interrupt_type irq_type_iosapic_level = {
-	typename:	"IO-SAPIC-level",
-	startup:	iosapic_startup_level_irq,
-	shutdown:	iosapic_shutdown_level_irq,
-	enable:		iosapic_enable_level_irq,
-	disable:	iosapic_disable_level_irq,
-	ack:		iosapic_ack_level_irq,
-	end:		iosapic_end_level_irq,
-	set_affinity:	iosapic_set_affinity
+	.typename =	"IO-SAPIC-level",
+	.startup =	iosapic_startup_level_irq,
+	.shutdown =	iosapic_shutdown_level_irq,
+	.enable =	iosapic_enable_level_irq,
+	.disable =	iosapic_disable_level_irq,
+	.ack =		iosapic_ack_level_irq,
+	.end =		iosapic_end_level_irq,
+	.set_affinity =	iosapic_set_affinity
 };
 
 /*
@@ -366,14 +366,14 @@ iosapic_ack_edge_irq (unsigned int irq)
 #define iosapic_end_edge_irq		nop
 
 struct hw_interrupt_type irq_type_iosapic_edge = {
-	typename:	"IO-SAPIC-edge",
-	startup:	iosapic_startup_edge_irq,
-	shutdown:	iosapic_disable_edge_irq,
-	enable:		iosapic_enable_edge_irq,
-	disable:	iosapic_disable_edge_irq,
-	ack:		iosapic_ack_edge_irq,
-	end:		iosapic_end_edge_irq,
-	set_affinity:	iosapic_set_affinity
+	.typename =	"IO-SAPIC-edge",
+	.startup =	iosapic_startup_edge_irq,
+	.shutdown =	iosapic_disable_edge_irq,
+	.enable =	iosapic_enable_edge_irq,
+	.disable =	iosapic_disable_edge_irq,
+	.ack =		iosapic_ack_edge_irq,
+	.end =		iosapic_end_edge_irq,
+	.set_affinity =	iosapic_set_affinity
 };
 
 unsigned int
@@ -679,11 +679,10 @@ iosapic_init_pci_irq (void)
 		       pci_irq.route[i].bus, pci_irq.route[i].pci_id>>16, pci_irq.route[i].pin,
 		       iosapic_irq[vector].base_irq + iosapic_irq[vector].pin, vector);
 #endif
-
 		/*
-		 * Forget not to program the IOSAPIC RTE per ACPI _PRT
+		 * NOTE: The IOSAPIC RTE will be programmed in iosapic_pci_fixup().  It
+		 * needs to be done there to ensure PCI hotplug works right.
 		 */
-		set_rte(vector, (ia64_get_lid() >> 16) & 0xffff);
 	}
 }
 

arch/ia64/kernel/irq_ia64.c

@@ -36,6 +36,10 @@
 #include <asm/pgtable.h>
 #include <asm/system.h>
 
+#ifdef CONFIG_PERFMON
+# include <asm/perfmon.h>
+#endif
+
 #define IRQ_DEBUG	0
 
 /* default base addr of IPI table */
@@ -144,9 +148,9 @@ ia64_handle_irq (ia64_vector vector, struct pt_regs *regs)
 extern void handle_IPI (int irq, void *dev_id, struct pt_regs *regs);
 
 static struct irqaction ipi_irqaction = {
-	handler:	handle_IPI,
-	flags:		SA_INTERRUPT,
-	name:		"IPI"
+	.handler =	handle_IPI,
+	.flags =	SA_INTERRUPT,
+	.name =		"IPI"
 };
 #endif
 
@@ -172,6 +176,9 @@ init_IRQ (void)
 	register_percpu_irq(IA64_SPURIOUS_INT_VECTOR, NULL);
 #ifdef CONFIG_SMP
 	register_percpu_irq(IA64_IPI_VECTOR, &ipi_irqaction);
+#endif
+#ifdef CONFIG_PERFMON
+	perfmon_init_percpu();
 #endif
 	platform_irq_init();
 }

arch/ia64/kernel/irq_lsapic.c

@@ -27,12 +27,12 @@ lsapic_noop (unsigned int irq)
 }
 
 struct hw_interrupt_type irq_type_ia64_lsapic = {
-	typename:	"LSAPIC",
-	startup:	lsapic_noop_startup,
-	shutdown:	lsapic_noop,
-	enable:		lsapic_noop,
-	disable:	lsapic_noop,
-	ack:		lsapic_noop,
-	end:		lsapic_noop,
-	set_affinity:	(void (*)(unsigned int, unsigned long)) lsapic_noop
+	.typename =	"LSAPIC",
+	.startup =	lsapic_noop_startup,
+	.shutdown =	lsapic_noop,
+	.enable =	lsapic_noop,
+	.disable =	lsapic_noop,
+	.ack =		lsapic_noop,
+	.end =		lsapic_noop,
+	.set_affinity =	(void (*)(unsigned int, unsigned long)) lsapic_noop
 };

arch/ia64/kernel/machvec.c

@@ -11,13 +11,16 @@
 struct ia64_machine_vector ia64_mv;
 
 /*
- * Most platforms use this routine for mapping page frame addresses
- * into a memory map index.
+ * Most platforms use this routine for mapping page frame addresses into a memory map
+ * index.
+ *
+ * Note: we can't use __pa() because map_nr_dense(X) MUST map to something >= max_mapnr if
+ * X is outside the identity mapped kernel space.
  */
 unsigned long
 map_nr_dense (unsigned long addr)
 {
-	return MAP_NR_DENSE(addr);
+	return (addr - PAGE_OFFSET) >> PAGE_SHIFT;
 }
 
 static struct ia64_machine_vector *

arch/ia64/kernel/mca.c

@@ -82,27 +82,27 @@ extern void			ia64_slave_init_handler (void);
 extern struct hw_interrupt_type	irq_type_iosapic_level;
 
 static struct irqaction cmci_irqaction = {
-	handler:    ia64_mca_cmc_int_handler,
-	flags:      SA_INTERRUPT,
-	name:       "cmc_hndlr"
+	.handler =	ia64_mca_cmc_int_handler,
+	.flags =	SA_INTERRUPT,
+	.name =		"cmc_hndlr"
 };
 
 static struct irqaction mca_rdzv_irqaction = {
-	handler:    ia64_mca_rendez_int_handler,
-	flags:      SA_INTERRUPT,
-	name:       "mca_rdzv"
+	.handler =	ia64_mca_rendez_int_handler,
+	.flags =	SA_INTERRUPT,
+	.name =		"mca_rdzv"
 };
 
 static struct irqaction mca_wkup_irqaction = {
-	handler:    ia64_mca_wakeup_int_handler,
-	flags:      SA_INTERRUPT,
-	name:       "mca_wkup"
+	.handler =	ia64_mca_wakeup_int_handler,
+	.flags =	SA_INTERRUPT,
+	.name =		"mca_wkup"
 };
 
 static struct irqaction mca_cpe_irqaction = {
-	handler:    ia64_mca_cpe_int_handler,
-	flags:      SA_INTERRUPT,
-	name:       "cpe_hndlr"
+	.handler =	ia64_mca_cpe_int_handler,
+	.flags =	SA_INTERRUPT,
+	.name =		"cpe_hndlr"
 };
 
 /*

arch/ia64/kernel/mca_asm.S

@@ -684,9 +684,9 @@ ia64_os_mca_tlb_error_check:
 	movl	r3=SAL_GET_STATE_INFO;;
 	DATA_VA_TO_PA(r7);;		// convert to physical address
 	ld8	r8=[r7],8;;		// get pdesc function pointer
-	DATA_VA_TO_PA(r8)		// convert to physical address
+	dep	r8=0,r8,61,3;;		// convert SAL VA to PA
 	ld8	r1=[r7];;		// set new (ia64_sal) gp
-	DATA_VA_TO_PA(r1)		// convert to physical address
+	dep	r1=0,r1,61,3;;		// convert SAL VA to PA
 	mov	b6=r8
 
 	alloc	r5=ar.pfs,8,0,8,0;;	// allocate stack frame for SAL call

arch/ia64/kernel/pci.c

@@ -265,12 +265,37 @@ pcibios_fixup_pbus_ranges (struct pci_bus * bus, struct pbus_set_ranges_data * r
 int
 pcibios_enable_device (struct pci_dev *dev)
 {
+	u16 cmd, old_cmd;
+	int idx;
+	struct resource *r;
+
 	if (!dev)
 		return -EINVAL;
 
-	/* Not needed, since we enable all devices at startup.  */
+	pci_read_config_word(dev, PCI_COMMAND, &cmd);
+	old_cmd = cmd;
+	for (idx=0; idx<6; idx++) {
+		r = &dev->resource[idx];
+		if (!r->start && r->end) {
+			printk(KERN_ERR
+			       "PCI: Device %s not available because of resource collisions\n",
+			       dev->slot_name);
+			return -EINVAL;
+		}
+		if (r->flags & IORESOURCE_IO)
+			cmd |= PCI_COMMAND_IO;
+		if (r->flags & IORESOURCE_MEM)
+			cmd |= PCI_COMMAND_MEMORY;
+	}
+	if (dev->resource[PCI_ROM_RESOURCE].start)
+		cmd |= PCI_COMMAND_MEMORY;
+	if (cmd != old_cmd) {
+		printk("PCI: Enabling device %s (%04x -> %04x)\n", dev->slot_name, old_cmd, cmd);
+		pci_write_config_word(dev, PCI_COMMAND, cmd);
+	}
 
 	printk(KERN_INFO "PCI: Found IRQ %d for device %s\n", dev->irq, dev->slot_name);
+
 	return 0;
 }
 

arch/ia64/kernel/perfmon.c

@@ -106,6 +106,12 @@
 
 #define PFM_REG_RETFLAG_SET(flags, val)	do { flags &= ~PFM_REG_RETFL_MASK; flags |= (val); } while(0)
 
+#ifdef CONFIG_SMP
+#define cpu_is_online(i) (cpu_online_map & (1UL << i))
+#else
+#define cpu_is_online(i)        (i==0)
+#endif
+
 /*
  * debugging
  */
@@ -277,8 +283,8 @@ typedef struct {
 typedef struct {
 	pfm_pmu_reg_type_t	type;
 	int			pm_pos;
-	int			(*read_check)(struct task_struct *task, unsigned int cnum, unsigned long *val);
-	int			(*write_check)(struct task_struct *task, unsigned int cnum, unsigned long *val);
+	int			(*read_check)(struct task_struct *task, unsigned int cnum, unsigned long *val, struct pt_regs *regs);
+	int			(*write_check)(struct task_struct *task, unsigned int cnum, unsigned long *val, struct pt_regs *regs);
 	unsigned long		dep_pmd[4];
 	unsigned long		dep_pmc[4];
 } pfm_reg_desc_t;
@@ -396,7 +402,7 @@ static unsigned long reset_pmcs[IA64_NUM_PMC_REGS];	/* contains PAL reset values
 static void pfm_vm_close(struct vm_area_struct * area);
 
 static struct vm_operations_struct pfm_vm_ops={
-	close: pfm_vm_close
+	.close = pfm_vm_close
 };
 
