Merge bk://linux.bkbits.net/linux-2.5

into maxwell.earthlink.net:/usr/src/linus-2.5

Documentation/BK-usage/00-INDEX

+bk-kernel-howto.txt: Description of kernel workflow under BitKeeper
+
+bk-make-sum: Create summary of changesets in one repository and not
+another, typically in preparation to be sent to an upstream maintainer.
+Typical usage:
+	cd my-updated-repo
+	bk-make-sum ~/repo/original-repo
+	mv /tmp/linus.txt ../original-repo.txt
+
+bksend: Create readable text output containing summary of changes, GNU
+patch of the changes, and BK metadata of changes (as needed for proper
+importing into BitKeeper by an upstream maintainer).  This output is
+suitable for emailing BitKeeper changes.  The recipient of this output
+may pipe it directly to 'bk receive'.
+
+bz64wrap: helper script. Uncompressed input is piped to this script,
+which compresses its input, and then outputs the uu-/base64-encoded
+version of the compressed input.
+
+csets-to-patches: Produces a delta of two BK repositories, in the form
+of individual files, each containing a single cset as a GNU patch.
+Output is several files, each with the filename "/tmp/rev-$REV.patch"
+Typical usage:
+	cd my-updated-repo
+	bk changes -L ~/repo/original-repo 2>&1 | \
+		perl csets-to-patches
+
+cset-to-linus: Produces a delta of two BK repositories, in the form of
+changeset descriptions, with 'diffstat' output created for each
+individual changset.
+Typical usage:
+	cd my-updated-repo
+	bk changes -L ~/repo/original-repo 2>&1 | \
+		perl cset-to-linus > summary.txt
+
+unbz64wrap: Reverse an encoded, compressed data stream created by
+bz64wrap into an uncompressed, typically text/plain output.
+

Documentation/BK-usage/bk-kernel-howto.txt

@@ -32,7 +32,7 @@ land at the right destination... but I'm getting ahead of myself.
 
 Let's start with this progression:
 Each BitKeeper source tree on disk is a repository unto itself.
-Each repository has a parent.
+Each repository has a parent (except the root/original, of course).
 Each repository contains a set of a changesets ("csets").
 Each cset is one or more changed files, bundled together.
 

arch/ia64/Config.help

@@ -157,6 +157,11 @@ CONFIG_PCMCIA
   and ds.o.  If you want to compile it as a module, say M here and
   read <file:Documentation/modules.txt>.
 
+CONFIG_KALLSYMS
+  Say Y here to let the kernel print out symbolic crash information and
+  symbolic stack backtraces. This increases the size of the kernel
+  somewhat, as all symbols have to be loaded into the kernel image.
+
 CONFIG_KCORE_ELF
   If you enabled support for /proc file system then the file
   /proc/kcore will contain the kernel core image. This can be used

arch/ia64/Makefile

@@ -33,44 +33,48 @@ ifeq ($(CONFIG_ITANIUM_BSTEP_SPECIFIC),y)
 	CFLAGS += -mb-step
 endif
 
-HEAD := arch/$(ARCH)/kernel/head.o arch/ia64/kernel/init_task.o
+HEAD := arch/$(ARCH)/kernel/head.o arch/$(ARCH)/kernel/init_task.o
 
-core-$(CONFIG_IA64_GENERIC) 	+= arch/$(ARCH)/hp/	  arch/$(ARCH)/dig/
-core-$(CONFIG_IA64_HP_SIM)	+= arch/$(ARCH)/hp/
-core-$(CONFIG_IA64_HP_ZX1)	+= arch/$(ARCH)/hp/ 	  arch/$(ARCH)/dig/
-core-$(CONFIG_IA64_SGI_SN)	+= arch/$(ARCH)/sn/kernel arch/$(ARCH)/sn/io
+libs-y				+= arch/$(ARCH)/lib/
+core-y				+= arch/$(ARCH)/kernel/ arch/$(ARCH)/mm/
+core-$(CONFIG_IA32_SUPPORT)	+= arch/$(ARCH)/ia32/
 core-$(CONFIG_IA64_DIG) 	+= arch/$(ARCH)/dig/
+core-$(CONFIG_IA64_GENERIC) 	+= arch/$(ARCH)/dig/ arch/$(ARCH)/hp/common/ arch/$(ARCH)/hp/zx1/ \
+				   arch/$(ARCH)/hp/sim/
+core-$(CONFIG_IA64_HP_ZX1)	+= arch/$(ARCH)/dig/
+core-$(CONFIG_IA64_SGI_SN)	+= arch/$(ARCH)/sn/kernel arch/$(ARCH)/sn/io \
+				   arch/$(ARCH)/sn/fakeprom
+drivers-$(CONFIG_PCI)		+= arch/$(ARCH)/pci/
+drivers-$(CONFIG_IA64_HP_SIM)	+= arch/$(ARCH)/hp/sim/
+drivers-$(CONFIG_IA64_HP_ZX1)	+= arch/$(ARCH)/hp/common/ arch/$(ARCH)/hp/zx1/
 
 ifdef CONFIG_IA64_SGI_SN
 	CFLAGS			+= -DBRINGUP
         SUBDIRS         	+= arch/$(ARCH)/sn/fakeprom
 endif
 
-core-$(CONFIG_IA32_SUPPORT)	+= arch/$(ARCH)/ia32/
+makeboot = $(call descend,arch/ia64/boot,$(1))
+maketool = $(call descend,arch/ia64/tools,$(1))
 
-libs-y				+= arch/$(ARCH)/lib/
-core-y				+= arch/$(ARCH)/kernel/ arch/$(ARCH)/mm/
-drivers-$(CONFIG_PCI)		+= arch/$(ARCH)/pci/
-drivers-$(CONFIG_IA64_HP_ZX1)	+= arch/$(ARCH)/hp/common/ arch/$(ARCH)/hp/zx1/
+.PHONY: compressed archclean archmrproper $(TOPDIR)/include/asm-ia64/offsets.h
+
+all: compressed boot
 
-MAKEBOOT = $(MAKE) -C arch/$(ARCH)/boot
+boot: vmlinux
+	+@$(call makeboot,all)
 
 compressed: vmlinux
 	$(OBJCOPY) $(OBJCOPYFLAGS) vmlinux vmlinux-tmp
 	gzip vmlinux-tmp
 	mv vmlinux-tmp.gz vmlinux.gz
 
-rawboot:
-	@$(MAKEBOOT) rawboot
-
 archclean:
-	@$(MAKEBOOT) clean
+	$(MAKE) -rR -f scripts/Makefile.clean obj=arch/$(ARCH)/boot
 
 archmrproper:
-	@$(MAKE) -C arch/$(ARCH)/tools mrproper
 
 prepare: $(TOPDIR)/include/asm-ia64/offsets.h
 
 $(TOPDIR)/include/asm-ia64/offsets.h: include/asm include/linux/version.h \
 				      include/config/MARKER
-	@$(MAKE) -C arch/$(ARCH)/tools $@
+	+@$(call maketool,$@)

arch/ia64/boot/Makefile

@@ -8,9 +8,9 @@
 # Copyright (C) 1998 by David Mosberger-Tang <davidm@hpl.hp.com>
 #
 
-LINKFLAGS = -static -T bootloader.lds
+LINKFLAGS = -static -T $(src)/bootloader.lds
 
-OBJECTS	= bootloader.o
+OBJS	= $(obj)/bootloader.o
 
 targets-$(CONFIG_IA64_HP_SIM)	+= bootloader
 targets-$(CONFIG_IA64_GENERIC)	+= bootloader
@@ -19,8 +19,8 @@ CFLAGS	:= $(CFLAGS) $(CFLAGS_KERNEL)
 
 all:	$(targets-y)
 
-bootloader: $(OBJECTS)
-	$(LD) $(LINKFLAGS) $(OBJECTS) $(TOPDIR)/lib/lib.a $(TOPDIR)/arch/$(ARCH)/lib/lib.a \
+bootloader: $(OBJS)
+	$(LD) $(LINKFLAGS) $(OBJS) $(TOPDIR)/lib/lib.a $(TOPDIR)/arch/$(ARCH)/lib/lib.a \
 	-o bootloader
 
 clean:

arch/ia64/config.in

@@ -66,6 +66,10 @@ fi
 
 if [ "$CONFIG_IA64_GENERIC" = "y" -o "$CONFIG_IA64_DIG" = "y" -o "$CONFIG_IA64_HP_ZX1" = "y" ];
 then
+	bool '  Enable NUMA support' CONFIG_NUMA
+	if [ "$CONFIG_NUMA" = "y" ]; then
+		define_bool CONFIG_DISCONTIGMEM y
+	fi
 	bool '  Enable IA-64 Machine Check Abort' CONFIG_IA64_MCA
 	define_bool CONFIG_PM y
 	define_bool CONFIG_IOSAPIC y
@@ -267,6 +271,7 @@ choice 'Physical memory granularity'				\
 
 bool 'Kernel debugging' CONFIG_DEBUG_KERNEL
 if [ "$CONFIG_DEBUG_KERNEL" != "n" ]; then
+   bool '  Load all symbols for debugging/kksymoops' CONFIG_KALLSYMS
    bool '  Print possible IA64 hazards to console' CONFIG_IA64_PRINT_HAZARDS
    bool '  Disable VHPT' CONFIG_DISABLE_VHPT
    bool '  Magic SysRq key' CONFIG_MAGIC_SYSRQ

arch/ia64/hp/Makefile

-# arch/ia64/hp/Makefile
-# Copyright (c) 2002 Matthew Wilcox for Hewlett Packard
-
-obj-$(CONFIG_IA64_GENERIC)	+= sim/ zx1/ common/
-obj-$(CONFIG_IA64_HP_SIM)	+= sim/
-
-include $(TOPDIR)/Rules.make

arch/ia64/hp/sim/simscsi.c

@@ -361,7 +361,9 @@ simscsi_queuecommand (Scsi_Cmnd *sc, void (*done)(Scsi_Cmnd *))
 			break;
 
 		      case MODE_SENSE:
-			printk("MODE_SENSE\n");
+			/* sd.c uses this to determine whether disk does write-caching. */
+			memset(sc->request_buffer, 0, 128);
+			sc->result = GOOD;
 			break;
 
 		      case START_STOP:
@@ -391,6 +393,4 @@ simscsi_queuecommand (Scsi_Cmnd *sc, void (*done)(Scsi_Cmnd *))
 
 static Scsi_Host_Template driver_template = SIMSCSI;
 
-#define __initcall(fn)	late_initcall(fn)
-
 #include "../drivers/scsi/scsi_module.c"

arch/ia64/hp/sim/simserial.c

@@ -752,8 +752,7 @@ static int get_async_struct(int line, struct async_struct **ret_info)
 	info->flags = sstate->flags;
 	info->xmit_fifo_size = sstate->xmit_fifo_size;
 	info->line = line;
-	info->tqueue.routine = do_softint;
-	info->tqueue.data = info;
+	INIT_WORK(&info->work, do_softint, info);
 	info->state = sstate;
 	if (sstate->info) {
 		kfree(info);

arch/ia64/ia32/ia32_ioctl.c

@@ -26,6 +26,12 @@
 #include <linux/if_ppp.h>
 #include <linux/ixjuser.h>
 #include <linux/i2o-dev.h>
+#include <scsi/scsi.h>
+/* Ugly hack. */
+#undef	__KERNEL__
+#include <scsi/scsi_ioctl.h>
+#define	__KERNEL__
+#include <scsi/sg.h>
 
