Merge wailua.hpl.hp.com:/var/autofs/net/tiger/data1/bk/vanilla/linux-2.5

into wailua.hpl.hp.com:/var/autofs/net/tiger/data1/bk/lia64/to-linus-2.5

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/boot/Makefile

@@ -15,6 +15,8 @@ OBJECTS	= bootloader.o
 targets-$(CONFIG_IA64_HP_SIM)	+= bootloader
 targets-$(CONFIG_IA64_GENERIC)	+= bootloader
 
+CFLAGS	:= $(CFLAGS) $(CFLAGS_KERNEL)
+
 all:	$(targets-y)
 
 bootloader: $(OBJECTS)

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 +100,20 @@ 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
@@ -123,39 +123,26 @@ if [ "$CONFIG_NET" = "y" ]; then
 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
 
-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'
+  mainmenu_option next_comment
+  comment 'ATA/ATAPI/MFM/RLL support'
 
-tristate 'ATA/ATAPI/MFM/RLL support' CONFIG_IDE
+  tristate 'ATA/ATAPI/MFM/RLL support' CONFIG_IDE
 
-if [ "$CONFIG_IDE" != "n" ]; then
+  if [ "$CONFIG_IDE" != "n" ]; then
     source drivers/ide/Config.in
-else
+  else
     define_bool CONFIG_BLK_DEV_HD n
+  fi
+  endmenu
 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 +155,7 @@ 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'
 
@@ -178,36 +164,46 @@ if [ "$CONFIG_NET" = "y" ]; then
       source drivers/net/Config.in
     fi
     endmenu
-fi
-
-source net/ax25/Config.in
-
-source drivers/isdn/Config.in
+  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)'
+  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 [ "$CONFIG_CD_NO_IDESCSI" != "n" ]; then
+  bool 'Support non-SCSI/IDE/ATAPI CDROM drives' CONFIG_CD_NO_IDESCSI
+  if [ "$CONFIG_CD_NO_IDESCSI" != "n" ]; then
     source drivers/cdrom/Config.in
-fi
-endmenu
+  fi
+  endmenu
 
-fi # !HP_SIM
+  #
+  # input before char - char/joystick depends on it. As does USB.
+  #
+  source drivers/input/Config.in
+  source drivers/char/Config.in
 
-#
-# input before char - char/joystick depends on it. As does USB.
-#
-source drivers/input/Config.in
-source drivers/char/Config.in
+  #source drivers/misc/Config.in
 
-#source drivers/misc/Config.in
+  source drivers/media/Config.in
+else # HP_SIM
 
-source drivers/media/Config.in
+  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
 
 source fs/Config.in
 
-if [ "$CONFIG_VT" = "y" ]; then
+if [ "$CONFIG_IA64_HP_SIM" = "n" ]; then
+  if [ "$CONFIG_VT" = "y" ]; then
     mainmenu_option next_comment
     comment 'Console drivers'
     bool 'VGA text console' CONFIG_VGA_CONSOLE
@@ -216,32 +212,26 @@ if [ "$CONFIG_VT" = "y" ]; then
       define_bool CONFIG_PCI_CONSOLE y
     fi
     endmenu
-fi
-
-if [ "$CONFIG_IA64_HP_SIM" = "n" ]; then
+  fi
 
-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
-  source arch/ia64/hp/Config.in
+  source arch/ia64/hp/sim/Config.in
 fi
 
-
 mainmenu_option next_comment
 comment 'Kernel hacking'
 

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/hpsim_setup.c

 /*
  * Platform dependent support for HP simulator.
  *
- * Copyright (C) 1998, 1999 Hewlett-Packard Co
- * Copyright (C) 1998, 1999 David Mosberger-Tang <davidm@hpl.hp.com>
+ * Copyright (C) 1998, 1999, 2002 Hewlett-Packard Co
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
  * Copyright (C) 1999 Vijay Chander <vijay@engr.sgi.com>
  */
+#include <linux/console.h>
 #include <linux/init.h>
+#include <linux/kdev_t.h>
 #include <linux/kernel.h>
+#include <linux/major.h>
 #include <linux/param.h>
+#include <linux/root_dev.h>
 #include <linux/string.h>
 #include <linux/types.h>
-#include <linux/kdev_t.h>
-#include <linux/console.h>
-#include <linux/root_dev.h>
 
 #include <asm/delay.h>
 #include <asm/irq.h>
@@ -55,5 +56,5 @@ hpsim_setup (char **cmdline_p)
 {
 	ROOT_DEV = Root_SDA1;		/* default to first SCSI drive */
 
-	register_console (&hpsim_cons);
+	register_console(&hpsim_cons);
 }

arch/ia64/hp/sim/simscsi.c

@@ -62,7 +62,9 @@ struct disk_stat {
 
 extern long ia64_ssc (long arg0, long arg1, long arg2, long arg3, int nr);
 
-static int desc[8] = { -1, -1, -1, -1, -1, -1, -1, -1 };
+static int desc[16] = {
+	-1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1
+};
 
 static struct queue_entry {
 	Scsi_Cmnd *sc;
@@ -148,9 +150,9 @@ simscsi_biosparam (Disk *disk, struct block_device *n, int ip[])
 {
 	int size = disk->capacity;
 
-	ip[0] = 64;
-	ip[1] = 32;
-	ip[2] = size >> 11;
+	ip[0] = 64;		/* heads */
+	ip[1] = 32;		/* sectors */
+	ip[2] = size >> 11;	/* cylinders */
 	return 0;
 }
 
@@ -229,6 +231,29 @@ simscsi_readwrite6 (Scsi_Cmnd *sc, int mode)
 		simscsi_readwrite(sc, mode, offset, sc->cmnd[4]*512);
 }
 
+static size_t
+simscsi_get_disk_size (int fd)
+{
+	struct disk_stat stat;
+	size_t bit, sectors = 0;
+	struct disk_req req;
+	char buf[512];
+
+	/*
+	 * This is a bit kludgey: the simulator doesn't provide a direct way of determining
+	 * the disk size, so we do a binary search, assuming a maximum disk size of 4GB.
+	 */
+	for (bit = (4UL << 30)/512; bit != 0; bit >>= 1) {
+		req.addr = __pa(&buf);
+		req.len = sizeof(buf);
+		ia64_ssc(fd, 1, __pa(&req), ((sectors | bit) - 1)*512, SSC_READ);
+		stat.fd = fd;
+		ia64_ssc(__pa(&stat), 0, 0, 0, SSC_WAIT_COMPLETION);
+		if (stat.count == sizeof(buf))
+			sectors |= bit;
+	}
+	return sectors - 1;	/* return last valid sector number */
+}
 