 /*
@@ -902,8 +908,8 @@ pfx_is_sane(struct task_struct *task, pfarg_context_t *pfx)
 		/*
 		 * and it must be a valid CPU
 		 */
-		cpu = ffs(pfx->ctx_cpu_mask);
-		if (!cpu_online(cpu)) {
+		cpu = ffz(~pfx->ctx_cpu_mask);
+		if (cpu_is_online(cpu) == 0) {
 			DBprintk(("CPU%d is not online\n", cpu));
 			return -EINVAL;
 		}
@@ -925,11 +931,12 @@ pfx_is_sane(struct task_struct *task, pfarg_context_t *pfx)
 			DBprintk(("must have notify_pid when blocking for [%d]\n", task->pid));
 			return -EINVAL;
 		}
-
+#if 0
 		if ((ctx_flags & PFM_FL_NOTIFY_BLOCK) && pfx->ctx_notify_pid == task->pid) {
 			DBprintk(("cannot notify self when blocking for [%d]\n", task->pid));
 			return -EINVAL;
 		}
+#endif
 	}
 	/* probably more to add here */
 
@@ -968,7 +975,7 @@ pfm_context_create(struct task_struct *task, pfm_context_t *ctx, void *req, int
 	if (ctx_flags & PFM_FL_SYSTEM_WIDE) {
 
 		/* at this point, we know there is at least one bit set */
-		cpu = ffs(tmp.ctx_cpu_mask) - 1;
+		cpu = ffz(~tmp.ctx_cpu_mask);
 
 		DBprintk(("requesting CPU%d currently on CPU%d\n",cpu, smp_processor_id()));
 
@@ -1280,7 +1287,7 @@ pfm_write_pmcs(struct task_struct *task, pfm_context_t *ctx, void *arg, int coun
 		/*
 		 * execute write checker, if any
 		 */
-		if (PMC_WR_FUNC(cnum)) ret = PMC_WR_FUNC(cnum)(task, cnum, &tmp.reg_value);
+		if (PMC_WR_FUNC(cnum)) ret = PMC_WR_FUNC(cnum)(task, cnum, &tmp.reg_value, regs);
 abort_mission:
 		if (ret == -EINVAL) reg_retval = PFM_REG_RETFL_EINVAL;
 
@@ -1371,7 +1378,7 @@ pfm_write_pmds(struct task_struct *task, pfm_context_t *ctx, void *arg, int coun
 		/*
 		 * execute write checker, if any
 		 */
-		if (PMD_WR_FUNC(cnum)) ret = PMD_WR_FUNC(cnum)(task, cnum, &tmp.reg_value);
+		if (PMD_WR_FUNC(cnum)) ret = PMD_WR_FUNC(cnum)(task, cnum, &tmp.reg_value, regs);
 abort_mission:
 		if (ret == -EINVAL) reg_retval = PFM_REG_RETFL_EINVAL;
 
@@ -1394,6 +1401,8 @@ pfm_write_pmds(struct task_struct *task, pfm_context_t *ctx, void *arg, int coun
 
 		/* keep track of what we use */
 		CTX_USED_PMD(ctx, pmu_conf.pmd_desc[(cnum)].dep_pmd[0]);
+		/* mark this register as used as well */
+		CTX_USED_PMD(ctx, RDEP(cnum));
 
 		/* writes to unimplemented part is ignored, so this is safe */
 		ia64_set_pmd(cnum, tmp.reg_value & pmu_conf.perf_ovfl_val);
@@ -1438,7 +1447,7 @@ pfm_read_pmds(struct task_struct *task, pfm_context_t *ctx, void *arg, int count
 	DBprintk(("ctx_last_cpu=%d for [%d]\n", atomic_read(&ctx->ctx_last_cpu), task->pid));
 
 	for (i = 0; i < count; i++, req++) {
-		unsigned long reg_val = ~0UL, ctx_val = ~0UL;
+		unsigned long ctx_val = ~0UL;
 
 		if (copy_from_user(&tmp, req, sizeof(tmp))) return -EFAULT;
 
@@ -1462,7 +1471,7 @@ pfm_read_pmds(struct task_struct *task, pfm_context_t *ctx, void *arg, int count
 		 */
 		if (atomic_read(&ctx->ctx_last_cpu) == smp_processor_id()){
 			ia64_srlz_d();
-			val = reg_val = ia64_get_pmd(cnum);
+			val = ia64_get_pmd(cnum);
 			DBprintk(("reading pmd[%u]=0x%lx from hw\n", cnum, val));
 		} else {
 #ifdef CONFIG_SMP
@@ -1484,7 +1493,7 @@ pfm_read_pmds(struct task_struct *task, pfm_context_t *ctx, void *arg, int count
 			}
 #endif
 			/* context has been saved */
-			val = reg_val = th->pmd[cnum];
+			val = th->pmd[cnum];
 		}
 		if (PMD_IS_COUNTING(cnum)) {
 			/*
@@ -1493,8 +1502,6 @@ pfm_read_pmds(struct task_struct *task, pfm_context_t *ctx, void *arg, int count
 
 			val &= pmu_conf.perf_ovfl_val;
 			val += ctx_val = ctx->ctx_soft_pmds[cnum].val;
-		} else {
-			val = reg_val = ia64_get_pmd(cnum);
 		} 
 
 		tmp.reg_value = val;
@@ -1503,14 +1510,13 @@ pfm_read_pmds(struct task_struct *task, pfm_context_t *ctx, void *arg, int count
 		 * execute read checker, if any
 		 */
 		if (PMD_RD_FUNC(cnum)) {
-			ret = PMD_RD_FUNC(cnum)(task, cnum, &tmp.reg_value);
+			ret = PMD_RD_FUNC(cnum)(task, cnum, &tmp.reg_value, regs);
 		}
 
 		PFM_REG_RETFLAG_SET(tmp.reg_flags, ret);
 
-		DBprintk(("read pmd[%u] ret=%d soft_pmd=0x%lx reg=0x%lx pmc=0x%lx\n", 
-					cnum, ret, ctx_val, reg_val, 
-					ia64_get_pmc(cnum)));
+		DBprintk(("read pmd[%u] ret=%d value=0x%lx pmc=0x%lx\n", 
+					cnum, ret, val, ia64_get_pmc(cnum)));
 
 		if (copy_to_user(req, &tmp, sizeof(tmp))) return -EFAULT;
 	}
@@ -1553,15 +1559,11 @@ pfm_use_debug_registers(struct task_struct *task)
 	 */
 	if (ctx && ctx->ctx_fl_using_dbreg == 1) return -1;
 
-	/*
-	 * XXX: not pretty
-	 */
 	LOCK_PFS();
 
 	/*
-	 * We only allow the use of debug registers when there is no system
-	 * wide monitoring 
-	 * XXX: we could relax this by 
+	 * We cannot allow setting breakpoints when system wide monitoring
+	 * sessions are using the debug registers.
 	 */
 	if (pfm_sessions.pfs_sys_use_dbregs> 0)
 		ret = -1;
@@ -1921,7 +1923,6 @@ typedef union {
 	dbr_mask_reg_t dbr;
 } dbreg_t;
 
-
 static int
 pfm_write_ibr_dbr(int mode, struct task_struct *task, void *arg, int count, struct pt_regs *regs)
 {
@@ -1963,8 +1964,8 @@ pfm_write_ibr_dbr(int mode, struct task_struct *task, void *arg, int count, stru
 	if (ctx->ctx_fl_system) {
 		/* we mark ourselves as owner  of the debug registers */
 		ctx->ctx_fl_using_dbreg = 1;
-	} else {
-       		if (ctx->ctx_fl_using_dbreg == 0) {
+		DBprintk(("system-wide setting fl_using_dbreg for [%d]\n", task->pid));
+	} else if (first_time) {
 			ret= -EBUSY;
 			if ((thread->flags & IA64_THREAD_DBG_VALID) != 0) {
 				DBprintk(("debug registers already in use for [%d]\n", task->pid));
@@ -1973,6 +1974,7 @@ pfm_write_ibr_dbr(int mode, struct task_struct *task, void *arg, int count, stru
 			/* we mark ourselves as owner  of the debug registers */
 			ctx->ctx_fl_using_dbreg = 1;
 
+			DBprintk(("setting fl_using_dbreg for [%d]\n", task->pid));
 			/* 
 			 * Given debug registers cannot be used for both debugging 
 			 * and performance monitoring at the same time, we reuse
@@ -1980,10 +1982,18 @@ pfm_write_ibr_dbr(int mode, struct task_struct *task, void *arg, int count, stru
 			 */
 			memset(task->thread.dbr, 0, sizeof(task->thread.dbr));
 			memset(task->thread.ibr, 0, sizeof(task->thread.ibr));
+	}
 
+	if (first_time) {
+		DBprintk(("[%d] clearing ibrs,dbrs\n", task->pid));
 		/*
 	 	 * clear hardware registers to make sure we don't
-			 * pick up stale state
+	 	 * pick up stale state. 
+		 *
+		 * for a system wide session, we do not use
+		 * thread.dbr, thread.ibr because this process
+		 * never leaves the current CPU and the state
+		 * is shared by all processes running on it
 	 	 */
 		for (i=0; i < pmu_conf.num_ibrs; i++) {
 			ia64_set_ibr(i, 0UL);
@@ -1994,7 +2004,6 @@ pfm_write_ibr_dbr(int mode, struct task_struct *task, void *arg, int count, stru
 		}
 		ia64_srlz_d();
 	}
-	}
 
 	ret = -EFAULT;
 
@@ -2361,9 +2370,9 @@ sys_perfmonctl (pid_t pid, int cmd, void *arg, int count, long arg5, long arg6,
 {
 	struct pt_regs *regs = (struct pt_regs *)&stack;
 	struct task_struct *task = current;
-	pfm_context_t *ctx = task->thread.pfm_context;
+	pfm_context_t *ctx;
 	size_t sz;
-	int ret = -ESRCH, narg;
+	int ret, narg;
 
 	/* 
 	 * reject any call if perfmon was disabled at initialization time
@@ -2393,6 +2402,8 @@ sys_perfmonctl (pid_t pid, int cmd, void *arg, int count, long arg5, long arg6,
 
 		if (pid != current->pid) {
 
+			ret = -ESRCH;
+
 			read_lock(&tasklist_lock);
 
 			task = find_task_by_pid(pid);
@@ -2407,10 +2418,11 @@ sys_perfmonctl (pid_t pid, int cmd, void *arg, int count, long arg5, long arg6,
 				ret = check_task_state(task);
 				if (ret != 0) goto abort_call;
 			}
-			ctx = task->thread.pfm_context;
 		}
 	} 
 
+	ctx = task->thread.pfm_context;
+
 	if (PFM_CMD_USE_CTX(cmd)) {
 		ret = -EINVAL;
 	       if (ctx == NULL) {
@@ -2953,11 +2965,6 @@ perfmon_interrupt (int irq, void *arg, struct pt_regs *regs)
 static int
 perfmon_proc_info(char *page)
 {
-#ifdef CONFIG_SMP
-#define cpu_is_online(i) (cpu_online_map & (1UL << i))
-#else
-#define cpu_is_online(i)        1
-#endif
 	char *p = page;
 	int i;
 
@@ -4118,9 +4125,9 @@ pfm_cleanup_notifiers(struct task_struct *task)
 }
 
 static struct irqaction perfmon_irqaction = {
-	handler:	perfmon_interrupt,
-	flags:		SA_INTERRUPT,
-	name:		"perfmon"
+	.handler =	perfmon_interrupt,
+	.flags =	SA_INTERRUPT,
+	.name =		"perfmon"
 };
 