 #include <asm/ia32.h>
 
@@ -60,6 +66,235 @@ put_dirent32 (struct dirent *d, struct linux32_dirent *d32)
 		|| put_user(d->d_reclen, &d32->d_reclen)
 		|| copy_to_user(d32->d_name, d->d_name, namelen + 1));
 }
+/*
+ *  The transform code for the SG_IO ioctl was brazenly lifted from
+ *  the Sparc64 port in the file `arch/sparc64/kernel/ioctl32.c'.
+ *  Thanks to Jakub Jelinek & Eddie C. Dost.
+ */
+typedef struct sg_io_hdr32 {
+	int interface_id;	/* [i] 'S' for SCSI generic (required) */
+	int dxfer_direction;	/* [i] data transfer direction  */
+	char  cmd_len;		/* [i] SCSI command length ( <= 16 bytes) */
+	char  mx_sb_len;		/* [i] max length to write to sbp */
+	short iovec_count;	/* [i] 0 implies no scatter gather */
+	int dxfer_len;		/* [i] byte count of data transfer */
+	int dxferp;		/* [i], [*io] points to data transfer memory
+					      or scatter gather list */
+	int cmdp;		/* [i], [*i] points to command to perform */
+	int sbp;		/* [i], [*o] points to sense_buffer memory */
+	int timeout;		/* [i] MAX_UINT->no timeout (unit: millisec) */
+	int flags;		/* [i] 0 -> default, see SG_FLAG... */
+	int pack_id;		/* [i->o] unused internally (normally) */
+	int usr_ptr;		/* [i->o] unused internally */
+	char  status;		/* [o] scsi status */
+	char  masked_status;	/* [o] shifted, masked scsi status */
+	char  msg_status;	/* [o] messaging level data (optional) */
+	char  sb_len_wr;	/* [o] byte count actually written to sbp */
+	short host_status;	/* [o] errors from host adapter */
+	short driver_status;	/* [o] errors from software driver */
+	int resid;		/* [o] dxfer_len - actual_transferred */
+	int duration;		/* [o] time taken by cmd (unit: millisec) */
+	int info;		/* [o] auxiliary information */
+} sg_io_hdr32_t;  /* 64 bytes long (on IA32) */
+
+struct iovec32 { unsigned int iov_base; int iov_len; };
+
+static int alloc_sg_iovec(sg_io_hdr_t *sgp, int uptr32)
+{
+	struct iovec32 *uiov = (struct iovec32 *) P(uptr32);
+	sg_iovec_t *kiov;
+	int i;
+
+	sgp->dxferp = kmalloc(sgp->iovec_count *
+			      sizeof(sg_iovec_t), GFP_KERNEL);
+	if (!sgp->dxferp)
+		return -ENOMEM;
+	memset(sgp->dxferp, 0,
+	       sgp->iovec_count * sizeof(sg_iovec_t));
+
+	kiov = (sg_iovec_t *) sgp->dxferp;
+	for (i = 0; i < sgp->iovec_count; i++) {
+		int iov_base32;
+		if (__get_user(iov_base32, &uiov->iov_base) ||
+		    __get_user(kiov->iov_len, &uiov->iov_len))
+			return -EFAULT;
+
+		kiov->iov_base = kmalloc(kiov->iov_len, GFP_KERNEL);
+		if (!kiov->iov_base)
+			return -ENOMEM;
+		if (copy_from_user(kiov->iov_base,
+				   (void *) P(iov_base32),
+				   kiov->iov_len))
+			return -EFAULT;
+
+		uiov++;
+		kiov++;
+	}
+
+	return 0;
+}
+
+static int copy_back_sg_iovec(sg_io_hdr_t *sgp, int uptr32)
+{
+	struct iovec32 *uiov = (struct iovec32 *) P(uptr32);
+	sg_iovec_t *kiov = (sg_iovec_t *) sgp->dxferp;
+	int i;
+
+	for (i = 0; i < sgp->iovec_count; i++) {
+		int iov_base32;
+
+		if (__get_user(iov_base32, &uiov->iov_base))
+			return -EFAULT;
+
+		if (copy_to_user((void *) P(iov_base32),
+				 kiov->iov_base,
+				 kiov->iov_len))
+			return -EFAULT;
+
+		uiov++;
+		kiov++;
+	}
+
+	return 0;
+}
+
+static void free_sg_iovec(sg_io_hdr_t *sgp)
+{
+	sg_iovec_t *kiov = (sg_iovec_t *) sgp->dxferp;
+	int i;
+
+	for (i = 0; i < sgp->iovec_count; i++) {
+		if (kiov->iov_base) {
+			kfree(kiov->iov_base);
+			kiov->iov_base = NULL;
+		}
+		kiov++;
+	}
+	kfree(sgp->dxferp);
+	sgp->dxferp = NULL;
+}
+
+static int sg_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
+{
+	sg_io_hdr32_t *sg_io32;
+	sg_io_hdr_t sg_io64;
+	int dxferp32, cmdp32, sbp32;
+	mm_segment_t old_fs;
+	int err = 0;
+
+	sg_io32 = (sg_io_hdr32_t *)arg;
+	err = __get_user(sg_io64.interface_id, &sg_io32->interface_id);
+	err |= __get_user(sg_io64.dxfer_direction, &sg_io32->dxfer_direction);
+	err |= __get_user(sg_io64.cmd_len, &sg_io32->cmd_len);
+	err |= __get_user(sg_io64.mx_sb_len, &sg_io32->mx_sb_len);
+	err |= __get_user(sg_io64.iovec_count, &sg_io32->iovec_count);
+	err |= __get_user(sg_io64.dxfer_len, &sg_io32->dxfer_len);
+	err |= __get_user(sg_io64.timeout, &sg_io32->timeout);
+	err |= __get_user(sg_io64.flags, &sg_io32->flags);
+	err |= __get_user(sg_io64.pack_id, &sg_io32->pack_id);
+
+	sg_io64.dxferp = NULL;
+	sg_io64.cmdp = NULL;
+	sg_io64.sbp = NULL;
+
+	err |= __get_user(cmdp32, &sg_io32->cmdp);
+	sg_io64.cmdp = kmalloc(sg_io64.cmd_len, GFP_KERNEL);
+	if (!sg_io64.cmdp) {
+		err = -ENOMEM;
+		goto out;
+	}
+	if (copy_from_user(sg_io64.cmdp,
+			   (void *) P(cmdp32),
+			   sg_io64.cmd_len)) {
+		err = -EFAULT;
+		goto out;
+	}
+
+	err |= __get_user(sbp32, &sg_io32->sbp);
+	sg_io64.sbp = kmalloc(sg_io64.mx_sb_len, GFP_KERNEL);
+	if (!sg_io64.sbp) {
+		err = -ENOMEM;
+		goto out;
+	}
+	if (copy_from_user(sg_io64.sbp,
+			   (void *) P(sbp32),
+			   sg_io64.mx_sb_len)) {
+		err = -EFAULT;
+		goto out;
+	}
+
+	err |= __get_user(dxferp32, &sg_io32->dxferp);
+	if (sg_io64.iovec_count) {
+		int ret;
+
+		if ((ret = alloc_sg_iovec(&sg_io64, dxferp32))) {
+			err = ret;
+			goto out;
+		}
+	} else {
+		sg_io64.dxferp = kmalloc(sg_io64.dxfer_len, GFP_KERNEL);
+		if (!sg_io64.dxferp) {
+			err = -ENOMEM;
+			goto out;
+		}
+		if (copy_from_user(sg_io64.dxferp,
+				   (void *) P(dxferp32),
+				   sg_io64.dxfer_len)) {
+			err = -EFAULT;
+			goto out;
+		}
+	}
+
+	/* Unused internally, do not even bother to copy it over. */
+	sg_io64.usr_ptr = NULL;
+
+	if (err)
+		return -EFAULT;
+
+	old_fs = get_fs();
+	set_fs (KERNEL_DS);
+	err = sys_ioctl (fd, cmd, (unsigned long) &sg_io64);
+	set_fs (old_fs);
+
+	if (err < 0)
+		goto out;
+
+	err = __put_user(sg_io64.pack_id, &sg_io32->pack_id);
+	err |= __put_user(sg_io64.status, &sg_io32->status);
+	err |= __put_user(sg_io64.masked_status, &sg_io32->masked_status);
+	err |= __put_user(sg_io64.msg_status, &sg_io32->msg_status);
+	err |= __put_user(sg_io64.sb_len_wr, &sg_io32->sb_len_wr);
+	err |= __put_user(sg_io64.host_status, &sg_io32->host_status);
+	err |= __put_user(sg_io64.driver_status, &sg_io32->driver_status);
+	err |= __put_user(sg_io64.resid, &sg_io32->resid);
+	err |= __put_user(sg_io64.duration, &sg_io32->duration);
+	err |= __put_user(sg_io64.info, &sg_io32->info);
+	err |= copy_to_user((void *)P(sbp32), sg_io64.sbp, sg_io64.mx_sb_len);
+	if (sg_io64.dxferp) {
+		if (sg_io64.iovec_count)
+			err |= copy_back_sg_iovec(&sg_io64, dxferp32);
+		else
+			err |= copy_to_user((void *)P(dxferp32),
+					    sg_io64.dxferp,
+					    sg_io64.dxfer_len);
+	}
+	if (err)
+		err = -EFAULT;
+
+out:
+	if (sg_io64.cmdp)
+		kfree(sg_io64.cmdp);
+	if (sg_io64.sbp)
+		kfree(sg_io64.sbp);
+	if (sg_io64.dxferp) {
+		if (sg_io64.iovec_count) {
+			free_sg_iovec(&sg_io64);
+		} else {
+			kfree(sg_io64.dxferp);
+		}
+	}
+	return err;
+}
 
 asmlinkage long
 sys32_ioctl (unsigned int fd, unsigned int cmd, unsigned int arg)
@@ -271,6 +506,9 @@ sys32_ioctl (unsigned int fd, unsigned int cmd, unsigned int arg)
 	      default:
 		return sys_ioctl(fd, cmd, (unsigned long)arg);
 
+		case IOCTL_NR(SG_IO):
+			return(sg_ioctl_trans(fd, cmd, arg));
+
 	}
 	printk("%x:unimplemented IA32 ioctl system call\n", cmd);
 	return -EINVAL;

arch/ia64/ia32/sys_ia32.c

@@ -2842,20 +2842,6 @@ putreg (struct task_struct *child, int regno, unsigned int value)
 	}
 }
 
-static inline void
-ia32f2ia64f (void *dst, void *src)
-{
-	asm volatile ("ldfe f6=[%1];; stf.spill [%0]=f6" :: "r"(dst), "r"(src) : "memory");
-	return;
-}
-
-static inline void
-ia64f2ia32f (void *dst, void *src)
-{
-	asm volatile ("ldf.fill f6=[%1];; stfe [%0]=f6" :: "r"(dst),  "r"(src) : "memory");
-	return;
-}
-
 static void
 put_fpreg (int regno, struct _fpreg_ia32 *reg, struct pt_regs *ptp, struct switch_stack *swp,
 	   int tos)

arch/ia64/kernel/acpi.c

@@ -8,6 +8,9 @@
  *  Copyright (C) 2000 Intel Corp.
  *  Copyright (C) 2000,2001 J.I. Lee <jung-ik.lee@intel.com>
  *  Copyright (C) 2001 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
+ *  Copyright (C) 2001 Jenna Hall <jenna.s.hall@intel.com>
+ *  Copyright (C) 2001 Takayoshi Kochi <t-kouchi@cq.jp.nec.com>
+ *  Copyright (C) 2002 Erich Focht <efocht@ess.nec.de>
  *
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  *
@@ -43,6 +46,7 @@
 #include <asm/machvec.h>
 #include <asm/page.h>
 #include <asm/system.h>
+#include <asm/numa.h>
 
 
 #define PREFIX			"ACPI: "
@@ -447,6 +451,189 @@ acpi_parse_madt (unsigned long phys_addr, unsigned long size)
 }
 
 
+#ifdef CONFIG_ACPI_NUMA
+
+#define SLIT_DEBUG
+
+#define PXM_FLAG_LEN ((MAX_PXM_DOMAINS + 1)/32)
+
+static int __initdata srat_num_cpus;			/* number of cpus */
+static u32 __initdata pxm_flag[PXM_FLAG_LEN];
+#define pxm_bit_set(bit)	(set_bit(bit,(void *)pxm_flag))
+#define pxm_bit_test(bit)	(test_bit(bit,(void *)pxm_flag))
+/* maps to convert between proximity domain and logical node ID */
+int __initdata pxm_to_nid_map[MAX_PXM_DOMAINS];
+int __initdata nid_to_pxm_map[NR_NODES];
+static struct acpi_table_slit __initdata *slit_table;
+
+/*
+ * ACPI 2.0 SLIT (System Locality Information Table)
+ * http://devresource.hp.com/devresource/Docs/TechPapers/IA64/slit.pdf
+ */
+void __init
+acpi_numa_slit_init (struct acpi_table_slit *slit)
+{
+	u32 len;
+
+	len = sizeof(struct acpi_table_header) + 8
+		+ slit->localities * slit->localities;
+	if (slit->header.length != len) {
+		printk("ACPI 2.0 SLIT: size mismatch: %d expected, %d actual\n",
+		      len, slit->header.length);
+		memset(numa_slit, 10, sizeof(numa_slit));
+		return;
+	}
+	slit_table = slit;
+}
+
+void __init
+acpi_numa_processor_affinity_init (struct acpi_table_processor_affinity *pa)
+{
+	/* record this node in proximity bitmap */
+	pxm_bit_set(pa->proximity_domain);
+
+	node_cpuid[srat_num_cpus].phys_id = (pa->apic_id << 8) | (pa->lsapic_eid);
+	/* nid should be overridden as logical node id later */
+	node_cpuid[srat_num_cpus].nid = pa->proximity_domain;
+	srat_num_cpus++;
+}
+
+void __init
+acpi_numa_memory_affinity_init (struct acpi_table_memory_affinity *ma)
+{
+	unsigned long paddr, size, hole_size, min_hole_size;
+	u8 pxm;
+	struct node_memblk_s *p, *q, *pend;
+
+	pxm = ma->proximity_domain;
+
+	/* fill node memory chunk structure */
+	paddr = ma->base_addr_hi;
+	paddr = (paddr << 32) | ma->base_addr_lo;
+	size = ma->length_hi;
+	size = (size << 32) | ma->length_lo;
+
+	if (num_memblks >= NR_MEMBLKS) {
+		printk("Too many mem chunks in SRAT. Ignoring %ld MBytes at %lx\n",
+			size/(1024*1024), paddr);
+		return;
+	}
+
+	/* Ignore disabled entries */
+	if (!ma->flags.enabled)
+		return;
+
+	/*
+	 * When the chunk is not the first one in the node, check distance
+	 * from the other chunks. When the hole is too huge ignore the chunk.
+	 * This restriction should be removed when multiple chunks per node
+	 * is supported.
+	 */
+	pend = &node_memblk[num_memblks];
+	min_hole_size = 0;
+	for (p = &node_memblk[0]; p < pend; p++) {
+		if (p->nid != pxm)
+			continue;
+		if (p->start_paddr < paddr)
+			hole_size = paddr - (p->start_paddr + p->size);
+		else
+			hole_size = p->start_paddr - (paddr + size);
+
+		if (!min_hole_size || hole_size < min_hole_size)
+			min_hole_size = hole_size;
+	}
+
+	if (min_hole_size) {
+		if (min_hole_size > size) {
+			printk("Too huge memory hole. Ignoring %ld MBytes at %lx\n",
+				size/(1024*1024), paddr);
+			return;
+		}
+	}
+
+	/* record this node in proximity bitmap */
+	pxm_bit_set(pxm);
+
+	/* Insertion sort based on base address */
+	pend = &node_memblk[num_memblks];
+	for (p = &node_memblk[0]; p < pend; p++) {
+		if (paddr < p->start_paddr)
+			break;
+	}
+	if (p < pend) {
+		for (q = pend; q >= p; q--)
+			*(q + 1) = *q;
+	}
+	p->start_paddr = paddr;
+	p->size = size;
+	p->nid = pxm;
+	num_memblks++;
+}
+
+void __init
+acpi_numa_arch_fixup(void)
+{
+	int i, j, node_from, node_to;
+
+	/* calculate total number of nodes in system from PXM bitmap */
+	numnodes = 0;		/* init total nodes in system */
+
+	memset(pxm_to_nid_map, -1, sizeof(pxm_to_nid_map));
+	memset(nid_to_pxm_map, -1, sizeof(nid_to_pxm_map));
+	for (i = 0; i < MAX_PXM_DOMAINS; i++) {
+		if (pxm_bit_test(i)) {
+			pxm_to_nid_map[i] = numnodes;
+			nid_to_pxm_map[numnodes++] = i;
+		}
+	}
+
+	/* set logical node id in memory chunk structure */
+	for (i = 0; i < num_memblks; i++)
+		node_memblk[i].nid = pxm_to_nid_map[node_memblk[i].nid];
+
+	/* assign memory bank numbers for each chunk on each node */
+	for (i = 0; i < numnodes; i++) {
+		int bank;
+
+		bank = 0;
+		for (j = 0; j < num_memblks; j++)
+			if (node_memblk[j].nid == i)
+				node_memblk[j].bank = bank++;
+	}
+
+	/* set logical node id in cpu structure */
+	for (i = 0; i < srat_num_cpus; i++)
+		node_cpuid[i].nid = pxm_to_nid_map[node_cpuid[i].nid];
+
+	printk("Number of logical nodes in system = %d\n", numnodes);
+	printk("Number of memory chunks in system = %d\n", num_memblks);
+
+	if (!slit_table) return;
+	memset(numa_slit, -1, sizeof(numa_slit));
+	for (i=0; i<slit_table->localities; i++) {
+		if (!pxm_bit_test(i))
+			continue;
+		node_from = pxm_to_nid_map[i];
+		for (j=0; j<slit_table->localities; j++) {
+			if (!pxm_bit_test(j))
+				continue;
+			node_to = pxm_to_nid_map[j];
+			node_distance(node_from, node_to) = 
+				slit_table->entry[i*slit_table->localities + j];
+		}
+	}
+
+#ifdef SLIT_DEBUG
+	printk("ACPI 2.0 SLIT locality table:\n");
+	for (i = 0; i < numnodes; i++) {
+		for (j = 0; j < numnodes; j++)
+			printk("%03d ", node_distance(i,j));
+		printk("\n");
+	}
+#endif
+}
+#endif /* CONFIG_ACPI_NUMA */
+
 static int __init
 acpi_parse_fadt (unsigned long phys_addr, unsigned long size)
 {
@@ -556,12 +743,6 @@ acpi_parse_spcr (unsigned long phys_addr, unsigned long size)
 int __init
 acpi_boot_init (char *cmdline)
 {
-	int result;
-
-	/* Initialize the ACPI boot-time table parser */
-	result = acpi_table_init(cmdline);
-	if (result)
-		return result;
 