 static void
 simscsi_readwrite10 (Scsi_Cmnd *sc, int mode)
@@ -247,6 +272,7 @@ int
 simscsi_queuecommand (Scsi_Cmnd *sc, void (*done)(Scsi_Cmnd *))
 {
 	char fname[MAX_ROOT_LEN+16];
+	size_t disk_size;
 	char *buf;
 #if DEBUG_SIMSCSI
 	register long sp asm ("sp");
@@ -258,14 +284,14 @@ simscsi_queuecommand (Scsi_Cmnd *sc, void (*done)(Scsi_Cmnd *))
 
 	sc->result = DID_BAD_TARGET << 16;
 	sc->scsi_done = done;
-	if (sc->target <= 7 && sc->lun == 0) {
+	if (sc->target <= 15 && sc->lun == 0) {
 		switch (sc->cmnd[0]) {
 		      case INQUIRY:
 			if (sc->request_bufflen < 35) {
 				break;
 			}
 			sprintf (fname, "%s%c", simscsi_root, 'a' + sc->target);
-			desc[sc->target] = ia64_ssc (__pa(fname), SSC_READ_ACCESS|SSC_WRITE_ACCESS,
+			desc[sc->target] = ia64_ssc(__pa(fname), SSC_READ_ACCESS|SSC_WRITE_ACCESS,
 						    0, 0, SSC_OPEN);
 			if (desc[sc->target] < 0) {
 				/* disk doesn't exist... */
@@ -319,11 +345,13 @@ simscsi_queuecommand (Scsi_Cmnd *sc, void (*done)(Scsi_Cmnd *))
 			}
 			buf = sc->request_buffer;
 
+			disk_size = simscsi_get_disk_size(desc[sc->target]);
+
 			/* pretend to be a 1GB disk (partition table contains real stuff): */
-			buf[0] = 0x00;
-			buf[1] = 0x1f;
-			buf[2] = 0xff;
-			buf[3] = 0xff;
+			buf[0] = (disk_size >> 24) & 0xff;
+			buf[1] = (disk_size >> 16) & 0xff;
+			buf[2] = (disk_size >>  8) & 0xff;
+			buf[3] = (disk_size >>  0) & 0xff;
 			/* set block size of 512 bytes: */
 			buf[4] = 0;
 			buf[5] = 0;

arch/ia64/hp/sim/simserial.c

@@ -13,6 +13,7 @@
  *
  * 02/04/00 D. Mosberger	Merged in serial.c bug fixes in rs_close().
  * 02/25/00 D. Mosberger	Synced up with 2.3.99pre-5 version of serial.c.
+ * 07/30/02 D. Mosberger	Replace sti()/cli() with explicit spinlocks & local irq masking
  */
 
 #include <linux/config.h>
@@ -31,6 +32,7 @@
 #include <linux/serialP.h>
 
 #include <asm/irq.h>
+#include <asm/hw_irq.h>
 #include <asm/uaccess.h>
 
 #ifdef CONFIG_KDB
@@ -61,6 +63,7 @@ extern void ia64_ssc_connect_irq (long intr, long irq);
 
 static char *serial_name = "SimSerial driver";
 static char *serial_version = "0.6";
+static spinlock_t serial_lock = SPIN_LOCK_UNLOCKED;
 
 /*
  * This has been extracted from asm/serial.h. We need one eventually but
@@ -232,14 +235,14 @@ static void rs_put_char(struct tty_struct *tty, unsigned char ch)
 
 	if (!tty || !info->xmit.buf) return;
 
-	save_flags(flags); cli();
+	spin_lock_irqsave(&serial_lock, flags);
 	if (CIRC_SPACE(info->xmit.head, info->xmit.tail, SERIAL_XMIT_SIZE) == 0) {
-		restore_flags(flags);
+		spin_unlock_irqrestore(&serial_lock, flags);
 		return;
 	}
 	info->xmit.buf[info->xmit.head] = ch;
 	info->xmit.head = (info->xmit.head + 1) & (SERIAL_XMIT_SIZE-1);
-	restore_flags(flags);
+	spin_unlock_irqrestore(&serial_lock, flags);
 }
 
 static _INLINE_ void transmit_chars(struct async_struct *info, int *intr_done)
@@ -247,7 +250,7 @@ static _INLINE_ void transmit_chars(struct async_struct *info, int *intr_done)
 	int count;
 	unsigned long flags;
 
-	save_flags(flags); cli();
+	spin_lock_irqsave(&serial_lock, flags);
 
 	if (info->x_char) {
 		char c = info->x_char;
@@ -290,7 +293,7 @@ static _INLINE_ void transmit_chars(struct async_struct *info, int *intr_done)
 		info->xmit.tail += count;
 	}
 out:
-	restore_flags(flags);
+	spin_unlock_irqrestore(&serial_lock, flags);
 }
 
 static void rs_flush_chars(struct tty_struct *tty)
@@ -314,7 +317,6 @@ static int rs_write(struct tty_struct * tty, int from_user,
 
 	if (!tty || !info->xmit.buf || !tmp_buf) return 0;
 
-	save_flags(flags);
 	if (from_user) {
 		down(&tmp_buf_sem);
 		while (1) {
@@ -331,21 +333,26 @@ static int rs_write(struct tty_struct * tty, int from_user,
 					ret = -EFAULT;
 				break;
 			}
-			cli();
-			c1 = CIRC_SPACE_TO_END(info->xmit.head, info->xmit.tail, SERIAL_XMIT_SIZE);
+
+			spin_lock_irqsave(&serial_lock, flags);
+			{
+				c1 = CIRC_SPACE_TO_END(info->xmit.head, info->xmit.tail,
+						       SERIAL_XMIT_SIZE);
 				if (c1 < c)
 					c = c1;
 				memcpy(info->xmit.buf + info->xmit.head, tmp_buf, c);
 				info->xmit.head = ((info->xmit.head + c) &
 						   (SERIAL_XMIT_SIZE-1));
-			restore_flags(flags);
+			}
+			spin_unlock_irqrestore(&serial_lock, flags);
+
 			buf += c;
 			count -= c;
 			ret += c;
 		}
 		up(&tmp_buf_sem);
 	} else {
-		cli();
+		spin_lock_irqsave(&serial_lock, flags);
 		while (1) {
 			c = CIRC_SPACE_TO_END(info->xmit.head, info->xmit.tail, SERIAL_XMIT_SIZE);
 			if (count < c)
@@ -360,7 +367,7 @@ static int rs_write(struct tty_struct * tty, int from_user,
 			count -= c;
 			ret += c;
 		}
-		restore_flags(flags);
+		spin_unlock_irqrestore(&serial_lock, flags);
 	}
 	/*
 	 * Hey, we transmit directly from here in our case
@@ -391,9 +398,9 @@ static void rs_flush_buffer(struct tty_struct *tty)
 	struct async_struct *info = (struct async_struct *)tty->driver_data;
 	unsigned long flags;
 