 
@@ -4150,11 +4157,6 @@ perfmon_init (void)
 	pal_perf_mon_info_u_t pm_info;
 	s64 status;
 
-	register_percpu_irq(IA64_PERFMON_VECTOR, &perfmon_irqaction);
-
-	ia64_set_pmv(IA64_PERFMON_VECTOR);
-	ia64_srlz_d();
-
 	pmu_conf.pfm_is_disabled = 1;
 
 	printk("perfmon: version %u.%u (sampling format v%u.%u) IRQ %u\n", 
@@ -4232,6 +4234,9 @@ __initcall(perfmon_init);
 void
 perfmon_init_percpu (void)
 {
+	if (smp_processor_id() == 0)
+		register_percpu_irq(IA64_PERFMON_VECTOR, &perfmon_irqaction);
+
 	ia64_set_pmv(IA64_PERFMON_VECTOR);
 	ia64_srlz_d();
 }

arch/ia64/kernel/perfmon_itanium.h

+/*
+ * This file contains the Itanium PMU register description tables
+ * and pmc checker used by perfmon.c.
+ *
+ * Copyright (C) 2002  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, unsigned int cnum, unsigned long *val, struct pt_regs *regs);
+static int pfm_write_ibr_dbr(int mode, struct task_struct *task, void *arg, int count, struct pt_regs *regs);
+
+static pfm_reg_desc_t pmc_desc[256]={
+/* pmc0  */ { PFM_REG_CONTROL, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc1  */ { PFM_REG_CONTROL, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc2  */ { PFM_REG_CONTROL, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc3  */ { PFM_REG_CONTROL, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc4  */ { PFM_REG_COUNTING, 6, NULL, NULL, {RDEP(4),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc5  */ { PFM_REG_COUNTING, 6, NULL, NULL, {RDEP(5),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc6  */ { PFM_REG_COUNTING, 6, NULL, NULL, {RDEP(6),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc7  */ { PFM_REG_COUNTING, 6, NULL, NULL, {RDEP(7),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc8  */ { PFM_REG_CONFIG, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc9  */ { PFM_REG_CONFIG, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc10 */ { PFM_REG_MONITOR, 6, NULL, NULL, {RDEP(0)|RDEP(1),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc11 */ { PFM_REG_MONITOR, 6, NULL, pfm_ita_pmc_check, {RDEP(2)|RDEP(3)|RDEP(17),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc12 */ { PFM_REG_MONITOR, 6, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc13 */ { PFM_REG_CONFIG, 0, NULL, pfm_ita_pmc_check, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+	    { PFM_REG_NONE, 0, NULL, NULL, {0,}, {0,}}, /* end marker */
+};
+
+static pfm_reg_desc_t pmd_desc[256]={
+/* pmd0  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(1),0UL, 0UL, 0UL}, {RDEP(10),0UL, 0UL, 0UL}},
+/* pmd1  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(0),0UL, 0UL, 0UL}, {RDEP(10),0UL, 0UL, 0UL}},
+/* pmd2  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(3)|RDEP(17),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+/* pmd3  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(2)|RDEP(17),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+/* pmd4  */ { PFM_REG_COUNTING, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(4),0UL, 0UL, 0UL}},
+/* pmd5  */ { PFM_REG_COUNTING, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(5),0UL, 0UL, 0UL}},
+/* pmd6  */ { PFM_REG_COUNTING, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(6),0UL, 0UL, 0UL}},
+/* pmd7  */ { PFM_REG_COUNTING, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(7),0UL, 0UL, 0UL}},
+/* pmd8  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd9  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd10 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd11 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd12 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd13 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd14 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd15 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd16 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd17 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(2)|RDEP(3),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+	    { PFM_REG_NONE, 0, NULL, NULL, {0,}, {0,}}, /* end marker */
+};
+
+static int
+pfm_ita_pmc_check(struct task_struct *task, unsigned int cnum, unsigned long *val, struct pt_regs *regs)
+{
+	pfm_context_t *ctx = task->thread.pfm_context;
+	int ret;
+
+	/*
+	 * 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) {
+
+		/* don't mix debug with perfmon */
+		if ((task->thread.flags & IA64_THREAD_DBG_VALID) != 0) return -EINVAL;
+
+		/* 
+		 * a count of 0 will mark the debug registers as in use and also
+		 * ensure that they are properly cleared.
+		 */
+		ret = pfm_write_ibr_dbr(1, task, NULL, 0, regs);
+		if (ret) return ret;
+	}
+
+	/*
+	 * 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) {
+
+		/* don't mix debug with perfmon */
+		if ((task->thread.flags & IA64_THREAD_DBG_VALID) != 0) return -EINVAL;
+
+		/* 
+		 * a count of 0 will mark the debug registers as in use and also
+		 * ensure that they are properly cleared.
+		 */
+		ret = pfm_write_ibr_dbr(0, task, NULL, 0, regs);
+		if (ret) return ret;
+	}
+	return 0;
+}
+

arch/ia64/kernel/perfmon_mckinley.h

+/*
+ * This file contains the McKinley PMU register description tables
+ * and pmc checker used by perfmon.c.
+ *
+ * Copyright (C) 2002  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, unsigned int cnum, unsigned long *val, struct pt_regs *regs);
+static int pfm_write_ibr_dbr(int mode, struct task_struct *task, void *arg, int count, struct pt_regs *regs);
+
+static pfm_reg_desc_t pmc_desc[256]={
+/* pmc0  */ { PFM_REG_CONTROL, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc1  */ { PFM_REG_CONTROL, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc2  */ { PFM_REG_CONTROL, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc3  */ { PFM_REG_CONTROL, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc4  */ { PFM_REG_COUNTING, 6, NULL, pfm_mck_pmc_check, {RDEP(4),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc5  */ { PFM_REG_COUNTING, 6, NULL, NULL, {RDEP(5),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc6  */ { PFM_REG_COUNTING, 6, NULL, NULL, {RDEP(6),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc7  */ { PFM_REG_COUNTING, 6, NULL, NULL, {RDEP(7),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc8  */ { PFM_REG_CONFIG, 0, NULL, pfm_mck_pmc_check, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc9  */ { PFM_REG_CONFIG, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc10 */ { PFM_REG_MONITOR, 4, NULL, NULL, {RDEP(0)|RDEP(1),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc11 */ { PFM_REG_MONITOR, 6, NULL, NULL, {RDEP(2)|RDEP(3)|RDEP(17),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc12 */ { PFM_REG_MONITOR, 6, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc13 */ { PFM_REG_CONFIG, 0, NULL, pfm_mck_pmc_check, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc14 */ { PFM_REG_CONFIG, 0, NULL, pfm_mck_pmc_check, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc15 */ { PFM_REG_CONFIG, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+	    { PFM_REG_NONE, 0, NULL, NULL, {0,}, {0,}}, /* end marker */
+};
+
+static pfm_reg_desc_t pmd_desc[256]={
+/* pmd0  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(1),0UL, 0UL, 0UL}, {RDEP(10),0UL, 0UL, 0UL}},
+/* pmd1  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(0),0UL, 0UL, 0UL}, {RDEP(10),0UL, 0UL, 0UL}},
+/* pmd2  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(3)|RDEP(17),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+/* pmd3  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(2)|RDEP(17),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+/* pmd4  */ { PFM_REG_COUNTING, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(4),0UL, 0UL, 0UL}},
+/* pmd5  */ { PFM_REG_COUNTING, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(5),0UL, 0UL, 0UL}},
+/* pmd6  */ { PFM_REG_COUNTING, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(6),0UL, 0UL, 0UL}},
+/* pmd7  */ { PFM_REG_COUNTING, 0, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(7),0UL, 0UL, 0UL}},
+/* pmd8  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd9  */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd10 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd11 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd12 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(13)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd13 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(14)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd14 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(15)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd15 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(16),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd16 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(8)|RDEP(9)|RDEP(10)|RDEP(11)|RDEP(12)|RDEP(13)|RDEP(14)|RDEP(15),0UL, 0UL, 0UL}, {RDEP(12),0UL, 0UL, 0UL}},
+/* pmd17 */ { PFM_REG_BUFFER, 0, NULL, NULL, {RDEP(2)|RDEP(3),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+	    { PFM_REG_NONE, 0, NULL, NULL, {0,}, {0,}}, /* end marker */
+};
+
+static int
+pfm_mck_pmc_check(struct task_struct *task, unsigned int cnum, unsigned long *val, struct pt_regs *regs)
+{
+	struct thread_struct *th = &task->thread;
+	pfm_context_t *ctx = task->thread.pfm_context;
+	int ret = 0, check_case1 = 0;
+	unsigned long val8 = 0, val14 = 0, val13 = 0;
+
+	/*
+	 * we must clear the debug registers if any pmc13.ena_dbrpX bit is enabled 
+	 * before they are written (fl_using_dbreg==0) to avoid picking up stale information. 
+	 */
+	if (cnum == 13 && (*val & (0xfUL << 45)) && ctx->ctx_fl_using_dbreg == 0) {
+
+		/* don't mix debug with perfmon */
+		if ((task->thread.flags & IA64_THREAD_DBG_VALID) != 0) return -EINVAL;
+
+		/* 
+		 * a count of 0 will mark the debug registers as in use and also
+		 * ensure that they are properly cleared.
+		 */
+		ret = pfm_write_ibr_dbr(1, task, NULL, 0, regs);
+		if (ret) return ret;
+	}
+	/* 
+	 * we must clear the (instruction) debug registers if any pmc14.ibrpX bit is enabled 
+	 * before they are (fl_using_dbreg==0) to avoid picking up stale information. 
+	 */
+	if (cnum == 14 && ((*val & 0x2222) != 0x2222) && ctx->ctx_fl_using_dbreg == 0) {
+
+		/* don't mix debug with perfmon */
+		if ((task->thread.flags & IA64_THREAD_DBG_VALID) != 0) return -EINVAL;
+
+		/* 
+		 * a count of 0 will mark the debug registers as in use and also
+		 * ensure that they are properly cleared.
+		 */
+		ret = pfm_write_ibr_dbr(0, task, NULL, 0, regs);
+		if (ret) return ret;
+
+	}
+
+	switch(cnum) {
+		case  4: *val |= 1UL << 23; /* force power enable bit */
+			 break;
+		case  8: val8 = *val;
+			 val13 = th->pmc[13];
+			 val14 = th->pmc[14];
+			 check_case1 = 1;
+			 break;
+		case 13: val8  = th->pmc[8];
+			 val13 = *val;
+			 val14 = th->pmc[14];
+			 check_case1 = 1;
+			 break;
+		case 14: val8  = th->pmc[13];
+			 val13 = th->pmc[13];
+			 val14 = *val;
+			 check_case1 = 1;
+			 break;
+	}
+	/* check illegal configuration which can produce inconsistencies in tagging
+	 * i-side events in L1D and L2 caches
+	 */
+	if (check_case1) {
+		ret =   ((val13 >> 45) & 0xf) == 0 
+		   && ((val8 & 0x1) == 0)
+		   && ((((val14>>1) & 0x3) == 0x2 || ((val14>>1) & 0x3) == 0x0)
+		       ||(((val14>>4) & 0x3) == 0x2 || ((val14>>4) & 0x3) == 0x0));
+
+		if (ret) printk("perfmon: failure check_case1\n");
+	}
+
+	return ret ? -EINVAL : 0;
+}

arch/ia64/kernel/process.c

@@ -325,6 +325,11 @@ copy_thread (int nr, unsigned long clone_flags,
 
 	/* copy parts of thread_struct: */
 	p->thread.ksp = (unsigned long) child_stack - 16;
+
+	/* stop some PSR bits from being inherited: */
+	child_ptregs->cr_ipsr =  ((child_ptregs->cr_ipsr | IA64_PSR_BITS_TO_SET)
+				  & ~IA64_PSR_BITS_TO_CLEAR);
+
 	/*
 	 * NOTE: The calling convention considers all floating point
 	 * registers in the high partition (fph) to be scratch.  Since

arch/ia64/kernel/setup.c

@@ -455,10 +455,10 @@ c_stop (struct seq_file *m, void *v)
 }
 
 struct seq_operations cpuinfo_op = {
-	start:	c_start,
-	next:	c_next,
-	stop:	c_stop,
-	show:	show_cpuinfo
+	.start =	c_start,
+	.next =		c_next,
+	.stop =		c_stop,
+	.show =		show_cpuinfo
 };
 
 void
@@ -542,7 +542,18 @@ cpu_init (void)
 	extern char __per_cpu_end[];
 	int cpu = smp_processor_id();
 