 	/*
 	 * MADT
@@ -631,6 +812,9 @@ acpi_boot_init (char *cmdline)
 	smp_boot_data.cpu_count = total_cpus;
 
 	smp_build_cpu_map();
+#ifdef CONFIG_NUMA
+	build_cpu_to_node_map();
+#endif
 #endif
 	/* Make boot-up look pretty */
 	printk("%d CPUs available, %d CPUs total\n", available_cpus, total_cpus);

arch/ia64/kernel/efi.c

@@ -60,67 +60,157 @@ struct proc_dir_entry *efi_dir = NULL;
 
 static unsigned long mem_limit = ~0UL;
 
-static efi_status_t
-phys_get_time (efi_time_t *tm, efi_time_cap_t *tc)
-{
-	return efi_call_phys(__va(runtime->get_time), __pa(tm), __pa(tc));
+#define efi_call_virt(f, args...)	(*(f))(args)
+
+#define STUB_GET_TIME(prefix, adjust_arg)							\
+static efi_status_t										\
+prefix##_get_time (efi_time_t *tm, efi_time_cap_t *tc)						\
+{												\
+	struct ia64_fpreg fr[6];								\
+	efi_status_t ret;									\
+												\
+	ia64_save_scratch_fpregs(fr);								\
+	ret = efi_call_##prefix((efi_get_time_t *) __va(runtime->get_time), adjust_arg(tm),	\
+				adjust_arg(tc));						\
+	ia64_load_scratch_fpregs(fr);								\
+	return ret;										\
 }
 
-static efi_status_t
-phys_set_time (efi_time_t *tm)
-{
-	return efi_call_phys(__va(runtime->set_time), __pa(tm));
+#define STUB_SET_TIME(prefix, adjust_arg)							\
+static efi_status_t										\
+prefix##_set_time (efi_time_t *tm)								\
+{												\
+	struct ia64_fpreg fr[6];								\
+	efi_status_t ret;									\
+												\
+	ia64_save_scratch_fpregs(fr);								\
+	ret = efi_call_##prefix((efi_set_time_t *) __va(runtime->set_time), adjust_arg(tm));	\
+	ia64_load_scratch_fpregs(fr);								\
+	return ret;										\
 }
 
-static efi_status_t
-phys_get_wakeup_time (efi_bool_t *enabled, efi_bool_t *pending, efi_time_t *tm)
-{
-	return efi_call_phys(__va(runtime->get_wakeup_time), __pa(enabled), __pa(pending),
-			     __pa(tm));
+#define STUB_GET_WAKEUP_TIME(prefix, adjust_arg)						\
+static efi_status_t										\
+prefix##_get_wakeup_time (efi_bool_t *enabled, efi_bool_t *pending, efi_time_t *tm)		\
+{												\
+	struct ia64_fpreg fr[6];								\
+	efi_status_t ret;									\
+												\
+	ia64_save_scratch_fpregs(fr);								\
+	ret = efi_call_##prefix((efi_get_wakeup_time_t *) __va(runtime->get_wakeup_time),	\
+				adjust_arg(enabled), adjust_arg(pending), adjust_arg(tm));	\
+	ia64_load_scratch_fpregs(fr);								\
+	return ret;										\
 }
 
-static efi_status_t
-phys_set_wakeup_time (efi_bool_t enabled, efi_time_t *tm)
-{
-	return efi_call_phys(__va(runtime->set_wakeup_time), enabled, __pa(tm));
+#define STUB_SET_WAKEUP_TIME(prefix, adjust_arg)						\
+static efi_status_t										\
+prefix##_set_wakeup_time (efi_bool_t enabled, efi_time_t *tm)					\
+{												\
+	struct ia64_fpreg fr[6];								\
+	efi_status_t ret;									\
+												\
+	ia64_save_scratch_fpregs(fr);								\
+	ret = efi_call_##prefix((efi_set_wakeup_time_t *) __va(runtime->set_wakeup_time),	\
+				enabled, adjust_arg(tm));					\
+	ia64_load_scratch_fpregs(fr);								\
+	return ret;										\
 }
 
-static efi_status_t
-phys_get_variable (efi_char16_t *name, efi_guid_t *vendor, u32 *attr,
-		   unsigned long *data_size, void *data)
-{
-	return efi_call_phys(__va(runtime->get_variable), __pa(name), __pa(vendor), __pa(attr),
-			     __pa(data_size), __pa(data));
+#define STUB_GET_VARIABLE(prefix, adjust_arg)						\
+static efi_status_t									\
+prefix##_get_variable (efi_char16_t *name, efi_guid_t *vendor, u32 *attr,		\
+		       unsigned long *data_size, void *data)				\
+{											\
+	struct ia64_fpreg fr[6];							\
+	efi_status_t ret;								\
+											\
+	ia64_save_scratch_fpregs(fr);							\
+	ret = efi_call_##prefix((efi_get_variable_t *) __va(runtime->get_variable),	\
+				adjust_arg(name), adjust_arg(vendor), adjust_arg(attr),	\
+				adjust_arg(data_size), adjust_arg(data));		\
+	ia64_load_scratch_fpregs(fr);							\
+	return ret;									\
 }
 
-static efi_status_t
-phys_get_next_variable (unsigned long *name_size, efi_char16_t *name, efi_guid_t *vendor)
-{
-	return efi_call_phys(__va(runtime->get_next_variable), __pa(name_size), __pa(name),
-			     __pa(vendor));
+#define STUB_GET_NEXT_VARIABLE(prefix, adjust_arg)						\
+static efi_status_t										\
+prefix##_get_next_variable (unsigned long *name_size, efi_char16_t *name, efi_guid_t *vendor)	\
+{												\
+	struct ia64_fpreg fr[6];								\
+	efi_status_t ret;									\
+												\
+	ia64_save_scratch_fpregs(fr);								\
+	ret = efi_call_##prefix((efi_get_next_variable_t *) __va(runtime->get_next_variable),	\
+				adjust_arg(name_size), adjust_arg(name), adjust_arg(vendor));	\
+	ia64_load_scratch_fpregs(fr);								\
+	return ret;										\
 }
 
-static efi_status_t
-phys_set_variable (efi_char16_t *name, efi_guid_t *vendor, u32 attr,
-		   unsigned long data_size, void *data)
-{
-	return efi_call_phys(__va(runtime->set_variable), __pa(name), __pa(vendor), attr,
-			     data_size, __pa(data));
+#define STUB_SET_VARIABLE(prefix, adjust_arg)						\
+static efi_status_t									\
+prefix##_set_variable (efi_char16_t *name, efi_guid_t *vendor, u32 attr,		\
+		       unsigned long data_size, void *data)				\
+{											\
+	struct ia64_fpreg fr[6];							\
+	efi_status_t ret;								\
+											\
+	ia64_save_scratch_fpregs(fr);							\
+	ret = efi_call_##prefix((efi_set_variable_t *) __va(runtime->set_variable),	\
+				adjust_arg(name), adjust_arg(vendor), attr, data_size,	\
+				adjust_arg(data));					\
+	ia64_load_scratch_fpregs(fr);							\
+	return ret;									\
 }
 
-static efi_status_t
-phys_get_next_high_mono_count (u64 *count)
-{
-	return efi_call_phys(__va(runtime->get_next_high_mono_count), __pa(count));
+#define STUB_GET_NEXT_HIGH_MONO_COUNT(prefix, adjust_arg)					\
+static efi_status_t										\
+prefix##_get_next_high_mono_count (u64 *count)							\
+{												\
+	struct ia64_fpreg fr[6];								\
+	efi_status_t ret;									\
+												\
+	ia64_save_scratch_fpregs(fr);								\
+	ret = efi_call_##prefix((efi_get_next_high_mono_count_t *)				\
+				__va(runtime->get_next_high_mono_count), adjust_arg(count));	\
+	ia64_load_scratch_fpregs(fr);								\
+	return ret;										\
 }
 
-static void
-phys_reset_system (int reset_type, efi_status_t status,
-		   unsigned long data_size, efi_char16_t *data)
-{
-	efi_call_phys(__va(runtime->reset_system), status, data_size, __pa(data));
+#define STUB_RESET_SYSTEM(prefix, adjust_arg)					\
+static void									\
+prefix##_reset_system (int reset_type, efi_status_t status,			\
+		       unsigned long data_size, efi_char16_t *data)		\
+{										\
+	struct ia64_fpreg fr[6];						\
+										\
+	ia64_save_scratch_fpregs(fr);						\
+	efi_call_##prefix((efi_reset_system_t *) __va(runtime->reset_system),	\
+			  reset_type, status, data_size, adjust_arg(data));	\
+	/* should not return, but just in case... */				\
+	ia64_load_scratch_fpregs(fr);						\
 }
 
+STUB_GET_TIME(phys, __pa)
+STUB_SET_TIME(phys, __pa)
+STUB_GET_WAKEUP_TIME(phys, __pa)
+STUB_SET_WAKEUP_TIME(phys, __pa)
+STUB_GET_VARIABLE(phys, __pa)
+STUB_GET_NEXT_VARIABLE(phys, __pa)
+STUB_SET_VARIABLE(phys, __pa)
+STUB_GET_NEXT_HIGH_MONO_COUNT(phys, __pa)
+STUB_RESET_SYSTEM(phys, __pa)
+
+STUB_GET_TIME(virt, )
+STUB_SET_TIME(virt, )
+STUB_GET_WAKEUP_TIME(virt, )
+STUB_SET_WAKEUP_TIME(virt, )
+STUB_GET_VARIABLE(virt, )
+STUB_GET_NEXT_VARIABLE(virt, )
+STUB_SET_VARIABLE(virt, )
+STUB_GET_NEXT_HIGH_MONO_COUNT(virt, )
+STUB_RESET_SYSTEM(virt, )
+
 void
 efi_gettimeofday (struct timeval *tv)
 {
@@ -574,18 +664,17 @@ efi_enter_virtual_mode (void)
 	}
 
 	/*
-	 * Now that EFI is in virtual mode, we arrange for EFI functions to be
-	 * called directly:
+	 * Now that EFI is in virtual mode, we call the EFI functions more efficiently:
 	 */
-	efi.get_time = __va(runtime->get_time);
-	efi.set_time = __va(runtime->set_time);
-	efi.get_wakeup_time = __va(runtime->get_wakeup_time);
-	efi.set_wakeup_time = __va(runtime->set_wakeup_time);
-	efi.get_variable = __va(runtime->get_variable);
-	efi.get_next_variable = __va(runtime->get_next_variable);
-	efi.set_variable = __va(runtime->set_variable);
-	efi.get_next_high_mono_count = __va(runtime->get_next_high_mono_count);
-	efi.reset_system = __va(runtime->reset_system);
+	efi.get_time = virt_get_time;
+	efi.set_time = virt_set_time;
+	efi.get_wakeup_time = virt_get_wakeup_time;
+	efi.set_wakeup_time = virt_set_wakeup_time;
+	efi.get_variable = virt_get_variable;
+	efi.get_next_variable = virt_get_next_variable;
+	efi.set_variable = virt_set_variable;
+	efi.get_next_high_mono_count = virt_get_next_high_mono_count;
+	efi.reset_system = virt_reset_system;
 }
 
 /*

arch/ia64/kernel/fw-emu.c

@@ -2,7 +2,7 @@
  * PAL & SAL emulation.
  *
  * Copyright (C) 1998-2001 Hewlett-Packard Co
- * Copyright (C) 1998-2001 David Mosberger-Tang <davidm@hpl.hp.com>
+ *	David Mosberger-Tang <davidm@hpl.hp.com>
  *
  * For the HP simulator, this file gets include in boot/bootloader.c.
  * For SoftSDV, this file gets included in sys_softsdv.c.

arch/ia64/kernel/iosapic.c

@@ -133,12 +133,8 @@ find_iosapic (unsigned int gsi)
 	return -1;
 }
 
-/*
- * Translate GSI number to the corresponding IA-64 interrupt vector.  If no
- * entry exists, return -1.
- */
-int
-gsi_to_vector (unsigned int gsi)
+static inline int
+_gsi_to_vector (unsigned int gsi)
 {
 	struct iosapic_intr_info *info;
 