-	save_flags(flags); cli();
+	spin_lock_irqsave(&serial_lock, flags);
 	info->xmit.head = info->xmit.tail = 0;
-	restore_flags(flags);
+	spin_unlock_irqrestore(&serial_lock, flags);
 
 	wake_up_interruptible(&tty->write_wait);
 
@@ -566,8 +573,8 @@ static void shutdown(struct async_struct * info)
 	       state->irq);
 #endif
 
-	save_flags(flags); cli(); /* Disable interrupts */
-
+	spin_lock_irqsave(&serial_lock, flags);
+	{
 		/*
 		 * First unlink the serial port from the IRQ chain...
 		 */
@@ -603,7 +610,8 @@ static void shutdown(struct async_struct * info)
 		if (info->tty) set_bit(TTY_IO_ERROR, &info->tty->flags);
 
 		info->flags &= ~ASYNC_INITIALIZED;
-	restore_flags(flags);
+	}
+	spin_unlock_irqrestore(&serial_lock, flags);
 }
 
 /*
@@ -626,14 +634,13 @@ static void rs_close(struct tty_struct *tty, struct file * filp)
 
 	state = info->state;
 
-	save_flags(flags); cli();
-
+	spin_lock_irqsave(&serial_lock, flags);
 	if (tty_hung_up_p(filp)) {
 #ifdef SIMSERIAL_DEBUG
 		printk("rs_close: hung_up\n");
 #endif
 		MOD_DEC_USE_COUNT;
-		restore_flags(flags);
+		spin_unlock_irqrestore(&serial_lock, flags);
 		return;
 	}
 #ifdef SIMSERIAL_DEBUG
@@ -658,11 +665,11 @@ static void rs_close(struct tty_struct *tty, struct file * filp)
 	}
 	if (state->count) {
 		MOD_DEC_USE_COUNT;
-		restore_flags(flags);
+		spin_unlock_irqrestore(&serial_lock, flags);
 		return;
 	}
 	info->flags |= ASYNC_CLOSING;
-	restore_flags(flags);
+	spin_unlock_irqrestore(&serial_lock, flags);
 
 	/*
 	 * Now we wait for the transmit buffer to clear; and we notify
@@ -770,7 +777,7 @@ startup(struct async_struct *info)
 	if (!page)
 		return -ENOMEM;
 
-	save_flags(flags); cli();
+	spin_lock_irqsave(&serial_lock, flags);
 
 	if (info->flags & ASYNC_INITIALIZED) {
 		free_page(page);
@@ -851,11 +858,11 @@ startup(struct async_struct *info)
 	}
 
 	info->flags |= ASYNC_INITIALIZED;
-	restore_flags(flags);
+	spin_unlock_irqrestore(&serial_lock, flags);
 	return 0;
 
 errout:
-	restore_flags(flags);
+	spin_unlock_irqrestore(&serial_lock, flags);
 	return retval;
 }
 

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/hp/zx1/hpzx1_misc.c

@@ -27,6 +27,7 @@
 #include "../drivers/acpi/include/acstruct.h"
 #include "../drivers/acpi/include/acnamesp.h"
 #include "../drivers/acpi/include/acutils.h"
+#include "../drivers/acpi/acpi_bus.h"
 
 #define PFX "hpzx1: "
 
@@ -109,7 +110,7 @@ static int hp_cfg_write##sz (struct pci_dev *dev, int where, u##bits value) \
 	\
 	switch (where) { \
 	case PCI_BASE_ADDRESS_0: \
-		if (value == ~0) \
+		if (value == (u##bits) ~0) \
 			fake_dev->sizing = 1; \
 		break; \
 	default: \
@@ -177,7 +178,7 @@ hpzx1_fake_pci_dev(unsigned long addr, unsigned int bus, unsigned int size)
 	 * Drivers should ioremap what they need, but we have to do
 	 * it here, too, so PCI config accesses work.
 	 */
-	dev->mapped_csrs = ioremap(dev->csr_base, dev->csr_size);
+	dev->mapped_csrs = (unsigned long) ioremap(dev->csr_base, dev->csr_size);
 
 	return dev;
 }
@@ -303,7 +304,7 @@ hpzx1_lba_probe(acpi_handle obj, u32 depth, void *context, void **ret)
 	if ((dev = hpzx1_fake_pci_dev(csr_base, busnum, csr_length)))
 		printk(KERN_INFO PFX "%s LBA at 0x%lx, _BBN 0x%02x; "
 			"pci dev %02x:%02x.%d\n",
-			name, csr_base, busnum, dev->bus,
+			name, csr_base, (unsigned int) busnum, dev->bus,
 			PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
 
 	return AE_OK;

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/ia32/ia32_ioctl.c

@@ -30,6 +30,8 @@
 #include <asm/ia32.h>
 
 #include <../drivers/char/drm/drm.h>
+#include <../drivers/char/drm/mga_drm.h>
+#include <../drivers/char/drm/i810_drm.h>
 
 
 #define IOCTL_NR(a)	((a) & ~(_IOC_SIZEMASK << _IOC_SIZESHIFT))

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/entry.S

@@ -175,6 +175,7 @@ GLOBAL_ENTRY(ia64_switch_to)
 (p6)	srlz.d
 	ld8 sp=[r21]			// load kernel stack pointer of new task
 	mov IA64_KR(CURRENT)=r20	// update "current" application register
+	mov r8=r13			// return pointer to previously running task
 	mov r13=in0			// set "current" pointer
 	;;
 	DO_LOAD_SWITCH_STACK

arch/ia64/kernel/ia64_ksyms.c

@@ -88,12 +88,6 @@ EXPORT_SYMBOL(ia64_cpu_to_sapicid);
 #include <asm/smplock.h>
 EXPORT_SYMBOL(kernel_flag);
 
-/* #include <asm/system.h> */
-EXPORT_SYMBOL(__global_sti);
-EXPORT_SYMBOL(__global_cli);
-EXPORT_SYMBOL(__global_save_flags);
-EXPORT_SYMBOL(__global_restore_flags);
-
 #else /* !CONFIG_SMP */
 
 EXPORT_SYMBOL(__flush_tlb_all);

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
@@ -422,6 +422,7 @@ register_irq (u32 global_vector, int vector, int pin, unsigned char delivery,
 	irq_desc_t *idesc;
 	struct hw_interrupt_type *irq_type;
 