+	if (__per_cpu_end - __per_cpu_start > PAGE_SIZE)
+		panic("Per-cpu data area too big! (%Zu > %Zu)",
+		      __per_cpu_end - __per_cpu_start, PAGE_SIZE);
+
+	/*
+	 * On the BSP, the page allocator isn't initialized by the time we get here.  On
+	 * the APs, the bootmem allocator is no longer available...
+	 */
+	if (cpu == 0)
 		my_cpu_data = alloc_bootmem_pages(__per_cpu_end - __per_cpu_start);
+	else
+		my_cpu_data = (void *) get_free_page(GFP_KERNEL);
 	memcpy(my_cpu_data, __phys_per_cpu_start, __per_cpu_end - __per_cpu_start);
 	__per_cpu_offset[cpu] = (char *) my_cpu_data - __per_cpu_start;
 	my_cpu_info = my_cpu_data + ((char *) &cpu_info - __per_cpu_start);

arch/ia64/kernel/signal.c

@@ -146,6 +146,7 @@ copy_siginfo_to_user (siginfo_t *to, siginfo_t *from)
 	if (from->si_code < 0) {
 		if (__copy_to_user(to, from, sizeof(siginfo_t)))
 			return -EFAULT;
+		return 0;
 	} else {
 		int err;
 

arch/ia64/kernel/smpboot.c

@@ -425,7 +425,7 @@ do_boot_cpu (int sapicid)
 
 	task_for_booting_cpu = idle;
 
-	Dprintk("Sending wakeup vector %u to AP 0x%x/0x%x.\n", ap_wakeup_vector, cpu, sapicid);
+	Dprintk("Sending wakeup vector %lu to AP 0x%x/0x%x.\n", ap_wakeup_vector, cpu, sapicid);
 
 	platform_send_ipi(cpu, ap_wakeup_vector, IA64_IPI_DM_INT, 0);
 
@@ -537,7 +537,7 @@ smp_boot_cpus (void)
 
 		printk("Before bogomips.\n");
 		if (!cpucount) {
-			printk(KERN_ERR "Error: only one processor found.\n");
+			printk(KERN_WARNING "Warning: only one processor found.\n");
 		} else {
 			unsigned long bogosum = 0;
   			for (cpu = 0; cpu < NR_CPUS; cpu++)

arch/ia64/kernel/time.c

@@ -41,10 +41,12 @@ do_profile (unsigned long ip)
 	extern unsigned long prof_cpu_mask;
 	extern char _stext;
 
+	if (!prof_buffer)
+		return;
+
 	if (!((1UL << smp_processor_id()) & prof_cpu_mask))
 		return;
 
-	if (prof_buffer && current->pid) {
 	ip -= (unsigned long) &_stext;
 	ip >>= prof_shift;
 	/*
@@ -55,7 +57,6 @@ do_profile (unsigned long ip)
 		ip = prof_len - 1;
 
 	atomic_inc((atomic_t *) &prof_buffer[ip]);
-	}
 }
 
 /*
@@ -285,9 +286,9 @@ ia64_init_itm (void)
 }
 
 static struct irqaction timer_irqaction = {
-	handler:	timer_interrupt,
-	flags:		SA_INTERRUPT,
-	name:		"timer"
+	.handler =	timer_interrupt,
+	.flags =	SA_INTERRUPT,
+	.name =		"timer"
 };
 
 void __init

arch/ia64/kernel/traps.c

@@ -93,9 +93,9 @@ die (const char *str, struct pt_regs *regs, long err)
 		int lock_owner;
 		int lock_owner_depth;
 	} die = {
-		lock:			SPIN_LOCK_UNLOCKED,
-		lock_owner:		-1,
-		lock_owner_depth:	0
+		.lock =			SPIN_LOCK_UNLOCKED,
+		.lock_owner =		-1,
+		.lock_owner_depth =	0
 	};
 
 	if (die.lock_owner != smp_processor_id()) {
@@ -435,7 +435,7 @@ ia64_fault (unsigned long vector, unsigned long isr, unsigned long ifa,
 	unsigned long code, error = isr;
 	struct siginfo siginfo;
 	char buf[128];
-	int result;
+	int result, sig;
 	static const char *reason[] = {
 		"IA-64 Illegal Operation fault",
 		"IA-64 Privileged Operation fault",
@@ -479,6 +479,30 @@ ia64_fault (unsigned long vector, unsigned long isr, unsigned long ifa,
 		break;
 
 	      case 26: /* NaT Consumption */
+		if (user_mode(regs)) {
+			if (((isr >> 4) & 0xf) == 2) {
+				/* NaT page consumption */
+				sig = SIGSEGV;
+				code = SEGV_ACCERR;
+			} else {
+				/* register NaT consumption */
+				sig = SIGILL;
+				code = ILL_ILLOPN;
+			}
+			siginfo.si_signo = sig;
+			siginfo.si_code = code;
+			siginfo.si_errno = 0;
+			siginfo.si_addr = (void *) (regs->cr_iip + ia64_psr(regs)->ri);
+			siginfo.si_imm = vector;
+			siginfo.si_flags = __ISR_VALID;
+			siginfo.si_isr = isr;
+			force_sig_info(sig, &siginfo, current);
+			return;
+		} else if (done_with_exception(regs))
+			return;
+		sprintf(buf, "NaT consumption");
+		break;
+
 	      case 31: /* Unsupported Data Reference */
 		if (user_mode(regs)) {
 			siginfo.si_signo = SIGILL;
@@ -491,7 +515,7 @@ ia64_fault (unsigned long vector, unsigned long isr, unsigned long ifa,
 			force_sig_info(SIGILL, &siginfo, current);
 			return;
 		}
-		sprintf(buf, (vector == 26) ? "NaT consumption" : "Unsupported data reference");
+		sprintf(buf, "Unsupported data reference");
 		break;
 
 	      case 29: /* Debug */
@@ -508,16 +532,15 @@ ia64_fault (unsigned long vector, unsigned long isr, unsigned long ifa,
 			if (ia64_psr(regs)->is == 0)
 			  ifa = regs->cr_iip;
 #endif
-			siginfo.si_addr = (void *) ifa;
 			break;
-		      case 35: siginfo.si_code = TRAP_BRANCH; break;
-		      case 36: siginfo.si_code = TRAP_TRACE; break;
+		      case 35: siginfo.si_code = TRAP_BRANCH; ifa = 0; break;
+		      case 36: siginfo.si_code = TRAP_TRACE; ifa = 0; break;
 		}
 		siginfo.si_signo = SIGTRAP;
 		siginfo.si_errno = 0;
 		siginfo.si_flags = 0;
 		siginfo.si_isr = 0;
-		siginfo.si_addr = 0;
+		siginfo.si_addr = (void *) ifa;
 		siginfo.si_imm = 0;
 		force_sig_info(SIGTRAP, &siginfo, current);
 		return;

arch/ia64/kernel/unwind.c

@@ -140,13 +140,13 @@ static struct {
 	} stat;
 # endif
 } unw = {
-	tables: &unw.kernel_table,
-	lock: SPIN_LOCK_UNLOCKED,
-	save_order: {
+	.tables = &unw.kernel_table,
+	.lock = SPIN_LOCK_UNLOCKED,
+	.save_order = {
 		UNW_REG_RP, UNW_REG_PFS, UNW_REG_PSP, UNW_REG_PR,
 		UNW_REG_UNAT, UNW_REG_LC, UNW_REG_FPSR, UNW_REG_PRI_UNAT_GR
 	},
-	preg_index: {
+	.preg_index = {
 		struct_offset(struct unw_frame_info, pri_unat_loc)/8,	/* PRI_UNAT_GR */
 		struct_offset(struct unw_frame_info, pri_unat_loc)/8,	/* PRI_UNAT_MEM */
 		struct_offset(struct unw_frame_info, bsp_loc)/8,
@@ -189,9 +189,9 @@ static struct {
 		struct_offset(struct unw_frame_info, fr_loc[30 - 16])/8,
 		struct_offset(struct unw_frame_info, fr_loc[31 - 16])/8,
 	},
-	hash : { [0 ... UNW_HASH_SIZE - 1] = -1 },
+	.hash = { [0 ... UNW_HASH_SIZE - 1] = -1 },
 #if UNW_DEBUG
-	preg_name: {
+	.preg_name = {
 		"pri_unat_gr", "pri_unat_mem", "bsp", "bspstore", "ar.pfs", "ar.rnat", "psp", "rp",
 		"r4", "r5", "r6", "r7",
 		"ar.unat", "pr", "ar.lc", "ar.fpsr",
@@ -634,8 +634,8 @@ alloc_spill_area (unsigned long *offp, unsigned long regsize,
 	for (reg = hi; reg >= lo; --reg) {
 		if (reg->where == UNW_WHERE_SPILL_HOME) {
 			reg->where = UNW_WHERE_PSPREL;
-			reg->val = 0x10 - *offp;
-			*offp += regsize;
+			*offp -= regsize;
+			reg->val = *offp;
 		}
 	}
 }
@@ -814,7 +814,8 @@ desc_frgr_mem (unsigned char grmask, unw_word frmask, struct unw_state_record *s
 	}
 	for (i = 0; i < 20; ++i) {
 		if ((frmask & 1) != 0) {
-			set_reg(sr->curr.reg + UNW_REG_F2 + i, UNW_WHERE_SPILL_HOME,
+			int base = (i < 4) ? UNW_REG_F2 : UNW_REG_F16 - 4;
+			set_reg(sr->curr.reg + base + i, UNW_WHERE_SPILL_HOME,
 				sr->region_start + sr->region_len - 1, 0);
 			sr->any_spills = 1;
 		}

arch/ia64/lib/Makefile

@@ -9,12 +9,12 @@ export-objs := io.o swiotlb.o
 obj-y := __divsi3.o __udivsi3.o __modsi3.o __umodsi3.o					\
 	__divdi3.o __udivdi3.o __moddi3.o __umoddi3.o					\
 	checksum.o clear_page.o csum_partial_copy.o copy_page.o				\
-	copy_user.o clear_user.o strncpy_from_user.o strlen_user.o strnlen_user.o	\
+	clear_user.o strncpy_from_user.o strlen_user.o strnlen_user.o	\
 	flush.o io.o ip_fast_csum.o do_csum.o						\
-	memcpy.o memset.o strlen.o swiotlb.o
+	memset.o strlen.o swiotlb.o
 