@@ -148,6 +144,26 @@ gsi_to_vector (unsigned int gsi)
 	return -1;
 }
 
+/*
+ * Translate GSI number to the corresponding IA-64 interrupt vector.  If no
+ * entry exists, return -1.
+ */
+inline int
+gsi_to_vector (unsigned int gsi)
+{
+	return _gsi_to_vector(gsi);
+}
+
+int
+gsi_to_irq (unsigned int gsi)
+{
+	/*
+	 * XXX fix me: this assumes an identity mapping vetween IA-64 vector and Linux irq
+	 * numbers...
+	 */
+	return _gsi_to_vector(gsi);
+}
+
 static void
 set_rte (unsigned int vector, unsigned int dest)
 {
@@ -157,7 +173,7 @@ set_rte (unsigned int vector, unsigned int dest)
 	int rte_index;
 	char redir;
 
-	DBG(KERN_DEBUG"IOSAPIC: routing vector %d to %x\n", vector, dest);
+	DBG(KERN_DEBUG"IOSAPIC: routing vector %d to 0x%x\n", vector, dest);
 
 	rte_index = iosapic_intr_info[vector].rte_index;
 	if (rte_index < 0)

arch/ia64/kernel/ivt.S

@@ -116,8 +116,8 @@ ENTRY(vhpt_miss)
 	;;
 (p8)	dep r25=r18,r25,2,6
 (p8)	shr r22=r22,HPAGE_SHIFT-PAGE_SHIFT
-	;;
 #endif
+	;;
 	cmp.eq p6,p7=5,r17			// is IFA pointing into to region 5?
 	shr.u r18=r22,PGDIR_SHIFT		// get bits 33-63 of the faulting address
 	;;

arch/ia64/kernel/mca.c

@@ -1012,28 +1012,13 @@ ia64_log_prt_oem_data (int header_len, int sect_len, u8 *p_data, prfunc_t prfunc
 void
 ia64_log_rec_header_print (sal_log_record_header_t *lh, prfunc_t prfunc)
 {
-	char str_buf[32];
-
-	sprintf(str_buf, "%2d.%02d",
-		(lh->revision.major >> 4) * 10 + (lh->revision.major & 0xf),
-		(lh->revision.minor >> 4) * 10 + (lh->revision.minor & 0xf));
-	prfunc("+Err Record ID: %d    SAL Rev: %s\n", lh->id, str_buf);
-	sprintf(str_buf, "%02d/%02d/%04d/ %02d:%02d:%02d",
-		(lh->timestamp.slh_month >> 4) * 10 +
-		(lh->timestamp.slh_month & 0xf),
-		(lh->timestamp.slh_day >> 4) * 10 +
-		(lh->timestamp.slh_day & 0xf),
-		(lh->timestamp.slh_century >> 4) * 1000 +
-		(lh->timestamp.slh_century & 0xf) * 100 +
-		(lh->timestamp.slh_year >> 4) * 10 +
-		(lh->timestamp.slh_year & 0xf),
-		(lh->timestamp.slh_hour >> 4) * 10 +
-		(lh->timestamp.slh_hour & 0xf),
-		(lh->timestamp.slh_minute >> 4) * 10 +
-		(lh->timestamp.slh_minute & 0xf),
-		(lh->timestamp.slh_second >> 4) * 10 +
-		(lh->timestamp.slh_second & 0xf));
-	prfunc("+Time: %s    Severity %d\n", str_buf, lh->severity);
+	prfunc("+Err Record ID: %d    SAL Rev: %2x.%02x\n", lh->id,
+			lh->revision.major, lh->revision.minor);
+	prfunc("+Time: %02x/%02x/%02x%02x %02x:%02x:%02x    Severity %d\n",
+			lh->timestamp.slh_month, lh->timestamp.slh_day,
+			lh->timestamp.slh_century, lh->timestamp.slh_year,
+			lh->timestamp.slh_hour, lh->timestamp.slh_minute,
+			lh->timestamp.slh_second, lh->severity);
 }
 
 /*

arch/ia64/kernel/perfmon.c

@@ -63,6 +63,26 @@
  * Misc macros and definitions
  */
 #define PMU_FIRST_COUNTER	4
+#define PMU_MAX_PMCS		256
+#define PMU_MAX_PMDS		256
+
+/*
+ * type of a PMU register (bitmask).
+ * bitmask structure:
+ * 	bit0   : register implemented
+ * 	bit1   : end marker 
+ * 	bit2-3 : reserved
+ * 	bit4-7 : register type
+ * 	bit8-31: reserved
+ */
+#define PFM_REG_IMPL		0x1 /* register implemented */
+#define PFM_REG_END		0x2 /* end marker */
+#define PFM_REG_MONITOR		(0x1<<4|PFM_REG_IMPL) /* a PMC with a pmc.pm field only */
+#define PFM_REG_COUNTING	(0x2<<4|PFM_REG_IMPL) /* a PMC with a pmc.pm AND pmc.oi, a PMD used as a counter */
+#define PFM_REG_CONTROL		(0x3<<4|PFM_REG_IMPL) /* PMU control register */
+#define	PFM_REG_CONFIG		(0x4<<4|PFM_REG_IMPL) /* refine configuration */
+#define PFM_REG_BUFFER	 	(0x5<<4|PFM_REG_IMPL) /* PMD used as buffer */
+
 
 #define PFM_IS_DISABLED() pmu_conf.pfm_is_disabled
 
@@ -70,13 +90,18 @@
 #define PFM_FL_INHERIT_MASK	(PFM_FL_INHERIT_NONE|PFM_FL_INHERIT_ONCE|PFM_FL_INHERIT_ALL)
 
 /* i assume unsigned */
-#define PMC_IS_IMPL(i)	  (i<pmu_conf.num_pmcs && pmu_conf.impl_regs[i>>6] & (1UL<< (i) %64))
-#define PMD_IS_IMPL(i)	  (i<pmu_conf.num_pmds &&  pmu_conf.impl_regs[4+(i>>6)] & (1UL<<(i) % 64))
+#define PMC_IS_IMPL(i)	  (i< PMU_MAX_PMCS && (pmu_conf.pmc_desc[i].type & PFM_REG_IMPL))
+#define PMD_IS_IMPL(i)	  (i< PMU_MAX_PMDS && (pmu_conf.pmd_desc[i].type & PFM_REG_IMPL))
 
 /* XXX: these three assume that register i is implemented */
 #define PMD_IS_COUNTING(i) (pmu_conf.pmd_desc[i].type == PFM_REG_COUNTING)
 #define PMC_IS_COUNTING(i) (pmu_conf.pmc_desc[i].type == PFM_REG_COUNTING)
 #define PMC_IS_MONITOR(i)  (pmu_conf.pmc_desc[i].type == PFM_REG_MONITOR)
+#define PMC_DFL_VAL(i)     pmu_conf.pmc_desc[i].default_value
+#define PMC_RSVD_MASK(i)   pmu_conf.pmc_desc[i].reserved_mask
+#define PMD_PMD_DEP(i)	   pmu_conf.pmd_desc[i].dep_pmd[0]
+#define PMC_PMD_DEP(i)	   pmu_conf.pmc_desc[i].dep_pmd[0]
+
 
 /* k assume unsigned */
 #define IBR_IS_IMPL(k)	  (k<pmu_conf.num_ibrs)
@@ -175,19 +200,6 @@ typedef struct _pfm_smpl_buffer_desc {
 #define LOCK_PSB(p)	spin_lock(&(p)->psb_lock)
 #define UNLOCK_PSB(p)	spin_unlock(&(p)->psb_lock)
 
-/*
- * The possible type of a PMU register
- */
-typedef enum { 
-	PFM_REG_NOTIMPL, /* not implemented */
-	PFM_REG_NONE, 	 /* end marker */
-	PFM_REG_MONITOR, /* a PMC with a pmc.pm field only */
-	PFM_REG_COUNTING,/* a PMC with a pmc.pm AND pmc.oi, a PMD used as a counter */
-	PFM_REG_CONTROL, /* PMU control register */
-	PFM_REG_CONFIG,  /* refine configuration */
-	PFM_REG_BUFFER	 /* PMD used as buffer */
-} pfm_pmu_reg_type_t;
-
 /*
  * 64-bit software counter structure
  */
@@ -283,13 +295,16 @@ typedef struct {
  * dep_pmc[]: a bitmask of dependent PMC registers
  */
 typedef struct {
-	pfm_pmu_reg_type_t	type;
+	unsigned int		type;
 	int			pm_pos;
+	unsigned long		default_value;	/* power-on default value */
+	unsigned long		reserved_mask;	/* bitmask of reserved bits */
 	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;
+
 /* assume cnum is a valid monitor */
 #define PMC_PM(cnum, val)	(((val) >> (pmu_conf.pmc_desc[cnum].pm_pos)) & 0x1)
 #define PMC_WR_FUNC(cnum)	(pmu_conf.pmc_desc[cnum].write_check)
@@ -401,8 +416,6 @@ static ctl_table pfm_sysctl_root[] = {
 };
 static struct ctl_table_header *pfm_sysctl_header;
 
-static unsigned long reset_pmcs[IA64_NUM_PMC_REGS];	/* contains PAL reset values for PMCS */
-
 static void pfm_vm_close(struct vm_area_struct * area);
 
 static struct vm_operations_struct pfm_vm_ops={
@@ -422,7 +435,7 @@ static struct {
 /*
  * forward declarations
  */
-static void ia64_reset_pmu(struct task_struct *);
+static void pfm_reset_pmu(struct task_struct *);
 #ifdef CONFIG_SMP
 static void pfm_fetch_regs(int cpu, struct task_struct *task, pfm_context_t *ctx);
 #endif
@@ -2244,7 +2257,7 @@ pfm_enable(struct task_struct *task, pfm_context_t *ctx, void *arg, int count,
 		pfm_lazy_save_regs(PMU_OWNER());
 
 	/* reset all registers to stable quiet state */
-	ia64_reset_pmu(task);
+	pfm_reset_pmu(task);
 
 	/* make sure nothing starts */
 	if (ctx->ctx_fl_system) {
@@ -2307,7 +2320,7 @@ pfm_get_pmc_reset(struct task_struct *task, pfm_context_t *ctx, void *arg, int c
 
 		if (!PMC_IS_IMPL(cnum)) goto abort_mission;
 
-		tmp.reg_value = reset_pmcs[cnum];
+		tmp.reg_value = PMC_DFL_VAL(cnum);
 
 		PFM_REG_RETFLAG_SET(tmp.reg_flags, 0);
 
@@ -2998,6 +3011,8 @@ perfmon_proc_info(char *page)
 		p += sprintf(p, "CPU%-2d recorded samples : %lu\n", i, pfm_stats[i].pfm_recorded_samples_count);
 		p += sprintf(p, "CPU%-2d smpl buffer full : %lu\n", i, pfm_stats[i].pfm_full_smpl_buffer_count);
 		p += sprintf(p, "CPU%-2d owner            : %d\n", i, pmu_owners[i].owner ? pmu_owners[i].owner->pid: -1);
+		p += sprintf(p, "CPU%-2d syst_wide        : %d\n", i, per_cpu(pfm_syst_wide, i));
+		p += sprintf(p, "CPU%-2d dcr_pp           : %d\n", i, per_cpu(pfm_dcr_pp, i));
 	}
 
 	LOCK_PFS();
@@ -3398,11 +3413,10 @@ pfm_load_regs (struct task_struct *task)
  * XXX: make this routine able to work with non current context
  */
 static void
-ia64_reset_pmu(struct task_struct *task)
+pfm_reset_pmu(struct task_struct *task)
 {
 	struct thread_struct *t = &task->thread;
 	pfm_context_t *ctx = t->pfm_context;
-	unsigned long mask;
 	int i;
 
 	if (task != current) {
@@ -3415,30 +3429,27 @@ ia64_reset_pmu(struct task_struct *task)
 
 	/*
 	 * install reset values for PMC. We skip PMC0 (done above)
-	 * XX: good up to 64 PMCS
 	 */
-	mask = pmu_conf.impl_regs[0] >> 1;
-	for(i=1; mask; mask>>=1, i++) {
-		if (mask & 0x1) {
-			ia64_set_pmc(i, reset_pmcs[i]);
-			/*
-			 * When restoring context, we must restore ALL pmcs, even the ones 
-			 * that the task does not use to avoid leaks and possibly corruption
-			 * of the sesion because of configuration conflicts. So here, we 
-			 * initialize the entire set used in the context switch restore routine.
-	 		 */
-			t->pmc[i] = reset_pmcs[i];
-			DBprintk((" pmc[%d]=0x%lx\n", i, reset_pmcs[i]));
-						 
-		}
+	for (i=1; (pmu_conf.pmc_desc[i].type & PFM_REG_END) == 0; i++) {
+		if ((pmu_conf.pmc_desc[i].type & PFM_REG_IMPL) == 0) continue;
+		ia64_set_pmc(i, PMC_DFL_VAL(i));
+		/*
+		 * When restoring context, we must restore ALL pmcs, even the ones 
+		 * that the task does not use to avoid leaks and possibly corruption
+		 * of the sesion because of configuration conflicts. So here, we 
+		 * initialize the entire set used in the context switch restore routine.
+	 	 */
+		t->pmc[i] = PMC_DFL_VAL(i);
+		DBprintk(("pmc[%d]=0x%lx\n", i, t->pmc[i]));
 	}
+
 	/*
 	 * clear reset values for PMD. 
 	 * XXX: good up to 64 PMDS. Suppose that zero is a valid value.
 	 */
-	mask = pmu_conf.impl_regs[4];
-	for(i=0; mask; mask>>=1, i++) {
-		if (mask & 0x1) ia64_set_pmd(i, 0UL);
+	for (i=0; (pmu_conf.pmd_desc[i].type & PFM_REG_END) == 0; i++) {
+		if ((pmu_conf.pmd_desc[i].type & PFM_REG_IMPL) == 0) continue;
+		ia64_set_pmd(i, 0UL);
 		t->pmd[i] = 0UL;
 	}
 
@@ -4119,23 +4130,6 @@ static struct irqaction perfmon_irqaction = {
 };
 