+	gsi_to_vector(global_vector) = vector;
 	iosapic_irq[vector].pin	= pin;
 	iosapic_irq[vector].polarity = polarity ? IOSAPIC_POL_HIGH : IOSAPIC_POL_LOW;
 	iosapic_irq[vector].dmode    = delivery;
@@ -640,7 +641,7 @@ iosapic_init_pci_irq (void)
 	unsigned int irq;
 	char *addr;
 
-	if (0 != acpi_get_prt(&pci_irq.route, &pci_irq.num_routes))
+	if (acpi_get_prt(&pci_irq.route, &pci_irq.num_routes))
 		return;
 
 	for (i = 0; i < pci_irq.num_routes; i++) {
@@ -679,11 +680,10 @@ 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.c

@@ -200,277 +200,12 @@ int show_interrupts(struct seq_file *p, void *v)
 	return 0;
 }
 
-
-/*
- * Global interrupt locks for SMP. Allow interrupts to come in on any
- * CPU, yet make cli/sti act globally to protect critical regions..
- */
-
-#ifdef CONFIG_SMP
-unsigned int global_irq_holder = NO_PROC_ID;
-unsigned volatile long global_irq_lock; /* pedantic: long for set_bit --RR */
-
-extern void show_stack(unsigned long* esp);
-
-static void show(char * str)
-{
-	int i;
-	int cpu = smp_processor_id();
-
-	printk("\n%s, CPU %d:\n", str, cpu);
-	printk("irq:  %d [",irqs_running());
-	for(i=0;i < NR_CPUS;i++)
-		printk(" %d",irq_count(i));
-	printk(" ]\nbh:   %d [",spin_is_locked(&global_bh_lock) ? 1 : 0);
-	for(i=0;i < NR_CPUS;i++)
-		printk(" %d",bh_count(i));
-
-	printk(" ]\nStack dumps:");
-#if defined(CONFIG_IA64)
-	/*
-	 * We can't unwind the stack of another CPU without access to
-	 * the registers of that CPU.  And sending an IPI when we're
-	 * in a potentially wedged state doesn't sound like a smart
-	 * idea.
-	 */
-#elif defined(CONFIG_X86)
-	for(i=0;i< NR_CPUS;i++) {
-		unsigned long esp;
-		if(i==cpu)
-			continue;
-		printk("\nCPU %d:",i);
-		esp = init_tss[i].esp0;
-		if(esp==NULL) {
-			/* tss->esp0 is set to NULL in cpu_init(),
-			 * it's initialized when the cpu returns to user
-			 * space. -- manfreds
-			 */
-			printk(" <unknown> ");
-			continue;
-		}
-		esp &= ~(THREAD_SIZE-1);
-		esp += sizeof(struct task_struct);
-		show_stack((void*)esp);
-	}
-#else
-	You lose...
-#endif
-	printk("\nCPU %d:",cpu);
-	show_stack(NULL);
-	printk("\n");
-}
-
-#define MAXCOUNT 100000000
-
-/*
- * I had a lockup scenario where a tight loop doing
- * spin_unlock()/spin_lock() on CPU#1 was racing with
- * spin_lock() on CPU#0. CPU#0 should have noticed spin_unlock(), but
- * apparently the spin_unlock() information did not make it
- * through to CPU#0 ... nasty, is this by design, do we have to limit
- * 'memory update oscillation frequency' artificially like here?
- *
- * Such 'high frequency update' races can be avoided by careful design, but
- * some of our major constructs like spinlocks use similar techniques,
- * it would be nice to clarify this issue. Set this define to 0 if you
- * want to check whether your system freezes.  I suspect the delay done
- * by SYNC_OTHER_CORES() is in correlation with 'snooping latency', but
- * i thought that such things are guaranteed by design, since we use
- * the 'LOCK' prefix.
- */
-#define SUSPECTED_CPU_OR_CHIPSET_BUG_WORKAROUND 0
-
-#if SUSPECTED_CPU_OR_CHIPSET_BUG_WORKAROUND
-# define SYNC_OTHER_CORES(x) udelay(x+1)
-#else
-/*
- * We have to allow irqs to arrive between local_irq_enable and local_irq_disable
- */
-# ifdef CONFIG_IA64
-#  define SYNC_OTHER_CORES(x) __asm__ __volatile__ ("nop 0")
-# else
-#  define SYNC_OTHER_CORES(x) __asm__ __volatile__ ("nop")
-# endif
-#endif
-
-static inline void wait_on_irq(void)
-{
-	int count = MAXCOUNT;
-
-	for (;;) {
-
-		/*
-		 * Wait until all interrupts are gone. Wait
-		 * for bottom half handlers unless we're
-		 * already executing in one..
-		 */
-		if (!irqs_running())
-			if (really_local_bh_count() || !spin_is_locked(&global_bh_lock))
-				break;
-
-		/* Duh, we have to loop. Release the lock to avoid deadlocks */
-		smp_mb__before_clear_bit();	/* need barrier before releasing lock... */
-		clear_bit(0,&global_irq_lock);
-
-		for (;;) {
-			if (!--count) {
-				show("wait_on_irq");
-				count = ~0;
-			}
-			local_irq_enable();
-			SYNC_OTHER_CORES(smp_processor_id());
-			local_irq_disable();
-			if (irqs_running())
-				continue;
-			if (global_irq_lock)
-				continue;
-			if (!really_local_bh_count() && spin_is_locked(&global_bh_lock))
-				continue;
-			if (!test_and_set_bit(0,&global_irq_lock))
-				break;
-		}
-	}
-}
-
-/*
- * This is called when we want to synchronize with
- * interrupts. We may for example tell a device to
- * stop sending interrupts: but to make sure there
- * are no interrupts that are executing on another
- * CPU we need to call this function.
- */
-void synchronize_irq(void)
-{
-	if (irqs_running()) {
-		/* Stupid approach */
-		cli();
-		sti();
-	}
-}
-
-static inline void get_irqlock(void)
-{
-	if (test_and_set_bit(0,&global_irq_lock)) {
-		/* do we already hold the lock? */
-		if (smp_processor_id() == global_irq_holder)
-			return;
-		/* Uhhuh.. Somebody else got it. Wait.. */
-		do {
-			do {
-#ifdef CONFIG_X86
-				rep_nop();
-#endif
-			} while (test_bit(0,&global_irq_lock));
-		} while (test_and_set_bit(0,&global_irq_lock));
-	}
-	/*
-	 * We also to make sure that nobody else is running
-	 * in an interrupt context.
-	 */
-	wait_on_irq();
-
-	/*
-	 * Ok, finally..
-	 */
-	global_irq_holder = smp_processor_id();
-}
-
-#define EFLAGS_IF_SHIFT 9
-
-/*
- * A global "cli()" while in an interrupt context
- * turns into just a local cli(). Interrupts
- * should use spinlocks for the (very unlikely)
- * case that they ever want to protect against
- * each other.
- *
- * If we already have local interrupts disabled,
- * this will not turn a local disable into a
- * global one (problems with spinlocks: this makes
- * save_flags+cli+sti usable inside a spinlock).
- */
-void __global_cli(void)
-{
-	unsigned int flags;
-
-#ifdef CONFIG_IA64
-	local_save_flags(flags);
-	if (flags & IA64_PSR_I) {
-		local_irq_disable();
-		if (!really_local_irq_count())
-			get_irqlock();
-	}
-#else
-	local_save_flags(flags);
-	if (flags & (1 << EFLAGS_IF_SHIFT)) {
-		local_irq_disable();
-		if (!really_local_irq_count())
-			get_irqlock();
-	}
-#endif
-}
-
-void __global_sti(void)
-{
-	if (!really_local_irq_count())
-		release_irqlock(smp_processor_id());
-	local_irq_enable();
-}
-
-/*
- * SMP flags value to restore to:
- * 0 - global cli
- * 1 - global sti
- * 2 - local cli
- * 3 - local sti
- */
-unsigned long __global_save_flags(void)
-{
-	int retval;
-	int local_enabled;
-	unsigned long flags;
-	int cpu = smp_processor_id();
-
-	local_save_flags(flags);
-#ifdef CONFIG_IA64
-	local_enabled = (flags & IA64_PSR_I) != 0;
-#else
-	local_enabled = (flags >> EFLAGS_IF_SHIFT) & 1;
-#endif
-	/* default to local */
-	retval = 2 + local_enabled;
-
-	/* check for global flags if we're not in an interrupt */
-	if (!really_local_irq_count()) {
-		if (local_enabled)
-			retval = 1;
-		if (global_irq_holder == cpu)
-			retval = 0;
-	}
-	return retval;
-}
-
-void __global_restore_flags(unsigned long flags)
+#if CONFIG_SMP
+inline void synchronize_irq(unsigned int irq)
 {
-	switch (flags) {
-	case 0:
-		__global_cli();
-		break;
-	case 1:
-		__global_sti();
-		break;
-	case 2:
-		local_irq_disable();
-		break;
-	case 3:
-		local_irq_enable();
-		break;
-	default:
-		printk("global_restore_flags: %08lx (%08lx)\n",
-			flags, (&flags)[-1]);
-	}
+	while (irq_desc(irq)->status & IRQ_INPROGRESS)
+		cpu_relax();
 }
-
 #endif
 