-obj-$(CONFIG_ITANIUM) += copy_page.o
-obj-$(CONFIG_MCKINLEY) += copy_page_mck.o
+obj-$(CONFIG_ITANIUM) += copy_page.o copy_user.o memcpy.o
+obj-$(CONFIG_MCKINLEY) += copy_page_mck.o memcpy_mck.o
 
 IGNORE_FLAGS_OBJS =	__divsi3.o __udivsi3.o __modsi3.o __umodsi3.o \
 			__divdi3.o __udivdi3.o __moddi3.o __umoddi3.o

arch/ia64/lib/copy_user.S

@@ -237,15 +237,17 @@ GLOBAL_ENTRY(__copy_user)
 .copy_user_bit##rshift:						\
 1:								\
 	EX(.failure_out,(EPI) st8 [dst1]=tmp,8);		\
-(EPI_1) shrp tmp=val1[PIPE_DEPTH-3],val1[PIPE_DEPTH-2],rshift;	\
-	EX(3f,(p16) ld8 val1[0]=[src1],8);			\
+(EPI_1) shrp tmp=val1[PIPE_DEPTH-2],val1[PIPE_DEPTH-1],rshift;	\
+	EX(3f,(p16) ld8 val1[1]=[src1],8);			\
+(p16)	mov val1[0]=r0;						\
 	br.ctop.dptk 1b;					\
 	;;							\
 	br.cond.sptk.many .diff_align_do_tail;			\
 2:								\
 (EPI)	st8 [dst1]=tmp,8;					\
-(EPI_1)	shrp tmp=val1[PIPE_DEPTH-3],val1[PIPE_DEPTH-2],rshift;	\
+(EPI_1)	shrp tmp=val1[PIPE_DEPTH-2],val1[PIPE_DEPTH-1],rshift;	\
 3:								\
+(p16)	mov val1[1]=r0;						\
 (p16)	mov val1[0]=r0;						\
 	br.ctop.dptk 2b;					\
 	;;							\

arch/ia64/lib/io.c

@@ -87,6 +87,12 @@ ia64_outl (unsigned int val, unsigned long port)
 	__ia64_outl(val, port);
 }
 
+void
+ia64_mmiob (void)
+{
+	__ia64_mmiob();
+}
+
 /* define aliases: */
 
 asm (".global __ia64_inb, __ia64_inw, __ia64_inl");
@@ -99,4 +105,7 @@ asm ("__ia64_outb = ia64_outb");
 asm ("__ia64_outw = ia64_outw");
 asm ("__ia64_outl = ia64_outl");
 
+asm (".global __ia64_mmiob");
+asm ("__ia64_mmiob = ia64_mmiob");
+
 #endif /* CONFIG_IA64_GENERIC */

arch/ia64/lib/swiotlb.c

@@ -415,18 +415,20 @@ int
 swiotlb_map_sg (struct pci_dev *hwdev, struct scatterlist *sg, int nelems, int direction)
 {
 	void *addr;
+	unsigned long pci_addr;
 	int i;
 
 	if (direction == PCI_DMA_NONE)
 		BUG();
 
 	for (i = 0; i < nelems; i++, sg++) {
-		sg->orig_address = SG_ENT_VIRT_ADDRESS(sg);
-		if ((SG_ENT_PHYS_ADDRESS(sg) & ~hwdev->dma_mask) != 0) {
-			addr = map_single(hwdev, sg->orig_address, sg->length, direction);
-			sg->page = virt_to_page(addr);
-			sg->offset = (u64) addr & ~PAGE_MASK;
-		}
+		addr = SG_ENT_VIRT_ADDRESS(sg);
+		pci_addr = virt_to_phys(addr);
+		if ((pci_addr & ~hwdev->dma_mask) != 0)
+			sg->dma_address = map_single(hwdev, addr, sg->length, direction);
+		else
+			sg->dma_address = pci_addr;
+		sg->dma_length = sg->length;
 	}
 	return nelems;
 }
@@ -444,12 +446,10 @@ swiotlb_unmap_sg (struct pci_dev *hwdev, struct scatterlist *sg, int nelems, int
 		BUG();
 
 	for (i = 0; i < nelems; i++, sg++)
-		if (sg->orig_address != SG_ENT_VIRT_ADDRESS(sg)) {
-			unmap_single(hwdev, SG_ENT_VIRT_ADDRESS(sg), sg->length, direction);
-			sg->page = virt_to_page(sg->orig_address);
-			sg->offset = (u64) sg->orig_address & ~PAGE_MASK;
-		} else if (direction == PCI_DMA_FROMDEVICE)
-			mark_clean(SG_ENT_VIRT_ADDRESS(sg), sg->length);
+		if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg))
+			unmap_single(hwdev, sg->dma_address, sg->dma_length, direction);
+		else if (direction == PCI_DMA_FROMDEVICE)
+			mark_clean(SG_ENT_VIRT_ADDRESS(sg), sg->dma_length);
 }
 
 /*
@@ -468,14 +468,14 @@ swiotlb_sync_sg (struct pci_dev *hwdev, struct scatterlist *sg, int nelems, int
 		BUG();
 
 	for (i = 0; i < nelems; i++, sg++)
-		if (sg->orig_address != SG_ENT_VIRT_ADDRESS(sg))
-			sync_single(hwdev, SG_ENT_VIRT_ADDRESS(sg), sg->length, direction);
+		if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg))
+			sync_single(hwdev, sg->dma_address, sg->dma_length, direction);
 }
 
 unsigned long
 swiotlb_dma_address (struct scatterlist *sg)
 {
-	return SG_ENT_PHYS_ADDRESS(sg);
+	return sg->dma_address;
 }
 
 /*

arch/ia64/mm/init.c

@@ -10,6 +10,7 @@
 
 #include <linux/bootmem.h>
 #include <linux/mm.h>
+#include <linux/personality.h>
 #include <linux/reboot.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
@@ -68,10 +69,9 @@ ia64_init_addr_space (void)
 	struct vm_area_struct *vma;
 
 	/*
-	 * If we're out of memory and kmem_cache_alloc() returns NULL,
-	 * we simply ignore the problem.  When the process attempts to
-	 * write to the register backing store for the first time, it
-	 * will get a SEGFAULT in this case.
+	 * If we're out of memory and kmem_cache_alloc() returns NULL, we simply ignore
+	 * the problem.  When the process attempts to write to the register backing store
+	 * for the first time, it will get a SEGFAULT in this case.
 	 */
 	vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
 	if (vma) {
@@ -86,6 +86,19 @@ ia64_init_addr_space (void)
 		vma->vm_private_data = NULL;
 		insert_vm_struct(current->mm, vma);
 	}
+
+	/* map NaT-page at address zero to speed up speculative dereferencing of NULL: */
+	if (!(current->personality & MMAP_PAGE_ZERO)) {
+		vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
+		if (vma) {
+			memset(vma, 0, sizeof(*vma));
+			vma->vm_mm = current->mm;
+			vma->vm_end = PAGE_SIZE;
+			vma->vm_page_prot = __pgprot(pgprot_val(PAGE_READONLY) | _PAGE_MA_NAT);
+			vma->vm_flags = VM_READ | VM_MAYREAD | VM_IO | VM_RESERVED;
+			insert_vm_struct(current->mm, vma);
+		}
+	}
 }
 
 void

arch/ia64/mm/tlb.c

@@ -35,10 +35,10 @@
 		1 << _PAGE_SIZE_4K )
 
 struct ia64_ctx ia64_ctx = {
-	lock:	SPIN_LOCK_UNLOCKED,
-	next:	1,
-	limit:	(1 << 15) - 1,		/* start out with the safe (architected) limit */
-	max_ctx: ~0U
+	.lock =		SPIN_LOCK_UNLOCKED,
+	.next =		1,
+	.limit =	(1 << 15) - 1,		/* start out with the safe (architected) limit */
+	.max_ctx =	~0U
 };
 
 /*
@@ -49,6 +49,7 @@ wrap_mmu_context (struct mm_struct *mm)
 {
 	unsigned long tsk_context, max_ctx = ia64_ctx.max_ctx;
 	struct task_struct *tsk;
+	int i;
 
 	if (ia64_ctx.next > max_ctx)
 		ia64_ctx.next = 300;	/* skip daemons */
@@ -77,7 +78,11 @@ wrap_mmu_context (struct mm_struct *mm)
 			ia64_ctx.limit = tsk_context;
 	}
 	read_unlock(&tasklist_lock);
-	flush_tlb_all();
+	/* can't call flush_tlb_all() here because of race condition with O(1) scheduler [EF] */
+	for (i = 0; i < smp_num_cpus; ++i)
+		if (i != smp_processor_id())
+			per_cpu(ia64_need_tlb_flush, i) = 1;
+	__flush_tlb_all();
 }
 
 void

arch/ia64/sn/io/ifconfig_net.c

@@ -279,9 +279,9 @@ static int ifconfig_net_ioctl(struct inode * inode, struct file * file,
 }
 
 struct file_operations ifconfig_net_fops = {
-	ioctl:ifconfig_net_ioctl,	/* ioctl */
-	open:ifconfig_net_open,		/* open */
-	release:ifconfig_net_close	/* release */
+	.ioctl =ifconfig_net_ioctl,	/* ioctl */
+	.open =ifconfig_net_open,		/* open */
+	.release =ifconfig_net_close	/* release */
 };
 
 

arch/ia64/sn/io/pciba.c

@@ -210,31 +210,31 @@ static void dump_allocations(struct list_head * dalp);
 
 /* file operations for each type of node */
 static struct file_operations rom_fops = {
-	owner:		THIS_MODULE,
-	mmap:		rom_mmap,
-	open:		generic_open,
-	release:	rom_release
+	.owner =	THIS_MODULE,
+	.mmap =		rom_mmap,
+	.open =		generic_open,
+	.release =	rom_release
 };
  
 
 static struct file_operations base_fops = {
-	owner:		THIS_MODULE,
-	mmap:		base_mmap,
-	open:		generic_open
+	.owner =	THIS_MODULE,
+	.mmap =		base_mmap,
+	.open =		generic_open
 };
 
 
 static struct file_operations config_fops = {
-	owner:		THIS_MODULE,
-	ioctl:		config_ioctl,
-	open:		generic_open
+	.owner =	THIS_MODULE,
+	.ioctl =	config_ioctl,
+	.open =		generic_open
 };	
 
 static struct file_operations dma_fops = {
-	owner:		THIS_MODULE,
-	ioctl:		dma_ioctl,
-	mmap:		dma_mmap,
-	open:		generic_open
+	.owner =	THIS_MODULE,
+	.ioctl =	dma_ioctl,
+	.mmap =		dma_mmap,
+	.open =		generic_open
 };	
 
 

arch/ia64/sn/io/sn1/hubcounters.c

@@ -24,7 +24,7 @@ extern void hubni_error_handler(char *, int); /* huberror.c */
 
 static int hubstats_ioctl(struct inode *, struct file *, unsigned int, unsigned long);
 struct file_operations hub_mon_fops = {
-        ioctl:          hubstats_ioctl,
+        .ioctl =        hubstats_ioctl,
 };
 