 
-static void
-pfm_pmu_snapshot(void)
-{
-	int i;
-
-	for (i=0; i < IA64_NUM_PMC_REGS; i++) {
-		if (i >= pmu_conf.num_pmcs) break;
-		if (PMC_IS_IMPL(i)) reset_pmcs[i] = ia64_get_pmc(i);
-	}
-#ifdef CONFIG_MCKINLEY
-	/*
-	 * set the 'stupid' enable bit to power the PMU!
-	 */
-	reset_pmcs[4] |= 1UL << 23;
-#endif
-}
-
 /*
  * perfmon initialization routine, called from the initcall() table
  */
@@ -4160,6 +4154,9 @@ perfmon_init (void)
 	}
 
 	pmu_conf.perf_ovfl_val = (1UL << pm_info.pal_perf_mon_info_s.width) - 1;
+	/*
+	 * XXX: use the pfm_*_desc tables instead and simply verify with PAL
+	 */
 	pmu_conf.max_counters  = pm_info.pal_perf_mon_info_s.generic;
 	pmu_conf.num_pmcs      = find_num_pm_regs(pmu_conf.impl_regs);
 	pmu_conf.num_pmds      = find_num_pm_regs(&pmu_conf.impl_regs[4]);
@@ -4183,24 +4180,11 @@ perfmon_init (void)
 	pmu_conf.num_ibrs <<=1;
 	pmu_conf.num_dbrs <<=1;
 
-	/*
-	 * take a snapshot of all PMU registers. PAL is supposed
-	 * to configure them with stable/safe values, i.e., not
-	 * capturing anything.
-	 * We take a snapshot now, before we make any modifications. This
-	 * will become our master copy. Then we will reuse the snapshot
-	 * to reset the PMU in pfm_enable(). Using this technique, perfmon
-	 * does NOT have to know about the specific values to program for
-	 * the PMC/PMD. The safe values may be different from one CPU model to
-	 * the other.
-	 */
-	pfm_pmu_snapshot();
-
 	/*
 	 * setup the register configuration descriptions for the CPU
 	 */
-	pmu_conf.pmc_desc = pmc_desc;
-	pmu_conf.pmd_desc = pmd_desc;
+	pmu_conf.pmc_desc = pfm_pmc_desc;
+	pmu_conf.pmd_desc = pfm_pmd_desc;
 
 	/* we are all set */
 	pmu_conf.pfm_is_disabled = 0;
@@ -4222,11 +4206,34 @@ __initcall(perfmon_init);
 void
 perfmon_init_percpu (void)
 {
+	int i;
+
 	if (smp_processor_id() == 0)
 		register_percpu_irq(IA64_PERFMON_VECTOR, &perfmon_irqaction);
 
 	ia64_set_pmv(IA64_PERFMON_VECTOR);
 	ia64_srlz_d();
+
+	/*
+	 * we first initialize the PMU to a stable state.
+	 * the values may have been changed from their power-up
+	 * values by software executed before the kernel took over.
+	 *
+	 * At this point, pmu_conf has not yet been initialized
+	 *
+	 * On McKinley, this code is ineffective until PMC4 is initialized.
+	 */
+	for (i=1; (pfm_pmc_desc[i].type & PFM_REG_END) == 0;  i++) {
+		if ((pfm_pmc_desc[i].type & PFM_REG_IMPL) == 0) continue;
+		ia64_set_pmc(i, pfm_pmc_desc[i].default_value);
+	}
+	for (i=0; (pfm_pmd_desc[i].type & PFM_REG_END) == 0; i++) {
+		if ((pfm_pmd_desc[i].type & PFM_REG_IMPL) == 0) continue;
+		ia64_set_pmd(i, 0UL);
+	}
+	ia64_set_pmc(0,1UL);
+	ia64_srlz_d();
+
 }
 
 #else /* !CONFIG_PERFMON */

arch/ia64/kernel/perfmon_generic.h

+#define RDEP(x)	(1UL<<(x))
+
+#if defined(CONFIG_ITANIUM) || defined(CONFIG_MCKINLEY)
+#error "This file should only be used when CONFIG_ITANIUM and CONFIG_MCKINLEY are not defined"
+#endif
+
+static pfm_reg_desc_t pmc_desc[PMU_MAX_PMCS]={
+/* pmc0  */ { PFM_REG_CONTROL , 0, 0x1UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc1  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc2  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc3  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc4  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(4),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc5  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(5),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc6  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(6),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc7  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {RDEP(7),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+	     { PFM_REG_END     , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}}, /* end marker */
+};
+
+static pfm_reg_desc_t pmd_desc[PMU_MAX_PMDS]={
+/* pmd0  */ { PFM_REG_NOTIMPL , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}},
+/* pmd1  */ { PFM_REG_NOTIMPL , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}},
+/* pmd2  */ { PFM_REG_NOTIMPL , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}},
+/* pmd3  */ { PFM_REG_NOTIMPL , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}},
+/* pmd4  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(4),0UL, 0UL, 0UL}},
+/* pmd5  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(5),0UL, 0UL, 0UL}},
+/* pmd6  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(6),0UL, 0UL, 0UL}},
+/* pmd7  */ { PFM_REG_COUNTING, 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(7),0UL, 0UL, 0UL}},
+	     { PFM_REG_END     , 0, 0x0UL, -1UL, NULL, NULL, {0,}, {0,}}, /* end marker */
+};

arch/ia64/kernel/perfmon_itanium.h

@@ -15,44 +15,44 @@
 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 pfm_pmc_desc[PMU_MAX_PMCS]={
+/* pmc0  */ { PFM_REG_CONTROL , 0, 0x1UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc1  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc2  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc3  */ { PFM_REG_CONTROL , 0, 0x0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc4  */ { PFM_REG_COUNTING, 6, 0x0UL, -1UL, NULL, NULL, {RDEP(4),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc5  */ { PFM_REG_COUNTING, 6, 0x0UL, -1UL, NULL, NULL, {RDEP(5),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc6  */ { PFM_REG_COUNTING, 6, 0x0UL, -1UL, NULL, NULL, {RDEP(6),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc7  */ { PFM_REG_COUNTING, 6, 0x0UL, -1UL, NULL, NULL, {RDEP(7),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc8  */ { PFM_REG_CONFIG  , 0, 0xf00000003ffffff8UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc9  */ { PFM_REG_CONFIG  , 0, 0xf00000003ffffff8UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc10 */ { PFM_REG_MONITOR , 6, 0x0UL, -1UL, NULL, NULL, {RDEP(0)|RDEP(1),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc11 */ { PFM_REG_MONITOR , 6, 0x0000000010000000UL, -1UL, NULL, pfm_ita_pmc_check, {RDEP(2)|RDEP(3)|RDEP(17),0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+/* pmc12 */ { PFM_REG_MONITOR , 6, 0x0UL, -1UL, 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, 0x0003ffff00000001UL, -1UL, NULL, pfm_ita_pmc_check, {0UL,0UL, 0UL, 0UL}, {0UL,0UL, 0UL, 0UL}},
+	    { PFM_REG_END     , 0, 0x0UL, -1UL, 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 pfm_reg_desc_t pfm_pmd_desc[PMU_MAX_PMDS]={
+/* pmd0  */ { PFM_REG_BUFFER  , 0, 0UL, -1UL, NULL, NULL, {RDEP(1),0UL, 0UL, 0UL}, {RDEP(10),0UL, 0UL, 0UL}},
+/* pmd1  */ { PFM_REG_BUFFER  , 0, 0UL, -1UL, NULL, NULL, {RDEP(0),0UL, 0UL, 0UL}, {RDEP(10),0UL, 0UL, 0UL}},
+/* pmd2  */ { PFM_REG_BUFFER  , 0, 0UL, -1UL, NULL, NULL, {RDEP(3)|RDEP(17),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+/* pmd3  */ { PFM_REG_BUFFER  , 0, 0UL, -1UL, NULL, NULL, {RDEP(2)|RDEP(17),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+/* pmd4  */ { PFM_REG_COUNTING, 0, 0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(4),0UL, 0UL, 0UL}},
+/* pmd5  */ { PFM_REG_COUNTING, 0, 0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(5),0UL, 0UL, 0UL}},
+/* pmd6  */ { PFM_REG_COUNTING, 0, 0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(6),0UL, 0UL, 0UL}},
+/* pmd7  */ { PFM_REG_COUNTING, 0, 0UL, -1UL, NULL, NULL, {0UL,0UL, 0UL, 0UL}, {RDEP(7),0UL, 0UL, 0UL}},
+/* pmd8  */ { PFM_REG_BUFFER  , 0, 0UL, -1UL, 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, 0UL, -1UL, 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, 0UL, -1UL, 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, 0UL, -1UL, 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, 0UL, -1UL, 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, 0UL, -1UL, 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, 0UL, -1UL, 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, 0UL, -1UL, 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, 0UL, -1UL, 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, 0UL, -1UL, NULL, NULL, {RDEP(2)|RDEP(3),0UL, 0UL, 0UL}, {RDEP(11),0UL, 0UL, 0UL}},
+	    { PFM_REG_END     , 0, 0UL, -1UL, NULL, NULL, {0,}, {0,}}, /* end marker */
 };
 
 static int

arch/ia64/kernel/process.c

@@ -12,6 +12,7 @@
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <linux/module.h>
 #include <linux/personality.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
@@ -45,8 +46,9 @@ static void
 do_show_stack (struct unw_frame_info *info, void *arg)
 {
 	unsigned long ip, sp, bsp;
+	char buf[80];			/* don't make it so big that it overflows the stack! */
 
-	printk("\nCall Trace: ");
+	printk("\nCall Trace:\n");
 	do {
 		unw_get_ip(info, &ip);
 		if (ip == 0)
@@ -54,7 +56,9 @@ do_show_stack (struct unw_frame_info *info, void *arg)
 
 		unw_get_sp(info, &sp);
 		unw_get_bsp(info, &bsp);
-		printk("[<%016lx>] sp=0x%016lx bsp=0x%016lx\n", ip, sp, bsp);
+		snprintf(buf, sizeof(buf), " [<%016lx>] %%s sp=0x%016lx bsp=0x%016lx\n",
+			 ip, sp, bsp);
+		print_symbol(buf, ip);
 	} while (unw_unwind(info) >= 0);
 }
 
@@ -94,6 +98,7 @@ show_regs (struct pt_regs *regs)
 	printk("\nPid: %d, comm: %20s\n", current->pid, current->comm);
 	printk("psr : %016lx ifs : %016lx ip  : [<%016lx>]    %s\n",
 	       regs->cr_ipsr, regs->cr_ifs, ip, print_tainted());
+	print_symbol("ip is at %s\n", ip);
 	printk("unat: %016lx pfs : %016lx rsc : %016lx\n",
 	       regs->ar_unat, regs->ar_pfs, regs->ar_rsc);
 	printk("rnat: %016lx bsps: %016lx pr  : %016lx\n",
@@ -199,10 +204,8 @@ ia64_save_extra (struct task_struct *task)
 	if ((task->thread.flags & IA64_THREAD_PM_VALID) != 0)
 		pfm_save_regs(task);
 
-# ifdef CONFIG_SMP
 	if (__get_cpu_var(pfm_syst_wide))
 		pfm_syst_wide_update_task(task, 0);
-# endif
 #endif
 
 #ifdef CONFIG_IA32_SUPPORT
@@ -221,9 +224,8 @@ ia64_load_extra (struct task_struct *task)
 	if ((task->thread.flags & IA64_THREAD_PM_VALID) != 0)
 		pfm_load_regs(task);
 
-# ifdef CONFIG_SMP
-	if (__get_cpu_var(pfm_syst_wide)) pfm_syst_wide_update_task(task, 1);
-# endif
+	if (__get_cpu_var(pfm_syst_wide)) 
+		pfm_syst_wide_update_task(task, 1);
 #endif
 
 #ifdef CONFIG_IA32_SUPPORT

arch/ia64/kernel/setup.c

@@ -34,6 +34,7 @@
 
 #include <asm/ia32.h>
 #include <asm/page.h>
+#include <asm/pgtable.h>
 #include <asm/machvec.h>
 #include <asm/processor.h>
 #include <asm/sal.h>
@@ -49,9 +50,6 @@
 # error "struct cpuinfo_ia64 too big!"
 #endif
 
-#define MIN(a,b)	((a) < (b) ? (a) : (b))
-#define MAX(a,b)	((a) > (b) ? (a) : (b))
-
 extern char _end;
 
 #ifdef CONFIG_SMP
@@ -95,6 +93,10 @@ struct rsvd_region {
 static struct rsvd_region rsvd_region[IA64_MAX_RSVD_REGIONS + 1];
 static int num_rsvd_regions;
 
+#define IGNORE_PFN0	1	/* XXX fix me: ignore pfn 0 until TLB miss handler is updated... */
+
+#ifndef CONFIG_DISCONTIGMEM
+
 static unsigned long bootmap_start; /* physical address where the bootmem map is located */
 
 static int
@@ -108,17 +110,63 @@ find_max_pfn (unsigned long start, unsigned long end, void *arg)
 	return 0;
 }
 
-#define IGNORE_PFN0	1	/* XXX fix me: ignore pfn 0 until TLB miss handler is updated... */
+#else /* CONFIG_DISCONTIGMEM */
 