 /*
@@ -482,11 +217,7 @@ void __global_restore_flags(unsigned long flags)
  */
 int handle_IRQ_event(unsigned int irq, struct pt_regs * regs, struct irqaction * action)
 {
-	int status;
-
-	local_irq_enter(irq);
-
-	status = 1;	/* Force the "do bottom halves" bit */
+	int status = 1;	/* Force the "do bottom halves" bit */
 
 	if (!(action->flags & SA_INTERRUPT))
 		local_irq_enable();
@@ -500,11 +231,16 @@ int handle_IRQ_event(unsigned int irq, struct pt_regs * regs, struct irqaction *
 		add_interrupt_randomness(irq);
 	local_irq_disable();
 
-	local_irq_exit(irq);
-
 	return status;
 }
 
+/*
+ * Generic enable/disable code: this just calls
+ * down into the PIC-specific version for the actual
+ * hardware disable after having gotten the irq
+ * controller lock.
+ */
+
 /**
  *	disable_irq_nosync - disable an irq without waiting
  *	@irq: Interrupt to disable
@@ -546,14 +282,7 @@ inline void disable_irq_nosync(unsigned int irq)
 void disable_irq(unsigned int irq)
 {
 	disable_irq_nosync(irq);
-
-#ifdef CONFIG_SMP
-	if (!really_local_irq_count()) {
-		do {
-			barrier();
-		} while (irq_desc(irq)->status & IRQ_INPROGRESS);
-	}
-#endif
+	synchronize_irq(irq);
 }
 
 /**
@@ -616,6 +345,7 @@ unsigned int do_IRQ(unsigned long irq, struct pt_regs *regs)
 	struct irqaction * action;
 	unsigned int status;
 
+	irq_enter();
 	kstat.irqs[cpu][irq]++;
 
 	if (desc->status & IRQ_PER_CPU) {
@@ -682,6 +412,7 @@ unsigned int do_IRQ(unsigned long irq, struct pt_regs *regs)
 		desc->handler->end(irq);
 		spin_unlock(&desc->lock);
 	}
+	irq_exit();
 	return 1;
 }
 
@@ -811,7 +542,7 @@ void free_irq(unsigned int irq, void *dev_id)
 #ifdef CONFIG_SMP
 			/* Wait to make sure it's not being used on another CPU */
 			while (desc->status & IRQ_INPROGRESS)
-				barrier();
+				synchronize_irq(irq);
 #endif
 			kfree(action);
 			return;
@@ -864,7 +595,7 @@ unsigned long probe_irq_on(void)
 
 	/* Wait for longstanding interrupts to trigger. */
 	for (delay = jiffies + HZ/50; time_after(delay, jiffies); )
-		/* about 20ms delay */ synchronize_irq();
+		/* about 20ms delay */ barrier();
 
 	/*
 	 * enable any unassigned irqs
@@ -887,7 +618,7 @@ unsigned long probe_irq_on(void)
 	 * Wait for spurious interrupts to trigger
 	 */
 	for (delay = jiffies + HZ/10; time_after(delay, jiffies); )
-		/* about 100ms delay */ synchronize_irq();
+		/* about 100ms delay */ barrier();
 
 	/*
 	 * Now filter out any obviously spurious interrupts

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 */
@@ -50,6 +54,11 @@ __u8 isa_irq_to_vector_map[16] = {
 	0x28, 0x27, 0x26, 0x25, 0x24, 0x23, 0x22, 0x21
 };
 
+/*
+ * GSI to IA-64 vector translation table.
+ */
+__u8 gsi_to_vector_map[255];
+
 int
 ia64_alloc_irq (void)
 {
@@ -144,9 +153,9 @@ 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 +181,9 @@ 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"
 };
 
 /*
@@ -626,9 +626,12 @@ ia64_mca_wakeup_all(void)
 void
 ia64_mca_rendez_int_handler(int rendez_irq, void *arg, struct pt_regs *ptregs)
 {
-	int flags, cpu = 0;
+	unsigned long flags;
+	int cpu = 0;
+
 	/* Mask all interrupts */
-	save_and_cli(flags);
+#warning XXX fix me: this used to be: save_and_cli(flags);
+	local_irq_save(flags);
 