 #define HUB_CAPTURE_TICKS	(2 * HZ)

arch/ia64/sn/io/sn1/pcibr.c

@@ -307,22 +307,22 @@ extern void free_pciio_dmamap(pcibr_dmamap_t);
  * appropriate function name below.
  */
 struct file_operations pcibr_fops = {
-        owner:  THIS_MODULE,
-        llseek: NULL,
-        read: NULL,
-        write: NULL,
-        readdir: NULL,
-        poll: NULL,
-        ioctl: NULL,
-        mmap: NULL,
-        open: NULL,
-        flush: NULL,
-        release: NULL,
-        fsync: NULL,
-        fasync: NULL,
-        lock: NULL,
-        readv: NULL,
-        writev: NULL
+        .owner = THIS_MODULE,
+        .llseek = NULL,
+        .read = NULL,
+        .write = NULL,
+        .readdir = NULL,
+        .poll = NULL,
+        .ioctl = NULL,
+        .mmap = NULL,
+        .open = NULL,
+        .flush = NULL,
+        .release = NULL,
+        .fsync = NULL,
+        .fasync = NULL,
+        .lock = NULL,
+        .readv = NULL,
+        .writev = NULL
 };
 
 extern devfs_handle_t hwgraph_root;

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

@@ -64,22 +64,22 @@ int                     pcibr_devflag = D_MP;
  * appropriate function name below.
  */
 struct file_operations pcibr_fops = {
-	owner:  THIS_MODULE,
-	llseek: NULL,
-	read: NULL,
-	write: NULL,
-	readdir: NULL,
-	poll: NULL,
-	ioctl: NULL,
-	mmap: NULL,
-	open: NULL,
-	flush: NULL,
-	release: NULL,
-	fsync: NULL,
-	fasync: NULL,
-	lock: NULL,
-	readv: NULL,
-	writev: NULL
+	.owner =THIS_MODULE,
+	.llseek = NULL,
+	.read = NULL,
+	.write = NULL,
+	.readdir = NULL,
+	.poll = NULL,
+	.ioctl = NULL,
+	.mmap = NULL,
+	.open = NULL,
+	.flush = NULL,
+	.release = NULL,
+	.fsync = NULL,
+	.fasync = NULL,
+	.lock = NULL,
+	.readv = NULL,
+	.writev = NULL
 };
 
 #ifdef LATER

arch/ia64/sn/kernel/setup.c

@@ -109,14 +109,14 @@ irqpda_t		*irqpdaindr[NR_CPUS];
  * VGA color display.
  */
 struct screen_info sn1_screen_info = {
-	orig_x:			 0,
-	orig_y:			 0,
-	orig_video_mode:	 3,
-	orig_video_cols:	80,
-	orig_video_ega_bx:	 3,
-	orig_video_lines:	25,
-	orig_video_isVGA:	 1,
-	orig_video_points:	16
+	.orig_x =		 0,
+	.orig_y =		 0,
+	.orig_video_mode =	 3,
+	.orig_video_cols =	80,
+	.orig_video_ega_bx =	 3,
+	.orig_video_lines =	25,
+	.orig_video_isVGA =	 1,
+	.orig_video_points =	16
 };
 
 /*
@@ -170,9 +170,9 @@ early_sn1_setup(void)
 #ifdef NOT_YET_CONFIG_IA64_MCA
 extern void ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs);
 static struct irqaction mca_cpe_irqaction = { 
-	handler:    ia64_mca_cpe_int_handler,
-	flags:      SA_INTERRUPT,
-	name:       "cpe_hndlr"
+	.handler =  ia64_mca_cpe_int_handler,
+	.flags =    SA_INTERRUPT,
+	.name =     "cpe_hndlr"
 };
 #endif
 #ifdef CONFIG_IA64_MCA

include/asm-ia64/bitops.h

@@ -326,7 +326,7 @@ ia64_fls (unsigned long x)
 	return exp - 0xffff;
 }
 
-static int
+static inline int
 fls (int x)
 {
 	return ia64_fls((unsigned int) x);

include/asm-ia64/delay.h

@@ -53,7 +53,7 @@ ia64_get_itc (void)
 
 	__asm__ __volatile__("mov %0=ar.itc" : "=r"(result) :: "memory");
 #ifdef CONFIG_ITANIUM
-	while (unlikely ((__s32) result == -1)
+	while (unlikely((__s32) result == -1))
 		__asm__ __volatile__("mov %0=ar.itc" : "=r"(result) :: "memory");
 #endif
 	return result;

include/asm-ia64/keyboard.h

@@ -16,6 +16,7 @@
 #define KEYBOARD_IRQ			isa_irq_to_vector(1)
 #define DISABLE_KBD_DURING_INTERRUPTS	0
 
+extern unsigned char acpi_kbd_controller_present;
 extern int pckbd_setkeycode(unsigned int scancode, unsigned int keycode);
 extern int pckbd_getkeycode(unsigned int scancode);
 extern int pckbd_pretranslate(unsigned char scancode, char raw_mode);
@@ -26,6 +27,7 @@ extern void pckbd_leds(unsigned char leds);
 extern void pckbd_init_hw(void);
 extern unsigned char pckbd_sysrq_xlate[128];
 
+#define kbd_controller_present() acpi_kbd_controller_present
 #define kbd_setkeycode		pckbd_setkeycode
 #define kbd_getkeycode		pckbd_getkeycode
 #define kbd_pretranslate	pckbd_pretranslate

include/asm-ia64/kregs.h

@@ -64,6 +64,15 @@
 #define IA64_PSR_RI_BIT		41
 #define IA64_PSR_ED_BIT		43
 #define IA64_PSR_BN_BIT		44
+#define IA64_PSR_IA	(__IA64_UL(1) << IA64_PSR_IA_BIT)
+
+/* A mask of PSR bits that we generally don't want to inherit across a clone2() or an
+   execve().  Only list flags here that need to be cleared/set for BOTH clone2() and
+   execve().  */
+#define IA64_PSR_BITS_TO_CLEAR	(IA64_PSR_MFL | IA64_PSR_MFH | IA64_PSR_DB | IA64_PSR_LP | \
+				 IA64_PSR_TB  | IA64_PSR_ID  | IA64_PSR_DA | IA64_PSR_DD | \
+				 IA64_PSR_SS  | IA64_PSR_ED  | IA64_PSR_IA)
+#define IA64_PSR_BITS_TO_SET	(IA64_PSR_DFH)
 
 #define IA64_PSR_BE	(__IA64_UL(1) << IA64_PSR_BE_BIT)
 #define IA64_PSR_UP	(__IA64_UL(1) << IA64_PSR_UP_BIT)

include/asm-ia64/machvec.h

@@ -210,6 +210,7 @@ extern ia64_mv_pci_unmap_sg swiotlb_unmap_sg;
 extern ia64_mv_pci_dma_sync_single swiotlb_sync_single;
 extern ia64_mv_pci_dma_sync_sg swiotlb_sync_sg;
 extern ia64_mv_pci_dma_address swiotlb_dma_address;
+extern ia64_mv_pci_dma_supported swiotlb_pci_dma_supported;
 
 /*
  * Define default versions so we can extend machvec for new platforms without having

include/asm-ia64/machvec_init.h

@@ -16,6 +16,7 @@ extern ia64_mv_inl_t __ia64_inl;
 extern ia64_mv_outb_t __ia64_outb;
 extern ia64_mv_outw_t __ia64_outw;
 extern ia64_mv_outl_t __ia64_outl;
+extern ia64_mv_mmiob_t __ia64_mmiob;
 
 #define MACHVEC_HELPER(name)									\
  struct ia64_machine_vector machvec_##name __attribute__ ((unused, __section__ (".machvec")))	\

include/asm-ia64/mmu_context.h

@@ -2,8 +2,8 @@
 #define _ASM_IA64_MMU_CONTEXT_H
 
 /*
- * Copyright (C) 1998-2001 Hewlett-Packard Co
- * Copyright (C) 1998-2001 David Mosberger-Tang <davidm@hpl.hp.com>
+ * Copyright (C) 1998-2002 Hewlett-Packard Co
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
  */
 
 /*
@@ -13,8 +13,6 @@
  * consider the region number when performing a TLB lookup, we need to assign a unique
  * region id to each region in a process.  We use the least significant three bits in a
  * region id for this purpose.
- *
- * Copyright (C) 1998-2001 David Mosberger-Tang <davidm@hpl.hp.com>
  */
 
 #define IA64_REGION_ID_KERNEL	0 /* the kernel's region id (tlb.c depends on this being 0) */
@@ -23,6 +21,8 @@
 
 # ifndef __ASSEMBLY__
 
+#include <linux/compiler.h>
+#include <linux/percpu.h>
 #include <linux/sched.h>
 #include <linux/spinlock.h>
 
@@ -36,6 +36,7 @@ struct ia64_ctx {
 };
 
 extern struct ia64_ctx ia64_ctx;
+extern u8 ia64_need_tlb_flush __per_cpu_data;
 
 extern void wrap_mmu_context (struct mm_struct *mm);
 
@@ -44,6 +45,23 @@ enter_lazy_tlb (struct mm_struct *mm, struct task_struct *tsk, unsigned cpu)
 {
 }
 
+/*
+ * When the context counter wraps around all TLBs need to be flushed because an old
+ * context number might have been reused. This is signalled by the ia64_need_tlb_flush
+ * per-CPU variable, which is checked in the routine below. Called by activate_mm().
+ * <efocht@ess.nec.de>
+ */
+static inline void
+delayed_tlb_flush (void)
+{
+	extern void __flush_tlb_all (void);
+
+	if (unlikely(ia64_need_tlb_flush)) {
+		__flush_tlb_all();
+		ia64_need_tlb_flush = 0;
+	}
+}
+
 static inline void
 get_new_mmu_context (struct mm_struct *mm)
 {
@@ -54,7 +72,6 @@ get_new_mmu_context (struct mm_struct *mm)
 		mm->context = ia64_ctx.next++;
 	}
 	spin_unlock(&ia64_ctx.lock);
-
 }
 
 static inline void
@@ -109,6 +126,8 @@ reload_context (struct mm_struct *mm)
 static inline void
 activate_mm (struct mm_struct *prev, struct mm_struct *next)
 {
+	delayed_tlb_flush();
+
 	/*
 	 * We may get interrupts here, but that's OK because interrupt
 	 * handlers cannot touch user-space.