 /*
- * Free available memory based on the primitive map created from
- * the boot parameters. This routine does not assume the incoming
- * segments are sorted.
+ * efi_memmap_walk() knows nothing about layout of memory across nodes. Find
+ * out to which node a block of memory belongs.  Ignore memory that we cannot
+ * identify, and split blocks that run across multiple nodes.
+ *
+ * Take this opportunity to round the start address up and the end address
+ * down to page boundaries.
  */
-static int
-free_available_memory (unsigned long start, unsigned long end, void *arg)
+void
+call_pernode_memory (unsigned long start, unsigned long end, void *arg)
+{
+	unsigned long rs, re;
+	void (*func)(unsigned long, unsigned long, int, int);
+	int i;
+
+	start = PAGE_ALIGN(start);
+	end &= PAGE_MASK;
+	if (start >= end)
+		return;
+
+	func = arg;
+
+	if (!num_memblks) {
+		/*
+		 * This machine doesn't have SRAT, so call func with
+		 * nid=0, bank=0.
+		 */
+		if (start < end)
+			(*func)(start, end - start, 0, 0);
+		return;
+	}
+
+	for (i = 0; i < num_memblks; i++) {
+		rs = max(start, node_memblk[i].start_paddr);
+		re = min(end, node_memblk[i].start_paddr+node_memblk[i].size);
+
+		if (rs < re)
+			(*func)(rs, re-rs, node_memblk[i].nid,
+				node_memblk[i].bank);
+	}
+}
+
+#endif /* CONFIG_DISCONTIGMEM */
+
+/*
+ * Filter incoming memory segments based on the primitive map created from the boot
+ * parameters. Segments contained in the map are removed from the memory ranges. A
+ * caller-specified function is called with the memory ranges that remain after filtering.
+ * This routine does not assume the incoming segments are sorted.
+ */
+int
+filter_rsvd_memory (unsigned long start, unsigned long end, void *arg)
 {
 	unsigned long range_start, range_end, prev_start;
+	void (*func)(unsigned long, unsigned long);
 	int i;
 
 #if IGNORE_PFN0
@@ -132,13 +180,18 @@ free_available_memory (unsigned long start, unsigned long end, void *arg)
 	 * lowest possible address(walker uses virtual)
 	 */
 	prev_start = PAGE_OFFSET;
+	func = arg;
 
 	for (i = 0; i < num_rsvd_regions; ++i) {
-		range_start = MAX(start, prev_start);
-		range_end   = MIN(end, rsvd_region[i].start);
+		range_start = max(start, prev_start);
+		range_end   = min(end, rsvd_region[i].start);
 
 		if (range_start < range_end)
-			free_bootmem(__pa(range_start), range_end - range_start);
+#ifdef CONFIG_DISCONTIGMEM
+			call_pernode_memory(__pa(range_start), __pa(range_end), func);
+#else
+			(*func)(__pa(range_start), range_end - range_start);
+#endif
 
 		/* nothing more available in this segment */
 		if (range_end == end) return 0;
@@ -150,6 +203,7 @@ free_available_memory (unsigned long start, unsigned long end, void *arg)
 }
 
 
+#ifndef CONFIG_DISCONTIGMEM
 /*
  * Find a place to put the bootmap and return its starting address in bootmap_start.
  * This address must be page-aligned.
@@ -171,8 +225,8 @@ find_bootmap_location (unsigned long start, unsigned long end, void *arg)
 	free_start = PAGE_OFFSET;
 
 	for (i = 0; i < num_rsvd_regions; i++) {
-		range_start = MAX(start, free_start);
-		range_end   = MIN(end, rsvd_region[i].start & PAGE_MASK);
+		range_start = max(start, free_start);
+		range_end   = min(end, rsvd_region[i].start & PAGE_MASK);
 
 		if (range_end <= range_start) continue;	/* skip over empty range */
 
@@ -188,6 +242,7 @@ find_bootmap_location (unsigned long start, unsigned long end, void *arg)
 	}
 	return 0;
 }
+#endif /* !CONFIG_DISCONTIGMEM */
 
 static void
 sort_regions (struct rsvd_region *rsvd_region, int max)
@@ -252,6 +307,15 @@ find_memory (void)
 
 	sort_regions(rsvd_region, num_rsvd_regions);
 
+#ifdef CONFIG_DISCONTIGMEM
+	{
+		extern void discontig_mem_init (void);
+
+		bootmap_size = max_pfn = 0;	/* stop gcc warnings */
+		discontig_mem_init();
+	}
+#else /* !CONFIG_DISCONTIGMEM */
+
 	/* first find highest page frame number */
 	max_pfn = 0;
 	efi_memmap_walk(find_max_pfn, &max_pfn);
@@ -268,8 +332,9 @@ find_memory (void)
 	bootmap_size = init_bootmem(bootmap_start >> PAGE_SHIFT, max_pfn);
 
 	/* Free all available memory, then mark bootmem-map as being in use.  */
-	efi_memmap_walk(free_available_memory, 0);
+	efi_memmap_walk(filter_rsvd_memory, free_bootmem);
 	reserve_bootmem(bootmap_start, bootmap_size);
+#endif /* !CONFIG_DISCONTIGMEM */
 
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (ia64_boot_param->initrd_start) {
@@ -296,6 +361,16 @@ setup_arch (char **cmdline_p)
 
 	efi_init();
 
+#ifdef CONFIG_ACPI_BOOT
+	/* Initialize the ACPI boot-time table parser */
+	acpi_table_init(*cmdline_p);
+
+#ifdef CONFIG_ACPI_NUMA
+	acpi_numa_init();
+#endif
+
+#endif /* CONFIG_APCI_BOOT */
+
 	find_memory();
 
 #if 0
@@ -530,6 +605,7 @@ setup_per_cpu_areas (void)
 	/* start_kernel() requires this... */
 }
 
+
 /*
  * cpu_init() initializes state that is per-CPU.  This function acts
  * as a 'CPU state barrier', nothing should get across.
@@ -542,32 +618,40 @@ cpu_init (void)
 	unsigned long num_phys_stacked;
 	pal_vm_info_2_u_t vmi;
 	unsigned int max_ctx;
-	struct cpuinfo_ia64 *my_cpu_info;
-	void *my_cpu_data;
+	struct cpuinfo_ia64 *cpu_info;
+	void *cpu_data;
 
 #ifdef CONFIG_SMP
 	extern char __per_cpu_end[];
-	int cpu = smp_processor_id();
+	int cpu;
 
 	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...
+	 * get_free_pages() cannot be used before cpu_init() done.  BSP allocates
+	 * "NR_CPUS" pages for all CPUs to avoid that AP calls get_zeroed_page().
 	 */
-	if (cpu == 0)
-		my_cpu_data = alloc_bootmem_pages(__per_cpu_end - __per_cpu_start);
-	else
-		my_cpu_data = (void *) get_zeroed_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 *) &__get_cpu_var(cpu_info) - __per_cpu_start);
-#else
-	my_cpu_data = __phys_per_cpu_start;
+	if (smp_processor_id() == 0) {
+		cpu_data = (unsigned long)alloc_bootmem_pages(PAGE_SIZE * NR_CPUS);
+		for (cpu = 0; cpu < NR_CPUS; cpu++) {
+			memcpy(cpu_data, __phys_per_cpu_start, __per_cpu_end - __per_cpu_start);
+			__per_cpu_offset[cpu] = (char *) cpu_data - __per_cpu_start;
+			cpu_data += PAGE_SIZE;
+		}
+	}
+	cpu_data = __per_cpu_start + __per_cpu_offset[smp_processor_id()];
+#else /* !CONFIG_SMP */
+	cpu_data = __phys_per_cpu_start;
+#endif /* !CONFIG_SMP */
+
+	cpu_info = cpu_data + ((char *) &__get_cpu_var(cpu_info) - __per_cpu_start);
+#ifdef CONFIG_NUMA
+	cpu_info->node_data = get_node_data_ptr();
+	cpu_info->nodeid = boot_get_local_nodeid();
 #endif
-	my_cpu_info = my_cpu_data + ((char *) &__get_cpu_var(cpu_info) - __per_cpu_start);
 
 	/*
 	 * We can't pass "local_cpu_data" to identify_cpu() because we haven't called
@@ -575,14 +659,14 @@ cpu_init (void)
 	 * depends on the data returned by identify_cpu().  We break the dependency by
 	 * accessing cpu_data() the old way, through identity mapped space.
 	 */
-	identify_cpu(my_cpu_info);
+	identify_cpu(cpu_info);
 
 #ifdef CONFIG_MCKINLEY
 	{
-#define FEATURE_SET 16
+#		define FEATURE_SET 16
 		struct ia64_pal_retval iprv;
 
-		if (my_cpu_info->family == 0x1f) {
+		if (cpu_info->family == 0x1f) {
 			PAL_CALL_PHYS(iprv, PAL_PROC_GET_FEATURES, 0, FEATURE_SET, 0);
 			if ((iprv.status == 0) && (iprv.v0 & 0x80) && (iprv.v2 & 0x80))
 				PAL_CALL_PHYS(iprv, PAL_PROC_SET_FEATURES,
@@ -613,7 +697,7 @@ cpu_init (void)
 	if (current->mm)
 		BUG();
 
-	ia64_mmu_init(my_cpu_data);
+	ia64_mmu_init(cpu_data);
 
 #ifdef CONFIG_IA32_SUPPORT
 	/* initialize global ia32 state - CR0 and CR4 */

arch/ia64/kernel/signal.c

@@ -41,6 +41,16 @@
 # define GET_SIGSET(k,u)	__get_user((k)->sig[0], &(u)->sig[0])
 #endif
 
+register double f16 asm ("f16"); register double f17 asm ("f17");
+register double f18 asm ("f18"); register double f19 asm ("f19");
+register double f20 asm ("f20"); register double f21 asm ("f21");
+register double f22 asm ("f22"); register double f23 asm ("f23");
+
+register double f24 asm ("f24"); register double f25 asm ("f25");
+register double f26 asm ("f26"); register double f27 asm ("f27");
+register double f28 asm ("f28"); register double f29 asm ("f29");
+register double f30 asm ("f30"); register double f31 asm ("f31");
+
 long
 ia64_rt_sigsuspend (sigset_t *uset, size_t sigsetsize, struct sigscratch *scr)
 {
@@ -58,13 +68,13 @@ ia64_rt_sigsuspend (sigset_t *uset, size_t sigsetsize, struct sigscratch *scr)
 
 	sigdelsetmask(&set, ~_BLOCKABLE);
 
-	spin_lock_irq(&current->sigmask_lock);
+	spin_lock_irq(&current->sig->siglock);
 	{
 		oldset = current->blocked;
 		current->blocked = set;
 		recalc_sigpending();
 	}
-	spin_unlock_irq(&current->sigmask_lock);
+	spin_unlock_irq(&current->sig->siglock);
 
 	/*
 	 * The return below usually returns to the signal handler.  We need to
@@ -264,12 +274,12 @@ ia64_rt_sigreturn (struct sigscratch *scr)
 
 	sigdelsetmask(&set, ~_BLOCKABLE);
 
-	spin_lock_irq(&current->sigmask_lock);
+	spin_lock_irq(&current->sig->siglock);
 	{
 		current->blocked = set;
 		recalc_sigpending();
 	}
-	spin_unlock_irq(&current->sigmask_lock);
+	spin_unlock_irq(&current->sig->siglock);
 
 	if (restore_sigcontext(sc, scr))
 		goto give_sigsegv;
@@ -458,13 +468,13 @@ handle_signal (unsigned long sig, struct k_sigaction *ka, siginfo_t *info, sigse
 		ka->sa.sa_handler = SIG_DFL;
 
 	if (!(ka->sa.sa_flags & SA_NODEFER)) {
-		spin_lock_irq(&current->sigmask_lock);
+		spin_lock_irq(&current->sig->siglock);
 		{
 			sigorsets(&current->blocked, &current->blocked, &ka->sa.sa_mask);
 			sigaddset(&current->blocked, sig);
 			recalc_sigpending();
 		}
-		spin_unlock_irq(&current->sigmask_lock);
+		spin_unlock_irq(&current->sig->siglock);
 	}
 	return 1;
 }

arch/ia64/kernel/smpboot.c

@@ -16,6 +16,7 @@
 
 #include <linux/config.h>
 
+#include <linux/acpi.h>
 #include <linux/bootmem.h>
 #include <linux/delay.h>
 #include <linux/init.h>
@@ -427,6 +428,37 @@ smp_build_cpu_map (void)
 	}
 }
 
+#ifdef CONFIG_NUMA
+
+char cpu_to_node_map[NR_CPUS] __cacheline_aligned;
+
+/*
+ * Build cpu to node mapping.
+ */
+void __init
+build_cpu_to_node_map (void)
+{
+	int cpu, i;
+
+	for(cpu = 0; cpu < NR_CPUS; ++cpu) {
+		/*
+		 * All Itanium NUMA platforms I know use ACPI, so maybe we
+		 * can drop this ifdef completely.                    [EF]
+		 */
+#ifdef CONFIG_ACPI_NUMA
+		for (i = 0; i < NR_CPUS; ++i)
+			if (cpu_physical_id(cpu) == node_cpuid[i].phys_id) {
+				cpu_to_node_map[cpu] = node_cpuid[i].nid;
+				break;
+			}
+#else
+#		error Fixme: Dunno how to build CPU-to-node map.
+#endif
+	}
+}
+
+#endif /* CONFIG_NUMA */
+
 /*
  * Cycle through the APs sending Wakeup IPIs to boot each.
  */

arch/ia64/kernel/sys_ia64.c

@@ -282,18 +282,20 @@ sys_free_hugepages (unsigned long  addr)
 	extern int free_hugepages(struct vm_area_struct *);
 	int retval;
 
-	vma = find_vma(mm, addr);
-	if (!vma || !is_vm_hugetlb_page(vma) || (vma->vm_start != addr))
-		return -EINVAL;
-
 	down_write(&mm->mmap_sem);
 	{
+		vma = find_vma(mm, addr);
+		if (!vma || !is_vm_hugetlb_page(vma) || (vma->vm_start != addr))
+			retval = -EINVAL;
+			goto out;
+
 		spin_lock(&mm->page_table_lock);
 		{
 			retval = free_hugepages(vma);
 		}
 		spin_unlock(&mm->page_table_lock);
 	}
+out:
 	up_write(&mm->mmap_sem);
 	return retval;
 }

arch/ia64/kernel/time.c

 /*
  * linux/arch/ia64/kernel/time.c
  *
- * Copyright (C) 1998-2001 Hewlett-Packard Co
- * Copyright (C) 1998-2000 Stephane Eranian <eranian@hpl.hp.com>
- * Copyright (C) 1999-2001 David Mosberger <davidm@hpl.hp.com>
+ * Copyright (C) 1998-2002 Hewlett-Packard Co
+ *	Stephane Eranian <eranian@hpl.hp.com>
+ *	David Mosberger <davidm@hpl.hp.com>
  * Copyright (C) 1999 Don Dugger <don.dugger@intel.com>
  * Copyright (C) 1999-2000 VA Linux Systems
  * Copyright (C) 1999-2000 Walt Drummond <drummond@valinux.com>
@@ -12,6 +12,7 @@
 