 #ifdef CONFIG_SMP
 	cpu = cpu_logical_id(hard_smp_processor_id());
@@ -646,7 +649,7 @@ ia64_mca_rendez_int_handler(int rendez_irq, void *arg, struct pt_regs *ptregs)
 	ia64_mca_wakeup_ipi_wait();
 
 	/* Enable all interrupts */
-	restore_flags(flags);
+	local_irq_restore(flags);
 }
 
 

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

@@ -165,7 +165,7 @@ struct pci_ops pci_sal_ops = {
  */
 
 struct pci_bus *
-pcibios_scan_root(int seg, int bus)
+pcibios_scan_root(int bus)
 {
 	struct list_head *list = NULL;
 	struct pci_bus *pci_bus = NULL;
@@ -174,12 +174,12 @@ pcibios_scan_root(int seg, int bus)
 		pci_bus = pci_bus_b(list);
 		if (pci_bus->number == bus) {
 			/* Already scanned */
-			printk("PCI: Bus (%02x:%02x) already probed\n", seg, bus);
+			printk("PCI: Bus (%02x) already probed\n", bus);
 			return pci_bus;
 		}
 	}
 
-	printk("PCI: Probing PCI hardware on bus (%02x:%02x)\n", seg, bus);
+	printk("PCI: Probing PCI hardware on bus (%02x)\n", bus);
 
 	return pci_scan_bus(bus, pci_root_ops, NULL);
 }
@@ -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
 };
 
 /*
@@ -610,7 +616,7 @@ pfm_remove_smpl_mapping(struct task_struct *task)
 
 	down_write(&task->mm->mmap_sem);
 
-	r = do_munmap(task->mm, ctx->ctx_smpl_vaddr, psb->psb_size);
+	r = do_munmap(task->mm, ctx->ctx_smpl_vaddr, psb->psb_size, 0);
 
 	up_write(&task->mm->mmap_sem);
 	if (r !=0) {
@@ -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

@@ -436,7 +436,7 @@ static void *
 c_start (struct seq_file *m, loff_t *pos)
 {
 #ifdef CONFIG_SMP
-	while (*pos < NR_CPUS && !(cpu_online_map & (1 << *pos)))
+	while (*pos < NR_CPUS && !(cpu_online_map & (1UL << *pos)))
 		++*pos;
 #endif
 	return *pos < NR_CPUS ? cpu_data(*pos) : NULL;
@@ -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

 /*
  * SMP boot-related support
  *
- * Copyright (C) 2001 David Mosberger-Tang <davidm@hpl.hp.com>
+ * Copyright (C) 1998-2002 Hewlett-Packard Co
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
  *
  * 01/05/16 Rohit Seth <rohit.seth@intel.com>	Moved SMP booting functions from smp.c to here.
  * 01/04/27 David Mosberger <davidm@hpl.hp.com>	Added ITC synching code.
+ * 02/07/31 David Mosberger <davidm@hpl.hp.com>	Switch over to hotplug-CPU boot-sequence.
+ *						smp_boot_cpus()/smp_commence() is replaced by
+ *						smp_prepare_cpus()/__cpu_up()/smp_cpus_done().
  */
 
 
@@ -66,18 +70,16 @@ static volatile unsigned long go[SLAVE + 1];
 
 #define DEBUG_ITC_SYNC	0
 
-extern void __init calibrate_delay(void);
-extern void start_ap(void);
+extern void __init calibrate_delay (void);
+extern void start_ap (void);
 extern unsigned long ia64_iobase;
 
 int cpucount;
 task_t *task_for_booting_cpu;
 
-/* Setup configured maximum number of CPUs to activate */
-static int max_cpus = -1;
-
 /* Bitmask of currently online CPUs */
 volatile unsigned long cpu_online_map;
+unsigned long phys_cpu_present_map;
 
 /* which logical CPU number maps to which CPU (physical APIC ID) */
 volatile int ia64_cpu_to_sapicid[NR_CPUS];
@@ -86,44 +88,12 @@ static volatile unsigned long cpu_callin_map;
 
 struct smp_boot_data smp_boot_data __initdata;
 
-/* Set when the idlers are all forked */
-volatile int smp_threads_ready;
-
 unsigned long ap_wakeup_vector = -1; /* External Int use to wakeup APs */
 
 char __initdata no_int_routing;
 
 unsigned char smp_int_redirect; /* are INT and IPI redirectable by the chipset? */
 
-/*
- * Setup routine for controlling SMP activation
- *
- * Command-line option of "nosmp" or "maxcpus=0" will disable SMP
- * activation entirely (the MPS table probe still happens, though).
- *
- * Command-line option of "maxcpus=<NUM>", where <NUM> is an integer
- * greater than 0, limits the maximum number of CPUs activated in
- * SMP mode to <NUM>.
- */
-
-static int __init
-nosmp (char *str)
-{
-	max_cpus = 0;
-	return 1;
-}
-
-__setup("nosmp", nosmp);
-
-static int __init
-maxcpus (char *str)
-{
-	get_option(&str, &max_cpus);
-	return 1;
-}
-
-__setup("maxcpus=", maxcpus);
-
 static int __init
 nointroute (char *str)
 {
@@ -299,7 +269,7 @@ smp_setup_percpu_timer (void)
 
 static volatile atomic_t smp_commenced = ATOMIC_INIT(0);
 
-void __init
+static void __init
 smp_commence (void)
 {
 	/*
@@ -308,7 +278,7 @@ smp_commence (void)
 	Dprintk("Setting commenced=1, go go go\n");
 
 	wmb();
-	atomic_set(&smp_commenced,1);
+	atomic_set(&smp_commenced, 1);
 }
 
 
@@ -405,6 +375,9 @@ do_boot_cpu (int sapicid)
 	int timeout, cpu;
 
 	cpu = ++cpucount;
+
+	set_bit(cpu, &phys_cpu_present_map);
+
 	/*
 	 * We can't use kernel_thread since we must avoid to
 	 * reschedule the child.
@@ -425,7 +398,7 @@ 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);
 
@@ -466,8 +439,8 @@ smp_tune_scheduling (void)
 /*
  * Cycle through the APs sending Wakeup IPIs to boot each.
  */
-void __init
-smp_boot_cpus (void)
+static void __init
+smp_boot_cpus (unsigned int max_cpus)
 {
 	int sapicid, cpu;
 	int boot_cpu_id = hard_smp_processor_id();
@@ -486,13 +459,13 @@ smp_boot_cpus (void)
 	 */
 	set_bit(0, &cpu_online_map);
 	set_bit(0, &cpu_callin_map);
+	set_bit(0, &phys_cpu_present_map);
 
 	local_cpu_data->loops_per_jiffy = loops_per_jiffy;
 	ia64_cpu_to_sapicid[0] = boot_cpu_id;
 
 	printk("Boot processor id 0x%x/0x%x\n", 0, boot_cpu_id);
 