include/asm-ia64/offsets.h

-#ifndef _ASM_IA64_OFFSETS_H
-#define _ASM_IA64_OFFSETS_H
-/*
- * DO NOT MODIFY
- *
- * This file was generated by arch/ia64/tools/print_offsets.awk.
- *
- */
-#define IA64_TASK_SIZE			3952	/* 0xf70 */
-#define IA64_THREAD_INFO_SIZE		32	/* 0x20 */
-#define IA64_PT_REGS_SIZE		400	/* 0x190 */
-#define IA64_SWITCH_STACK_SIZE		560	/* 0x230 */
-#define IA64_SIGINFO_SIZE		128	/* 0x80 */
-#define IA64_CPU_SIZE			224	/* 0xe0 */
-#define SIGFRAME_SIZE			2816	/* 0xb00 */
-#define UNW_FRAME_INFO_SIZE		448	/* 0x1c0 */
-
-#define IA64_TASK_THREAD_KSP_OFFSET	1496	/* 0x5d8 */
-#define IA64_PT_REGS_CR_IPSR_OFFSET	0	/* 0x0 */
-#define IA64_PT_REGS_CR_IIP_OFFSET	8	/* 0x8 */
-#define IA64_PT_REGS_CR_IFS_OFFSET	16	/* 0x10 */
-#define IA64_PT_REGS_AR_UNAT_OFFSET	24	/* 0x18 */
-#define IA64_PT_REGS_AR_PFS_OFFSET	32	/* 0x20 */
-#define IA64_PT_REGS_AR_RSC_OFFSET	40	/* 0x28 */
-#define IA64_PT_REGS_AR_RNAT_OFFSET	48	/* 0x30 */
-#define IA64_PT_REGS_AR_BSPSTORE_OFFSET	56	/* 0x38 */
-#define IA64_PT_REGS_PR_OFFSET		64	/* 0x40 */
-#define IA64_PT_REGS_B6_OFFSET		72	/* 0x48 */
-#define IA64_PT_REGS_LOADRS_OFFSET	80	/* 0x50 */
-#define IA64_PT_REGS_R1_OFFSET		88	/* 0x58 */
-#define IA64_PT_REGS_R2_OFFSET		96	/* 0x60 */
-#define IA64_PT_REGS_R3_OFFSET		104	/* 0x68 */
-#define IA64_PT_REGS_R12_OFFSET		112	/* 0x70 */
-#define IA64_PT_REGS_R13_OFFSET		120	/* 0x78 */
-#define IA64_PT_REGS_R14_OFFSET		128	/* 0x80 */
-#define IA64_PT_REGS_R15_OFFSET		136	/* 0x88 */
-#define IA64_PT_REGS_R8_OFFSET		144	/* 0x90 */
-#define IA64_PT_REGS_R9_OFFSET		152	/* 0x98 */
-#define IA64_PT_REGS_R10_OFFSET		160	/* 0xa0 */
-#define IA64_PT_REGS_R11_OFFSET		168	/* 0xa8 */
-#define IA64_PT_REGS_R16_OFFSET		176	/* 0xb0 */
-#define IA64_PT_REGS_R17_OFFSET		184	/* 0xb8 */
-#define IA64_PT_REGS_R18_OFFSET		192	/* 0xc0 */
-#define IA64_PT_REGS_R19_OFFSET		200	/* 0xc8 */
-#define IA64_PT_REGS_R20_OFFSET		208	/* 0xd0 */
-#define IA64_PT_REGS_R21_OFFSET		216	/* 0xd8 */
-#define IA64_PT_REGS_R22_OFFSET		224	/* 0xe0 */
-#define IA64_PT_REGS_R23_OFFSET		232	/* 0xe8 */
-#define IA64_PT_REGS_R24_OFFSET		240	/* 0xf0 */
-#define IA64_PT_REGS_R25_OFFSET		248	/* 0xf8 */
-#define IA64_PT_REGS_R26_OFFSET		256	/* 0x100 */
-#define IA64_PT_REGS_R27_OFFSET		264	/* 0x108 */
-#define IA64_PT_REGS_R28_OFFSET		272	/* 0x110 */
-#define IA64_PT_REGS_R29_OFFSET		280	/* 0x118 */
-#define IA64_PT_REGS_R30_OFFSET		288	/* 0x120 */
-#define IA64_PT_REGS_R31_OFFSET		296	/* 0x128 */
-#define IA64_PT_REGS_AR_CCV_OFFSET	304	/* 0x130 */
-#define IA64_PT_REGS_AR_FPSR_OFFSET	312	/* 0x138 */
-#define IA64_PT_REGS_B0_OFFSET		320	/* 0x140 */
-#define IA64_PT_REGS_B7_OFFSET		328	/* 0x148 */
-#define IA64_PT_REGS_F6_OFFSET		336	/* 0x150 */
-#define IA64_PT_REGS_F7_OFFSET		352	/* 0x160 */
-#define IA64_PT_REGS_F8_OFFSET		368	/* 0x170 */
-#define IA64_PT_REGS_F9_OFFSET		384	/* 0x180 */
-#define IA64_SWITCH_STACK_CALLER_UNAT_OFFSET 0	/* 0x0 */
-#define IA64_SWITCH_STACK_AR_FPSR_OFFSET 8	/* 0x8 */
-#define IA64_SWITCH_STACK_F2_OFFSET	16	/* 0x10 */
-#define IA64_SWITCH_STACK_F3_OFFSET	32	/* 0x20 */
-#define IA64_SWITCH_STACK_F4_OFFSET	48	/* 0x30 */
-#define IA64_SWITCH_STACK_F5_OFFSET	64	/* 0x40 */
-#define IA64_SWITCH_STACK_F10_OFFSET	80	/* 0x50 */
-#define IA64_SWITCH_STACK_F11_OFFSET	96	/* 0x60 */
-#define IA64_SWITCH_STACK_F12_OFFSET	112	/* 0x70 */
-#define IA64_SWITCH_STACK_F13_OFFSET	128	/* 0x80 */
-#define IA64_SWITCH_STACK_F14_OFFSET	144	/* 0x90 */
-#define IA64_SWITCH_STACK_F15_OFFSET	160	/* 0xa0 */
-#define IA64_SWITCH_STACK_F16_OFFSET	176	/* 0xb0 */
-#define IA64_SWITCH_STACK_F17_OFFSET	192	/* 0xc0 */
-#define IA64_SWITCH_STACK_F18_OFFSET	208	/* 0xd0 */
-#define IA64_SWITCH_STACK_F19_OFFSET	224	/* 0xe0 */
-#define IA64_SWITCH_STACK_F20_OFFSET	240	/* 0xf0 */
-#define IA64_SWITCH_STACK_F21_OFFSET	256	/* 0x100 */
-#define IA64_SWITCH_STACK_F22_OFFSET	272	/* 0x110 */
-#define IA64_SWITCH_STACK_F23_OFFSET	288	/* 0x120 */
-#define IA64_SWITCH_STACK_F24_OFFSET	304	/* 0x130 */
-#define IA64_SWITCH_STACK_F25_OFFSET	320	/* 0x140 */
-#define IA64_SWITCH_STACK_F26_OFFSET	336	/* 0x150 */
-#define IA64_SWITCH_STACK_F27_OFFSET	352	/* 0x160 */
-#define IA64_SWITCH_STACK_F28_OFFSET	368	/* 0x170 */
-#define IA64_SWITCH_STACK_F29_OFFSET	384	/* 0x180 */
-#define IA64_SWITCH_STACK_F30_OFFSET	400	/* 0x190 */
-#define IA64_SWITCH_STACK_F31_OFFSET	416	/* 0x1a0 */
-#define IA64_SWITCH_STACK_R4_OFFSET	432	/* 0x1b0 */
-#define IA64_SWITCH_STACK_R5_OFFSET	440	/* 0x1b8 */
-#define IA64_SWITCH_STACK_R6_OFFSET	448	/* 0x1c0 */
-#define IA64_SWITCH_STACK_R7_OFFSET	456	/* 0x1c8 */
-#define IA64_SWITCH_STACK_B0_OFFSET	464	/* 0x1d0 */
-#define IA64_SWITCH_STACK_B1_OFFSET	472	/* 0x1d8 */
-#define IA64_SWITCH_STACK_B2_OFFSET	480	/* 0x1e0 */
-#define IA64_SWITCH_STACK_B3_OFFSET	488	/* 0x1e8 */
-#define IA64_SWITCH_STACK_B4_OFFSET	496	/* 0x1f0 */
-#define IA64_SWITCH_STACK_B5_OFFSET	504	/* 0x1f8 */
-#define IA64_SWITCH_STACK_AR_PFS_OFFSET	512	/* 0x200 */
-#define IA64_SWITCH_STACK_AR_LC_OFFSET	520	/* 0x208 */
-#define IA64_SWITCH_STACK_AR_UNAT_OFFSET 528	/* 0x210 */
-#define IA64_SWITCH_STACK_AR_RNAT_OFFSET 536	/* 0x218 */
-#define IA64_SWITCH_STACK_AR_BSPSTORE_OFFSET 544	/* 0x220 */
-#define IA64_SWITCH_STACK_PR_OFFSET	552	/* 0x228 */
-#define IA64_SIGCONTEXT_IP_OFFSET	40	/* 0x28 */
-#define IA64_SIGCONTEXT_AR_BSP_OFFSET	72	/* 0x48 */
-#define IA64_SIGCONTEXT_AR_FPSR_OFFSET	104	/* 0x68 */
-#define IA64_SIGCONTEXT_AR_RNAT_OFFSET	80	/* 0x50 */
-#define IA64_SIGCONTEXT_AR_UNAT_OFFSET	96	/* 0x60 */
-#define IA64_SIGCONTEXT_B0_OFFSET	136	/* 0x88 */
-#define IA64_SIGCONTEXT_CFM_OFFSET	48	/* 0x30 */
-#define IA64_SIGCONTEXT_FLAGS_OFFSET	0	/* 0x0 */
-#define IA64_SIGCONTEXT_FR6_OFFSET	560	/* 0x230 */
-#define IA64_SIGCONTEXT_PR_OFFSET	128	/* 0x80 */
-#define IA64_SIGCONTEXT_R12_OFFSET	296	/* 0x128 */
-#define IA64_SIGCONTEXT_RBS_BASE_OFFSET	2512	/* 0x9d0 */
-#define IA64_SIGCONTEXT_LOADRS_OFFSET	2520	/* 0x9d8 */
-#define IA64_SIGFRAME_ARG0_OFFSET	0	/* 0x0 */
-#define IA64_SIGFRAME_ARG1_OFFSET	8	/* 0x8 */
-#define IA64_SIGFRAME_ARG2_OFFSET	16	/* 0x10 */
-#define IA64_SIGFRAME_HANDLER_OFFSET	24	/* 0x18 */
-#define IA64_SIGFRAME_SIGCONTEXT_OFFSET	160	/* 0xa0 */
-#define IA64_CLONE_VFORK		16384	/* 0x4000 */
-#define IA64_CLONE_VM			256	/* 0x100 */
-
-#endif /* _ASM_IA64_OFFSETS_H */

include/asm-ia64/pci.h

@@ -90,7 +90,7 @@ pcibios_penalize_isa_irq (int irq)
 /* Return the index of the PCI controller for device PDEV. */
 #define pci_controller_num(PDEV)	(0)
 
-#define sg_dma_len(sg)		((sg)->length)
+#define sg_dma_len(sg)		((sg)->dma_length)
 
 #define HAVE_PCI_MMAP
 extern int pci_mmap_page_range (struct pci_dev *dev, struct vm_area_struct *vma,

include/asm-ia64/perfmon.h

@@ -172,9 +172,8 @@ extern int  pfm_use_debug_registers(struct task_struct *);
 extern int  pfm_release_debug_registers(struct task_struct *);
 extern int  pfm_cleanup_smpl_buf(struct task_struct *);
 extern void pfm_syst_wide_update_task(struct task_struct *, int);
-extern void pfm_ovfl_block_reset (void);
-
-extern int pfm_syst_wide;
+extern void pfm_ovfl_block_reset(void);
+extern void perfmon_init_percpu(void);
 