 #include <linux/init.h>
 #include <linux/kernel.h>
+#include <linux/profile.h>
 #include <linux/sched.h>
 #include <linux/time.h>
 #include <linux/interrupt.h>

arch/ia64/kernel/traps.c

@@ -54,8 +54,8 @@ trap_init (void)
 	if (ia64_boot_param->fpswa) {
 		/* FPSWA fixup: make the interface pointer a kernel virtual address: */
 		fpswa_interface = __va(ia64_boot_param->fpswa);
-		major = fpswa_interface->revision & 0xffff;
-		minor = fpswa_interface->revision >> 16;
+		major = fpswa_interface->revision >> 16;
+		minor = fpswa_interface->revision & 0xffff;
 	}
 	printk("fpswa interface at %lx (rev %d.%d)\n", ia64_boot_param->fpswa, major, minor);
 }
@@ -69,7 +69,6 @@ bust_spinlocks (int yes)
 {
 	int loglevel_save = console_loglevel;
 
-	spin_lock_init(&timerlist_lock);
 	if (yes) {
 		oops_in_progress = 1;
 		return;

arch/ia64/lib/Makefile

@@ -32,26 +32,26 @@ AFLAGS___udivsi3.o	= -DUNSIGNED
 AFLAGS___modsi3.o	=	     -DMODULO
 AFLAGS___umodsi3.o	= -DUNSIGNED -DMODULO
 
-__divdi3.o: idiv64.S
+$(obj)/__divdi3.o: $(src)/idiv64.S
 	$(cmd_as_o_S)
 
-__udivdi3.o: idiv64.S
+$(obj)/__udivdi3.o: $(src)/idiv64.S
 	$(cmd_as_o_S)
 
-__moddi3.o: idiv64.S
+$(obj)/__moddi3.o: $(src)/idiv64.S
 	$(cmd_as_o_S)
 
-__umoddi3.o: idiv64.S
+$(obj)/__umoddi3.o: $(src)/idiv64.S
 	$(cmd_as_o_S)
 
-__divsi3.o: idiv32.S
+$(obj)/__divsi3.o: $(src)/idiv32.S
 	$(cmd_as_o_S)
 
-__udivsi3.o: idiv32.S
+$(obj)/__udivsi3.o: $(src)/idiv32.S
 	$(cmd_as_o_S)
 
-__modsi3.o: idiv32.S
+$(obj)/__modsi3.o: $(src)/idiv32.S
 	$(cmd_as_o_S)
 
-__umodsi3.o: idiv32.S
+$(obj)/__umodsi3.o: $(src)/idiv32.S
 	$(cmd_as_o_S)

arch/ia64/mm/Makefile

@@ -9,5 +9,7 @@
 
 obj-y	 := init.o fault.o tlb.o extable.o
 obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
+obj-$(CONFIG_NUMA) += numa.o
+obj-$(CONFIG_DISCONTIGMEM) += discontig.o
 
 include $(TOPDIR)/Rules.make

arch/ia64/mm/discontig.c

+/*
+ * Copyright (c) 2000 Silicon Graphics, Inc.  All rights reserved.
+ * Copyright (c) 2001 Intel Corp.
+ * Copyright (c) 2001 Tony Luck <tony.luck@intel.com>
+ * Copyright (c) 2002 NEC Corp.
+ * Copyright (c) 2002 Kimio Suganuma <k-suganuma@da.jp.nec.com>
+ */
+
+/*
+ * Platform initialization for Discontig Memory
+ */
+
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/bootmem.h>
+#include <linux/mmzone.h>
+#include <linux/acpi.h>
+#include <linux/efi.h>
+
+
+/*
+ * Round an address upward to the next multiple of GRANULE size.
+ */
+#define GRANULEROUNDUP(n) (((n)+IA64_GRANULE_SIZE-1) & ~(IA64_GRANULE_SIZE-1))
+
+static struct ia64_node_data	*node_data[NR_NODES];
+static long			boot_pg_data[8*NR_NODES+sizeof(pg_data_t)]  __initdata;
+static pg_data_t		*pg_data_ptr[NR_NODES] __initdata;
+static bootmem_data_t		bdata[NR_NODES][NR_BANKS_PER_NODE+1] __initdata;
+
+extern int  filter_rsvd_memory (unsigned long start, unsigned long end, void *arg);
+
+/*
+ * Return the compact node number of this cpu. Used prior to
+ * setting up the cpu_data area.
+ *	Note - not fast, intended for boot use only!!
+ */
+int
+boot_get_local_nodeid(void)
+{
+	int	i;
+
+	for (i = 0; i < NR_CPUS; i++)
+		if (node_cpuid[i].phys_id == hard_smp_processor_id())
+			return node_cpuid[i].nid;
+
+	/* node info missing, so nid should be 0.. */
+	return 0;
+}
+
+/*
+ * Return a pointer to the pg_data structure for a node.
+ * This function is used ONLY in early boot before the cpu_data
+ * structure is available.
+ */
+pg_data_t* __init
+boot_get_pg_data_ptr(long node)
+{
+	return pg_data_ptr[node];
+}
+
+
+/*
+ * Return a pointer to the node data for the current node.
+ *	(boottime initialization only)
+ */
+struct ia64_node_data *
+get_node_data_ptr(void)
+{
+	return node_data[boot_get_local_nodeid()];
+}
+
+/*
+ * We allocate one of the bootmem_data_t structs for each piece of memory
+ * that we wish to treat as a contiguous block.  Each such block must start
+ * on a BANKSIZE boundary.  Multiple banks per node is not supported.
+ */
+static int __init
+build_maps(unsigned long pstart, unsigned long length, int node)
+{
+	bootmem_data_t	*bdp;
+	unsigned long cstart, epfn;
+
+	bdp = pg_data_ptr[node]->bdata;
+	epfn = GRANULEROUNDUP(pstart + length) >> PAGE_SHIFT;
+	cstart = pstart & ~(BANKSIZE - 1);
+
+	if (!bdp->node_low_pfn) {
+		bdp->node_boot_start = cstart;
+		bdp->node_low_pfn = epfn;
+	} else {
+		bdp->node_boot_start = min(cstart, bdp->node_boot_start);
+		bdp->node_low_pfn = max(epfn, bdp->node_low_pfn);
+	}
+
+	min_low_pfn = min(min_low_pfn, bdp->node_boot_start>>PAGE_SHIFT);
+	max_low_pfn = max(max_low_pfn, bdp->node_low_pfn);
+
+	return 0;
+}
+
+/*
+ * Find space on each node for the bootmem map.
+ *
+ * Called by efi_memmap_walk to find boot memory on each node. Note that
+ * only blocks that are free are passed to this routine (currently filtered by
+ * free_available_memory).
+ */
+static int __init
+find_bootmap_space(unsigned long pstart, unsigned long length, int node)
+{
+	unsigned long	mapsize, pages, epfn;
+	bootmem_data_t	*bdp;
+
+	epfn = (pstart + length) >> PAGE_SHIFT;
+	bdp = &pg_data_ptr[node]->bdata[0];
+
+	if (pstart < bdp->node_boot_start || epfn > bdp->node_low_pfn)
+		return 0;
+
+	if (!bdp->node_bootmem_map) {
+		pages = bdp->node_low_pfn - (bdp->node_boot_start>>PAGE_SHIFT);
+		mapsize = bootmem_bootmap_pages(pages) << PAGE_SHIFT;
+		if (length > mapsize) {
+			init_bootmem_node(
+				BOOT_NODE_DATA(node),
+				pstart>>PAGE_SHIFT, 
+				bdp->node_boot_start>>PAGE_SHIFT,
+				bdp->node_low_pfn);
+		}
+
+	}
+
+	return 0;
+}
+
+
+/*
+ * Free available memory to the bootmem allocator.
+ *
+ * Note that only blocks that are free are passed to this routine (currently 
+ * filtered by free_available_memory).
+ *
+ */
+static int __init
+discontig_free_bootmem_node(unsigned long pstart, unsigned long length, int node)
+{
+	free_bootmem_node(BOOT_NODE_DATA(node), pstart, length);
+
+	return 0;
+}
+
+
+/*
+ * Reserve the space used by the bootmem maps.
+ */
+static void __init
+discontig_reserve_bootmem(void)
+{
+	int		node;
+	unsigned long	mapbase, mapsize, pages;
+	bootmem_data_t	*bdp;
+
+	for (node = 0; node < numnodes; node++) {
+		bdp = BOOT_NODE_DATA(node)->bdata;
+
+		pages = bdp->node_low_pfn - (bdp->node_boot_start>>PAGE_SHIFT);
+		mapsize = bootmem_bootmap_pages(pages) << PAGE_SHIFT;
+		mapbase = __pa(bdp->node_bootmem_map);
+		reserve_bootmem_node(BOOT_NODE_DATA(node), mapbase, mapsize);
+	}
+}
+
+/*
+ * Allocate per node tables.
+ * 	- the pg_data structure is allocated on each node. This minimizes offnode 
+ *	  memory references
+ *	- the node data is allocated & initialized. Portions of this structure is read-only (after 
+ *	  boot) and contains node-local pointers to usefuls data structures located on
+ *	  other nodes.
+ *
+ * We also switch to using the "real" pg_data structures at this point. Earlier in boot, we
+ * use a different structure. The only use for pg_data prior to the point in boot is to get 
+ * the pointer to the bdata for the node.
+ */
+static void __init
+allocate_pernode_structures(void)
+{
+	pg_data_t	*pgdat=0, *new_pgdat_list=0;
+	int		node, mynode;
+
+	mynode = boot_get_local_nodeid();
+	for (node = numnodes - 1; node >= 0 ; node--) {
+		node_data[node] = alloc_bootmem_node(BOOT_NODE_DATA(node), sizeof (struct ia64_node_data));
+		pgdat = __alloc_bootmem_node(BOOT_NODE_DATA(node), sizeof(pg_data_t), SMP_CACHE_BYTES, 0);
+		pgdat->bdata = &(bdata[node][0]);
+		pg_data_ptr[node] = pgdat;
+		pgdat->pgdat_next = new_pgdat_list;
+		new_pgdat_list = pgdat;
+	}
+	
+	memcpy(node_data[mynode]->pg_data_ptrs, pg_data_ptr, sizeof(pg_data_ptr));
+	memcpy(node_data[mynode]->node_data_ptrs, node_data, sizeof(node_data));
+
+	pgdat_list = new_pgdat_list;
+}
+
+/*
+ * Called early in boot to setup the boot memory allocator, and to
+ * allocate the node-local pg_data & node-directory data structures..
+ */
+void __init
+discontig_mem_init(void)
+{
+	int	node;
+
+	if (numnodes == 0) {
+		printk("node info missing!\n");
+		numnodes = 1;
+	}
+
+	for (node = 0; node < numnodes; node++) {
+		pg_data_ptr[node] = (pg_data_t*) &boot_pg_data[node];
+		pg_data_ptr[node]->bdata = &bdata[node][0];
+	}
+
+	min_low_pfn = -1;
+	max_low_pfn = 0;
+
+        efi_memmap_walk(filter_rsvd_memory, build_maps);
+        efi_memmap_walk(filter_rsvd_memory, find_bootmap_space);
+        efi_memmap_walk(filter_rsvd_memory, discontig_free_bootmem_node);
+	discontig_reserve_bootmem();
+	allocate_pernode_structures();
+}
+
+/*
+ * Initialize the paging system.
+ *	- determine sizes of each node
+ *	- initialize the paging system for the node
+ *	- build the nodedir for the node. This contains pointers to
+ *	  the per-bank mem_map entries.
+ *	- fix the page struct "virtual" pointers. These are bank specific
+ *	  values that the paging system doesnt understand.
+ *	- replicate the nodedir structure to other nodes	
+ */ 
+
+void __init
+discontig_paging_init(void)
+{
+	int		node, mynode;
+	unsigned long	max_dma, zones_size[MAX_NR_ZONES];
+	unsigned long	kaddr, ekaddr, bid;
+	struct page	*page;
+	bootmem_data_t	*bdp;
+
+	max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;
+
+	mynode = boot_get_local_nodeid();
+	for (node = 0; node < numnodes; node++) {
+		long pfn, startpfn;
+
+		memset(zones_size, 0, sizeof(zones_size));
+
+		startpfn = -1;
+		bdp = BOOT_NODE_DATA(node)->bdata;
+		pfn = bdp->node_boot_start >> PAGE_SHIFT;
+		if (startpfn == -1)
+			startpfn = pfn;
+		if (pfn > max_dma)
+			zones_size[ZONE_NORMAL] += (bdp->node_low_pfn - pfn);
+		else if (bdp->node_low_pfn < max_dma)
+			zones_size[ZONE_DMA] += (bdp->node_low_pfn - pfn);
+		else {
+			zones_size[ZONE_DMA] += (max_dma - pfn);
+			zones_size[ZONE_NORMAL] += (bdp->node_low_pfn - max_dma);
+		}
+
+		free_area_init_node(node, NODE_DATA(node), NULL, zones_size, startpfn, 0);
+
+		page = NODE_DATA(node)->node_mem_map;
+
+		bdp = BOOT_NODE_DATA(node)->bdata;
+
+		kaddr = (unsigned long)__va(bdp->node_boot_start);
+		ekaddr = (unsigned long)__va(bdp->node_low_pfn << PAGE_SHIFT);
+		while (kaddr < ekaddr) {
+			bid = BANK_MEM_MAP_INDEX(kaddr);
+			node_data[mynode]->node_id_map[bid] = node;
+			node_data[mynode]->bank_mem_map_base[bid] = page;
+			kaddr += BANKSIZE;
+			page += BANKSIZE/PAGE_SIZE;
+		}
+	}
+
+	/*
+	 * Finish setting up the node data for this node, then copy it to the other nodes.
+	 */
+	for (node=0; node < numnodes; node++)
+		if (mynode != node) {
+			memcpy(node_data[node], node_data[mynode], sizeof(struct ia64_node_data));
+			node_data[node]->node = node;
+		}
+}
+

arch/ia64/mm/init.c

@@ -15,6 +15,7 @@
 #include <linux/slab.h>
 #include <linux/swap.h>
 #include <linux/efi.h>
+#include <linux/mmzone.h>
 