-	global_irq_holder = NO_PROC_ID;
 	current_thread_info()->cpu = 0;
 	smp_tune_scheduling();
 
@@ -537,11 +510,11 @@ 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++)
-				if (cpu_online_map & (1<<cpu))
+				if (cpu_online_map & (1UL << cpu))
 					bogosum += cpu_data(cpu)->loops_per_jiffy;
 
 			printk(KERN_INFO"Total of %d processors activated (%lu.%02lu BogoMIPS).\n",
@@ -552,6 +525,29 @@ smp_boot_cpus (void)
 	;
 }
 
+void __init
+smp_prepare_cpus (unsigned int max_cpus)
+{
+ 	smp_boot_cpus(max_cpus);
+}
+
+int __devinit
+__cpu_up (unsigned int cpu)
+{
+	/*
+	 * Yeah, that's cheesy, but it will do until there is real hotplug support and in
+	 * the meantime, this gives time for the interface changes to settle down...
+	 */
+	smp_commence();
+	return 0;
+}
+
+void __init
+smp_cpus_done (unsigned int max_cpus)
+{
+	/* nuthing... */
+}
+
 /*
  * Assume that CPU's have been discovered by some platform-dependant interface.  For
  * SoftSDV/Lion, that would be ACPI.
@@ -571,9 +567,6 @@ init_smp_config(void)
 	ap_startup = (struct fptr *) start_ap;
 	sal_ret = ia64_sal_set_vectors(SAL_VECTOR_OS_BOOT_RENDEZ,
 				       __pa(ap_startup->fp), __pa(ap_startup->gp), 0, 0, 0, 0);
-	if (sal_ret < 0) {
-		printk("SMP: Can't set SAL AP Boot Rendezvous: %s\n     Forcing UP mode\n",
-		       ia64_sal_strerror(sal_ret));
-		max_cpus = 0;
-	}
+	if (sal_ret < 0)
+		printk("SMP: Can't set SAL AP Boot Rendezvous: %s\n", ia64_sal_strerror(sal_ret));
 }

arch/ia64/kernel/sys_ia64.c

@@ -82,7 +82,6 @@ asmlinkage unsigned long
 ia64_shmat (int shmid, void *shmaddr, int shmflg, long arg3, long arg4, long arg5, long arg6,
 	    long arg7, long stack)
 {
-	extern int sys_shmat (int shmid, char *shmaddr, int shmflg, ulong *raddr);
 	struct pt_regs *regs = (struct pt_regs *) &stack;
 	unsigned long raddr;
 	int retval;
@@ -120,7 +119,7 @@ ia64_brk (unsigned long brk, long arg1, long arg2, long arg3,
 
 	/* Always allow shrinking brk. */
 	if (brk <= mm->brk) {
-		if (!do_munmap(mm, newbrk, oldbrk-newbrk))
+		if (!do_munmap(mm, newbrk, oldbrk-newbrk, 1))
 			goto set_brk;
 		goto out;
 	}
@@ -138,10 +137,6 @@ ia64_brk (unsigned long brk, long arg1, long arg2, long arg3,
 	if (find_vma_intersection(mm, oldbrk, newbrk+PAGE_SIZE))
 		goto out;
 
-	/* Check if we have enough memory.. */
-	if (!vm_enough_memory((newbrk-oldbrk) >> PAGE_SHIFT))
-		goto out;
-
 	/* Ok, looks good - let it rip. */
 	if (do_brk(oldbrk, newbrk-oldbrk) != oldbrk)
 		goto out;

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

@@ -62,14 +62,14 @@ trap_init (void)
 void
 bust_spinlocks (int yes)
 {
+	int loglevel_save = console_loglevel;
+
 	spin_lock_init(&timerlist_lock);
 	if (yes) {
 		oops_in_progress = 1;
-#ifdef CONFIG_SMP
-		global_irq_lock = 0;	/* Many serial drivers do __global_cli() */
-#endif
-	} else {
-		int loglevel_save = console_loglevel;
+		return;
+	}
+
 #ifdef CONFIG_VT
 	unblank_screen();
 #endif
@@ -82,7 +82,6 @@ bust_spinlocks (int yes)
 	console_loglevel = 15;		/* NMI oopser may have shut the console up */
 	printk(" ");
 	console_loglevel = loglevel_save;
-	}
 }
 
 void
@@ -93,9 +92,9 @@ 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 +434,7 @@ 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 +478,30 @@ 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 +514,7 @@ 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 +531,15 @@ 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

@@ -9,40 +9,48 @@ export-objs := 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
 
-$(L_TARGET): $(obj-y) $(export-objs)
+include $(TOPDIR)/Rules.make
+
+AFLAGS___divdi3.o	=
+AFLAGS___udivdi3.o	= -DUNSIGNED
+AFLAGS___moddi3.o	= 	     -DMODULO
+AFLAGS___umoddi3.o	= -DUNSIGNED -DMODULO
+
+AFLAGS___divsi3.o	=
+AFLAGS___udivsi3.o	= -DUNSIGNED
+AFLAGS___modsi3.o	=	     -DMODULO
+AFLAGS___umodsi3.o	= -DUNSIGNED -DMODULO
 
 __divdi3.o: idiv64.S
-	$(CC) $(AFLAGS) -c -o $@ $<
+	$(cmd_as_o_S)
 
 __udivdi3.o: idiv64.S
-	$(CC) $(AFLAGS) -c -DUNSIGNED -c -o $@ $<
+	$(cmd_as_o_S)
 
 __moddi3.o: idiv64.S
-	$(CC) $(AFLAGS) -c -DMODULO -c -o $@ $<
+	$(cmd_as_o_S)
 
 __umoddi3.o: idiv64.S
-	$(CC) $(AFLAGS) -c -DMODULO -DUNSIGNED -c -o $@ $<
+	$(cmd_as_o_S)
 
 __divsi3.o: idiv32.S
-	$(CC) $(AFLAGS) -c -o $@ $<
+	$(cmd_as_o_S)
 
 __udivsi3.o: idiv32.S
-	$(CC) $(AFLAGS) -c -DUNSIGNED -c -o $@ $<
+	$(cmd_as_o_S)
 
 __modsi3.o: idiv32.S
-	$(CC) $(AFLAGS) -c -DMODULO -c -o $@ $<
+	$(cmd_as_o_S)
 
 __umodsi3.o: idiv32.S
-	$(CC) $(AFLAGS) -c -DMODULO -DUNSIGNED -c -o $@ $<
-
-include $(TOPDIR)/Rules.make
+	$(cmd_as_o_S)