 #endif /* __KERNEL__ */
 

include/asm-ia64/processor.h

@@ -270,12 +270,8 @@ struct thread_struct {
 
 #define start_thread(regs,new_ip,new_sp) do {							\
 	set_fs(USER_DS);									\
-	ia64_psr(regs)->dfh = 1;	/* disable fph */					\
-	ia64_psr(regs)->mfh = 0;	/* clear mfh */						\
-	ia64_psr(regs)->cpl = 3;	/* set user mode */					\
-	ia64_psr(regs)->ri = 0;		/* clear return slot number */				\
-	ia64_psr(regs)->is = 0;		/* IA-64 instruction set */				\
-	ia64_psr(regs)->sp = 1;		/* enforce secure perfmon */				\
+	regs->cr_ipsr = ((regs->cr_ipsr | (IA64_PSR_BITS_TO_SET | IA64_PSR_CPL | IA64_PSR_SP))	\
+			 & ~(IA64_PSR_BITS_TO_CLEAR | IA64_PSR_RI | IA64_PSR_IS));		\
 	regs->cr_iip = new_ip;									\
 	regs->ar_rsc = 0xf;		/* eager mode, privilege level 3 */			\
 	regs->ar_rnat = 0;									\
@@ -284,7 +280,7 @@ struct thread_struct {
 	regs->loadrs = 0;									\
 	regs->r8 = current->mm->dumpable;	/* set "don't zap registers" flag */		\
 	regs->r12 = new_sp - 16;	/* allocate 16 byte scratch area */			\
-	if (!likely (current->mm->dumpable)) {					\
+	if (unlikely(!current->mm->dumpable)) {					\
 		/*										\
 		 * Zap scratch regs to avoid leaking bits between processes with different	\
 		 * uid/privileges.								\

include/asm-ia64/scatterlist.h

@@ -7,12 +7,12 @@
  */
 
 struct scatterlist {
-	char *orig_address;	/* for use by swiotlb */
-
-	/* These two are only valid if ADDRESS member of this struct is NULL.  */
 	struct page *page;
 	unsigned int offset;
 	unsigned int length;	/* buffer length */
+
+	dma_addr_t dma_address;
+	unsigned int dma_length;
 };
 
 #define ISA_DMA_THRESHOLD	(~0UL)

include/asm-ia64/softirq.h

@@ -8,6 +8,7 @@
  *	David Mosberger-Tang <davidm@hpl.hp.com>
  */
 #include <asm/hardirq.h>
+#include <linux/compiler.h>
 
 #define __local_bh_enable()	do { barrier(); really_local_bh_count()--; } while (0)
 

include/asm-ia64/system.h

@@ -13,6 +13,7 @@
  * Copyright (C) 1999 Don Dugger <don.dugger@intel.com>
  */
 #include <linux/config.h>
+#include <linux/percpu.h>
 
 #include <asm/kregs.h>
 #include <asm/page.h>
@@ -384,7 +385,8 @@ extern void ia64_save_extra (struct task_struct *task);
 extern void ia64_load_extra (struct task_struct *task);
 
 #if defined(CONFIG_SMP) && defined(CONFIG_PERFMON)
-# define PERFMON_IS_SYSWIDE() (local_cpu_data->pfm_syst_wide != 0)
+  extern int __per_cpu_data pfm_syst_wide;
+# define PERFMON_IS_SYSWIDE() (this_cpu(pfm_syst_wide) != 0)
 #else
 # define PERFMON_IS_SYSWIDE() (0)
 #endif

include/asm-ia64/tlb.h

-/* XXX fix me! */
+#ifndef _ASM_IA64_TLB_H
+#define _ASM_IA64_TLB_H
+/*
+ * Copyright (C) 2002 Hewlett-Packard Co
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
+ *
+ * This file was derived from asm-generic/tlb.h.
+ */
+/*
+ * Removing a translation from a page table (including TLB-shootdown) is a four-step
+ * procedure:
+ *
+ *	(1) Flush (virtual) caches --- ensures virtual memory is coherent with kernel memory
+ *	    (this is a no-op on ia64).
+ *	(2) Clear the relevant portions of the page-table
+ *	(3) Flush the TLBs --- ensures that stale content is gone from CPU TLBs
+ *	(4) Release the pages that were freed up in step (2).
+ *
+ * Note that the ordering of these steps is crucial to avoid races on MP machines.
+ *
+ * The Linux kernel defines several platform-specific hooks for TLB-shootdown.  When
+ * unmapping a portion of the virtual address space, these hooks are called according to
+ * the following template:
+ *
+ *	tlb <- tlb_gather_mmu(mm);		// start unmap for address space MM
+ *	{
+ *	  for each vma that needs a shootdown do {
+ *	    tlb_start_vma(tlb, vma);
+ *	      for each page-table-entry PTE that needs to be removed do {
+ *		tlb_remove_tlb_entry(tlb, pte, address);
+ *		if (pte refers to a normal page) {
+ *		  tlb_remove_page(tlb, page);
+ *		}
+ *	      }
+ *	    tlb_end_vma(tlb, vma);
+ *	  }
+ *	}
+ *	tlb_finish_mmu(tlb, start, end);	// finish unmap for address space MM
+ */
+#include <linux/config.h>
+#include <linux/mm.h>
+
+#include <asm/processor.h>
+#include <asm/tlbflush.h>
+
+#ifdef CONFIG_SMP
+# define FREE_PTE_NR		2048
+# define tlb_fast_mode(tlb)	((tlb)->nr == ~0UL)
+#else
+# define FREE_PTE_NR		0
+# define tlb_fast_mode(tlb)	(1)
+#endif
+
+typedef struct {
+	struct mm_struct	*mm;
+	unsigned long		nr;	/* == ~0UL => fast mode */
+	unsigned long		freed;	/* number of pages freed */
+	unsigned long		start_addr;
+	unsigned long		end_addr;
+	struct page 		*pages[FREE_PTE_NR];
+} mmu_gather_t;
+
+/* Users of the generic TLB shootdown code must declare this storage space. */
+extern mmu_gather_t	mmu_gathers[NR_CPUS];
+
+/*
+ * Flush the TLB for address range START to END and, if not in fast mode, release the
+ * freed pages that where gathered up to this point.
+ */
+static inline void
+ia64_tlb_flush_mmu (mmu_gather_t *tlb, unsigned long start, unsigned long end)
+{
+	unsigned long nr;
+
+	if (unlikely (end - start >= 1024*1024*1024*1024UL
+		      || rgn_index(start) != rgn_index(end - 1)))
+	{
+		/*
+		 * If we flush more than a tera-byte or across regions, we're probably
+		 * better off just flushing the entire TLB(s).  This should be very rare
+		 * and is not worth optimizing for.
+		 */
+		flush_tlb_all();
+	} else {
+		/*
+		 * XXX fix me: flush_tlb_range() should take an mm pointer instead of a
+		 * vma pointer.
+		 */
+		struct vm_area_struct vma;
+
+		vma.vm_mm = tlb->mm;
+		/* flush the address range from the tlb: */
+		flush_tlb_range(&vma, start, end);
+		/* now flush the virt. page-table area mapping the address range: */
+		flush_tlb_range(&vma, ia64_thash(start), ia64_thash(end));
+	}
+
+	/* lastly, release the freed pages */
+	nr = tlb->nr;
+	if (!tlb_fast_mode(tlb)) {
+		unsigned long i;
+		tlb->nr = 0;
+		tlb->start_addr = ~0UL;
+		for (i = 0; i < nr; ++i)
+			free_page_and_swap_cache(tlb->pages[i]);
+	}
+}
+
+/*
+ * Return a pointer to an initialized mmu_gather_t.
+ */
+static inline mmu_gather_t *
+tlb_gather_mmu (struct mm_struct *mm)
+{
+	mmu_gather_t *tlb = &mmu_gathers[smp_processor_id()];
+
+	tlb->mm = mm;
+	tlb->freed = 0;
+	tlb->start_addr = ~0UL;
+
+	/* Use fast mode if only one CPU is online */
+	tlb->nr = smp_num_cpus > 1 ? 0UL : ~0UL;
+	return tlb;
+}
+
+/*
+ * Called at the end of the shootdown operation to free up any resources that were
+ * collected.  The page table lock is still held at this point.
+ */
+static inline void
+tlb_finish_mmu (mmu_gather_t *tlb, unsigned long start, unsigned long end)
+{
+	unsigned long freed = tlb->freed;
+	struct mm_struct *mm = tlb->mm;
+	unsigned long rss = mm->rss;
+
+	if (rss < freed)
+		freed = rss;
+	mm->rss = rss - freed;
+	/*
+	 * Note: tlb->nr may be 0 at this point, so we can't rely on tlb->start_addr and
+	 * tlb->end_addr.
+	 */
+	ia64_tlb_flush_mmu(tlb, start, end);
+
+	/* keep the page table cache within bounds */
+	check_pgt_cache();
+}
+
+/*
+ * Remove TLB entry for PTE mapped at virtual address ADDRESS.  This is called for any
+ * PTE, not just those pointing to (normal) physical memory.
+ */
+static inline void
+tlb_remove_tlb_entry (mmu_gather_t *tlb, pte_t pte, unsigned long address)
+{
+	if (tlb->start_addr == ~0UL)
+		tlb->start_addr = address;
+	tlb->end_addr = address + PAGE_SIZE;
+}
+
+/*
+ * Logically, this routine frees PAGE.  On MP machines, the actual freeing of the page
+ * must be delayed until after the TLB has been flushed (see comments at the beginning of
+ * this file).
+ */
+static inline void
+tlb_remove_page (mmu_gather_t *tlb, struct page *page)
+{
+	if (tlb_fast_mode(tlb)) {
+		free_page_and_swap_cache(page);
+		return;
+	}
+	tlb->pages[tlb->nr++] = page;
+	if (tlb->nr >= FREE_PTE_NR)
+		ia64_tlb_flush_mmu(tlb, tlb->start_addr, tlb->end_addr);
+}
+
 #define tlb_start_vma(tlb, vma)			do { } while (0)
 #define tlb_end_vma(tlb, vma)			do { } while (0)
-#define tlb_remove_tlb_entry(tlb, ptep, address) do { } while (0)
-#define tlb_flush(tlb)				flush_tlb_mm((tlb)->mm)
 
-#include <asm-generic/tlb.h>
+#endif /* _ASM_IA64_TLB_H */

include/asm-ia64/tlbflush.h

@@ -70,12 +70,10 @@ flush_tlb_page (struct vm_area_struct *vma, unsigned long addr)
 static inline void
 flush_tlb_pgtables (struct mm_struct *mm, unsigned long start, unsigned long end)
 {
-	struct vm_area_struct vma;
-
-	if (REGION_NUMBER(start) != REGION_NUMBER(end))
-		printk("flush_tlb_pgtables: can't flush across regions!!\n");
-	vma.vm_mm = mm;
-	flush_tlb_range(&vma, ia64_thash(start), ia64_thash(end));
+	/*
+	 * Deprecated.  The virtual page table is now flushed via the normal gather/flush
+	 * interface (see tlb.h).
+	 */
 }
 
 #define flush_tlb_kernel_range(start, end)	flush_tlb_all()	/* XXX fix me */

include/asm-ia64/unistd.h

@@ -223,6 +223,10 @@
 #define __NR_sched_setaffinity		1231
 #define __NR_sched_getaffinity		1232
 #define __NR_security			1233
+#define __NR_get_large_pages		1234
+#define __NR_free_large_pages		1235
+#define __NR_share_large_pages		1236
+#define __NR_unshare_large_pages	1237
 
 #if !defined(__ASSEMBLY__) && !defined(ASSEMBLER)