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 +415,21 @@ 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 = (dma_addr_t)
+				map_single(hwdev, addr, sg->length, direction);
+		else
+			sg->dma_address = pci_addr;
+		sg->dma_length = sg->length;
 	}
 	return nelems;
 }
@@ -444,12 +447,10 @@ 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, (void *) 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 +469,14 @@ 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, (void *) 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

@@ -35,12 +35,14 @@
 		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
 };
 
+u8 ia64_need_tlb_flush __per_cpu_data;
+
 /*
  * Acquire the ia64_ctx.lock before calling this function!
  */
@@ -49,6 +51,7 @@ 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 +80,11 @@ 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 < NR_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

arch/ia64/tools/Makefile

@@ -4,7 +4,9 @@ TARGET	= $(TOPDIR)/include/asm-ia64/offsets.h
 
 all:
 
-mrproper:
+fastdep:
+
+mrproper: clean
 
 clean:
 	rm -f print_offsets.s print_offsets offsets.h

arch/ia64/vmlinux.lds.S

@@ -41,9 +41,6 @@ SECTIONS
 
   /* Read-only data */
 
-  . = ALIGN(16);
-  __gp = . + 0x200000;	/* gp must be 16-byte aligned for exc. table */
-
   /* Global data */
   _data = .;
 
@@ -146,6 +143,9 @@ SECTIONS
   .data : AT(ADDR(.data) - PAGE_OFFSET)
 	{ *(.data) *(.gnu.linkonce.d*) CONSTRUCTORS }
 
+  . = ALIGN(16);
+  __gp = . + 0x200000;	/* gp must be 16-byte aligned for exc. table */
+
   .got : AT(ADDR(.got) - PAGE_OFFSET)
 	{ *(.got.plt) *(.got) }
   /* We want the small data sections together, so single-instruction offsets

include/asm-ia64/acpi.h

@@ -30,11 +30,74 @@
 
 #ifdef __KERNEL__
 
-#define __acpi_map_table(phys_addr, size) __va(phys_addr)
+#define COMPILER_DEPENDENT_INT64	long
+#define COMPILER_DEPENDENT_UINT64	unsigned long
+
+/*
+ * Calling conventions:
+ *
+ * ACPI_SYSTEM_XFACE        - Interfaces to host OS (handlers, threads)
+ * ACPI_EXTERNAL_XFACE      - External ACPI interfaces
+ * ACPI_INTERNAL_XFACE      - Internal ACPI interfaces
+ * ACPI_INTERNAL_VAR_XFACE  - Internal variable-parameter list interfaces
+ */
+#define ACPI_SYSTEM_XFACE
+#define ACPI_EXTERNAL_XFACE
+#define ACPI_INTERNAL_XFACE
+#define ACPI_INTERNAL_VAR_XFACE
+
+/* Asm macros */
+
+#define ACPI_ASM_MACROS
+#define BREAKPOINT3
+#define ACPI_DISABLE_IRQS() local_irq_disable()
+#define ACPI_ENABLE_IRQS()  local_irq_enable()
+#define ACPI_FLUSH_CPU_CACHE()
+
+#define ACPI_ACQUIRE_GLOBAL_LOCK(GLptr, Acq) \
+	do { \
+	__asm__ volatile ("1:  ld4      r29=%1\n"  \
+		";;\n"                  \
+		"mov    ar.ccv=r29\n"   \
+		"mov    r2=r29\n"       \
+		"shr.u  r30=r29,1\n"    \
+		"and    r29=-4,r29\n"   \
+		";;\n"                  \
+		"add    r29=2,r29\n"    \
+		"and    r30=1,r30\n"    \
+		";;\n"                  \
+		"add    r29=r29,r30\n"  \
+		";;\n"                  \
+		"cmpxchg4.acq   r30=%1,r29,ar.ccv\n" \
+		";;\n"                  \
+		"cmp.eq p6,p7=r2,r30\n" \
+		"(p7) br.dpnt.few 1b\n" \
+		"cmp.gt p8,p9=3,r29\n"  \
+		";;\n"                  \
+		"(p8) mov %0=-1\n"      \
+		"(p9) mov %0=r0\n"      \
+		:"=r"(Acq):"m"(GLptr):"r2","r29","r30","memory"); \
+	} while (0)
+
+#define ACPI_RELEASE_GLOBAL_LOCK(GLptr, Acq) \
+	do { \
+	__asm__ volatile ("1:  ld4      r29=%1\n" \
+		";;\n"                  \
+		"mov    ar.ccv=r29\n"   \
+		"mov    r2=r29\n"       \
+		"and    r29=-4,r29\n"   \
+		";;\n"                  \
+		"cmpxchg4.acq   r30=%1,r29,ar.ccv\n" \
+		";;\n"                  \
+		"cmp.eq p6,p7=r2,r30\n" \
+		"(p7) br.dpnt.few 1b\n" \
+		"and    %0=1,r2\n"      \
+		";;\n"                  \
+		:"=r"(Acq):"m"(GLptr):"r2","r29","r30","memory"); \
+	} while (0)
 
 const char *acpi_get_sysname (void);
 int acpi_boot_init (char *cdline);
-int acpi_find_rsdp (unsigned long *phys_addr);
 int acpi_request_vector (u32 int_type);
 int acpi_get_prt (struct pci_vector_struct **vectors, int *count);
 int acpi_get_interrupt_model(int *type);

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/cacheflush.h

@@ -6,6 +6,8 @@
  *	David Mosberger-Tang <davidm@hpl.hp.com>
  */
 
+#include <linux/page-flags.h>
+
 #include <asm/bitops.h>
 #include <asm/page.h>
 
@@ -23,7 +25,7 @@
 
 #define flush_dcache_page(page)			\
 do {						\
-	clear_bit(PG_arch_1, &page->flags);	\
+	clear_bit(PG_arch_1, &(page)->flags);	\
 } while (0)
 
 extern void flush_icache_range (unsigned long start, unsigned long end);

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/hw_irq.h

@@ -2,10 +2,11 @@
 #define _ASM_IA64_HW_IRQ_H
 
 /*
- * Copyright (C) 2001 Hewlett-Packard Co
- * Copyright (C) 2001 David Mosberger-Tang <davidm@hpl.hp.com>
+ * Copyright (C) 2001-2002 Hewlett-Packard Co
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
  */
 
+#include <linux/interrupt.h>
 #include <linux/sched.h>
 #include <linux/types.h>
 
@@ -67,6 +68,8 @@ enum {
 
 extern __u8 isa_irq_to_vector_map[16];
 #define isa_irq_to_vector(x)	isa_irq_to_vector_map[(x)]
+extern __u8 gsi_to_vector_map[255];
+#define gsi_to_vector(x)	gsi_to_vector_map[(x)]
 
 extern unsigned long ipi_base_addr;
 

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