 #include <asm/a.out.h>
 #include <asm/bitops.h>
@@ -78,7 +79,7 @@ ia64_init_addr_space (void)
 		vma->vm_mm = current->mm;
 		vma->vm_start = IA64_RBS_BOT;
 		vma->vm_end = vma->vm_start + PAGE_SIZE;
-		vma->vm_page_prot = protection_map[VM_READ | VM_WRITE];
+		vma->vm_page_prot = protection_map[VM_DATA_DEFAULT_FLAGS & 0x7];
 		vma->vm_flags = VM_READ|VM_WRITE|VM_MAYREAD|VM_MAYWRITE|VM_GROWSUP;
 		vma->vm_ops = NULL;
 		vma->vm_pgoff = 0;
@@ -347,6 +348,15 @@ extern long htlbzone_pages;
 extern struct list_head htlbpage_freelist;
 #endif
 
+#ifdef CONFIG_DISCONTIGMEM
+void
+paging_init (void)
+{
+	extern void discontig_paging_init(void);
+
+	discontig_paging_init();
+}
+#else /* !CONFIG_DISCONTIGMEM */
 void
 paging_init (void)
 {
@@ -365,6 +375,7 @@ paging_init (void)
 	}
 	free_area_init(zones_size);
 }
+#endif /* !CONFIG_DISCONTIGMEM */
 
 static int
 count_pages (u64 start, u64 end, void *arg)
@@ -380,10 +391,9 @@ count_reserved_pages (u64 start, u64 end, void *arg)
 {
 	unsigned long num_reserved = 0;
 	unsigned long *count = arg;
-	struct page *pg;
 
-	for (pg = virt_to_page(start); pg < virt_to_page(end); ++pg)
-		if (PageReserved(pg))
+	for (; start < end; start += PAGE_SIZE)
+		if (PageReserved(virt_to_page(start)))
 			++num_reserved;
 	*count += num_reserved;
 	return 0;
@@ -395,6 +405,7 @@ mem_init (void)
 	extern char __start_gate_section[];
 	long reserved_pages, codesize, datasize, initsize;
 	unsigned long num_pgt_pages;
+	pg_data_t *pgdat;
 
 #ifdef CONFIG_PCI
 	/*
@@ -405,16 +416,19 @@ mem_init (void)
 	platform_pci_dma_init();
 #endif
 
+#ifndef CONFIG_DISCONTIGMEM
 	if (!mem_map)
 		BUG();
+	max_mapnr = max_low_pfn;
+#endif
 
 	num_physpages = 0;
 	efi_memmap_walk(count_pages, &num_physpages);
 
-	max_mapnr = max_low_pfn;
 	high_memory = __va(max_low_pfn * PAGE_SIZE);
 
-	totalram_pages += free_all_bootmem();
+	for_each_pgdat(pgdat)
+		totalram_pages += free_all_bootmem_node(pgdat);
 
 	reserved_pages = 0;
 	efi_memmap_walk(count_reserved_pages, &reserved_pages);
@@ -425,7 +439,7 @@ mem_init (void)
 
 	printk("Memory: %luk/%luk available (%luk code, %luk reserved, %luk data, %luk init)\n",
 	       (unsigned long) nr_free_pages() << (PAGE_SHIFT - 10),
-	       max_mapnr << (PAGE_SHIFT - 10), codesize >> 10, reserved_pages << (PAGE_SHIFT - 10),
+	       num_physpages << (PAGE_SHIFT - 10), codesize >> 10, reserved_pages << (PAGE_SHIFT - 10),
 	       datasize >> 10, initsize >> 10);
 
 	/*
@@ -441,6 +455,8 @@ mem_init (void)
 	if (num_pgt_pages > pgt_cache_water[1])
 		pgt_cache_water[1] = num_pgt_pages;
 
+	show_mem();
+
 	/* install the gate page in the global page table: */
 	put_gate_page(virt_to_page(__start_gate_section), GATE_ADDR);
 

arch/ia64/mm/numa.c

+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * This file contains NUMA specific variables and functions which can
+ * be split away from DISCONTIGMEM and are used on NUMA machines with
+ * contiguous memory.
+ * 
+ *                         2002/08/07 Erich Focht <efocht@ess.nec.de>
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/mmzone.h>
+#include <asm/numa.h>
+
+/*
+ * The following structures are usually initialized by ACPI or
+ * similar mechanisms and describe the NUMA characteristics of the machine.
+ */
+int num_memblks = 0;
+struct node_memblk_s node_memblk[NR_MEMBLKS];
+struct node_cpuid_s node_cpuid[NR_CPUS];
+/*
+ * This is a matrix with "distances" between nodes, they should be
+ * proportional to the memory access latency ratios.
+ */
+u8 numa_slit[NR_NODES * NR_NODES];
+
+/* Identify which cnode a physical address resides on */
+int
+paddr_to_nid(unsigned long paddr)
+{
+	int	i;
+
+	for (i = 0; i < num_memblks; i++)
+		if (paddr >= node_memblk[i].start_paddr &&
+		    paddr < node_memblk[i].start_paddr + node_memblk[i].size)
+			break;
+
+	return (i < num_memblks) ? node_memblk[i].nid : -1;
+}

arch/ia64/pci/pci.c

@@ -141,13 +141,13 @@ pcibios_scan_root (int bus)
 /*
  *  Called after each bus is probed, but before its children are examined.
  */
-void __init
+void __devinit
 pcibios_fixup_bus (struct pci_bus *b)
 {
 	return;
 }
 
-void __init
+void __devinit
 pcibios_update_resource (struct pci_dev *dev, struct resource *root,
 			 struct resource *res, int resource)
 {
@@ -163,7 +163,7 @@ pcibios_update_resource (struct pci_dev *dev, struct resource *root,
 	/* ??? FIXME -- record old value for shutdown.  */
 }
 
-void __init
+void __devinit
 pcibios_update_irq (struct pci_dev *dev, int irq)
 {
 	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
@@ -171,7 +171,7 @@ pcibios_update_irq (struct pci_dev *dev, int irq)
 	/* ??? FIXME -- record old value for shutdown.  */
 }
 
-void __init
+void __devinit
 pcibios_fixup_pbus_ranges (struct pci_bus * bus, struct pbus_set_ranges_data * ranges)
 {
 }

arch/ia64/tools/Makefile

@@ -2,6 +2,8 @@ CFLAGS	= -g -O2 -Wall $(CPPFLAGS)
 
 TARGET	= $(TOPDIR)/include/asm-ia64/offsets.h
 
+src = $(obj)
+
 all:
 
 fastdep:
@@ -9,13 +11,13 @@ fastdep:
 mrproper: clean
 
 clean:
-	rm -f print_offsets.s print_offsets offsets.h
+	rm -f $(obj)/print_offsets.s $(obj)/print_offsets $(obj)/offsets.h
 
-$(TARGET): offsets.h
-	@if ! cmp -s offsets.h ${TARGET}; then	\
+$(TARGET): $(obj)/offsets.h
+	@if ! cmp -s $(obj)/offsets.h ${TARGET}; then	\
 		echo -e "*** Updating ${TARGET}...";	\
-		cp offsets.h ${TARGET};		\
-	else					\
+		cp $(obj)/offsets.h ${TARGET};		\
+	else						\
 		echo "*** ${TARGET} is up to date";	\
 	fi
 
@@ -30,25 +32,26 @@ $(TARGET): offsets.h
 
 ifeq ($(CROSS_COMPILE),)
 
-offsets.h: print_offsets
-	./print_offsets > offsets.h
+$(obj)/offsets.h: $(obj)/print_offsets
+	$(obj)/print_offsets > $(obj)/offsets.h
 
 comma	:= ,
 
-print_offsets: print_offsets.c FORCE
+$(obj)/print_offsets: $(src)/print_offsets.c FORCE
+	[ -r $(TARGET) ] || echo "#define IA64_TASK_SIZE 0" > $(TARGET)
 	$(CC) $(CFLAGS) -DKBUILD_BASENAME=$(subst $(comma),_,$(subst -,_,$(*F))) \
-		print_offsets.c -o $@
+		$(src)/print_offsets.c -o $@
 
 FORCE:
 
 else
 
-offsets.h: print_offsets.s
-	$(AWK) -f print_offsets.awk $^ > $@
+$(obj)/offsets.h: $(obj)/print_offsets.s
+	$(AWK) -f $(src)/print_offsets.awk $^ > $@
 
-print_offsets.s: print_offsets.c
-	$(CC) $(CFLAGS) -DKBUILD_BASENAME=$(subst $(comma),_,$(subst -,_,$(*F))) -S \
-		print_offsets.c -o $@
+$(obj)/print_offsets.s: $(src)/print_offsets.c
+	[ -r $(TARGET) ] || echo "#define IA64_TASK_SIZE 0" > $(TARGET)
+	$(CC) $(CFLAGS) -DKBUILD_BASENAME=$(subst $(comma),_,$(subst -,_,$(*F))) -S $^ -o $@
 
 endif
 

arch/ia64/vmlinux.lds.S

@@ -12,11 +12,11 @@ SECTIONS
 {
   /* Sections to be discarded */
   /DISCARD/ : {
-	*(.text.exit)
-	*(.data.exit)
+	*(.exit.text)
+	*(.exit.data)
 	*(.exitcall.exit)
-	*(.IA_64.unwind.text.exit)
-	*(.IA_64.unwind_info.text.exit)
+	*(.IA_64.unwind.exit.text)
+	*(.IA_64.unwind_info.exit.text)
 	}
 
   v = PAGE_OFFSET;	/* this symbol is here to make debugging easier... */
@@ -65,6 +65,13 @@ SECTIONS
 	{ *(__ksymtab) }
   __stop___ksymtab = .;
 
+  __kallsyms : AT(ADDR(__kallsyms) - PAGE_OFFSET)
+	{
+	  __start___kallsyms = .;	/* All kernel symbols */
+	  *(__kallsyms)
+	  __stop___kallsyms = .;
+	}
+
   /* Unwind info & table: */
   . = ALIGN(8);
   .IA_64.unwind_info : AT(ADDR(.IA_64.unwind_info) - PAGE_OFFSET)
@@ -85,15 +92,15 @@ SECTIONS
 
   . = ALIGN(PAGE_SIZE);
   __init_begin = .;
-  .text.init : AT(ADDR(.text.init) - PAGE_OFFSET)
-	{ *(.text.init) }
+  .init.text : AT(ADDR(.init.text) - PAGE_OFFSET)
+	{ *(.init.text) }
 
-  .data.init : AT(ADDR(.data.init) - PAGE_OFFSET)
-	{ *(.data.init) }
+  .init.data : AT(ADDR(.init.data) - PAGE_OFFSET)
+	{ *(.init.data) }
    . = ALIGN(16);
   __setup_start = .;
-  .setup.init : AT(ADDR(.setup.init) - PAGE_OFFSET)
-        { *(.setup.init) }
+  .init.setup : AT(ADDR(.init.setup) - PAGE_OFFSET)
+        { *(.init.setup) }
   __setup_end = .;
   __initcall_start = .;
   .initcall.init : AT(ADDR(.initcall.init) - PAGE_OFFSET)

drivers/block/deadline-iosched.c

@@ -118,6 +118,8 @@ deadline_find_hash(struct deadline_data *dd, sector_t offset)
 
 	while ((entry = next) != hash_list) {
 		next = entry->next;
+		prefetch(next);
+		
 		drq = list_entry_hash(entry);
 
 		BUG_ON(!drq->hash_valid_count);
@@ -191,6 +193,8 @@ deadline_merge(request_queue_t *q, struct list_head **insert, struct bio *bio)
 	while ((entry = entry->prev) != sort_list) {
 		__rq = list_entry_rq(entry);
 
+		prefetch(entry->prev);
+
 		BUG_ON(__rq->flags & REQ_STARTED);
 
 		if (!(__rq->flags & REQ_CMD))
@@ -298,6 +302,8 @@ static void deadline_move_requests(struct deadline_data *dd, struct request *rq)
 		struct list_head *nxt = rq->queuelist.next;
 		int this_rq_cost;
 
+		prefetch(nxt);
+
 		/*
 		 * take it off the sort and fifo list, move
 		 * to dispatch queue

drivers/block/elevator.c

@@ -272,13 +272,27 @@ void elv_merge_requests(request_queue_t *q, struct request *rq,
 		e->elevator_merge_req_fn(q, rq, next);
 }
 
-/*
- * add_request and next_request are required to be supported, naturally
- */
-void __elv_add_request(request_queue_t *q, struct request *rq,
-			  struct list_head *insert_here)
+void __elv_add_request(request_queue_t *q, struct request *rq, int at_end,
+		       int plug)
+{
+	struct list_head *insert = &q->queue_head;
+
+	if (at_end)
+		insert = insert->prev;
+	if (plug)
+		blk_plug_device(q);
+
+	q->elevator.elevator_add_req_fn(q, rq, insert);
+}
+
+void elv_add_request(request_queue_t *q, struct request *rq, int at_end,
+		     int plug)
 {
-	q->elevator.elevator_add_req_fn(q, rq, insert_here);
+	unsigned long flags;
+
+	spin_lock_irqsave(q->queue_lock, flags);
+	__elv_add_request(q, rq, at_end, plug);
+	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 
 static inline struct request *__elv_next_request(request_queue_t *q)
@@ -289,8 +303,14 @@ static inline struct request *__elv_next_request(request_queue_t *q)
 struct request *elv_next_request(request_queue_t *q)
 {
 	struct request *rq;
+	int ret;
 
 	while ((rq = __elv_next_request(q))) {
+		/*
+		 * just mark as started even if we don't start it, a request
+		 * that has been delayed should not be passed by new incoming
+		 * requests
+		 */
 		rq->flags |= REQ_STARTED;
 
 		if (&rq->queuelist == q->last_merge)
@@ -299,20 +319,22 @@ struct request *elv_next_request(request_queue_t *q)
 		if ((rq->flags & REQ_DONTPREP) || !q->prep_rq_fn)
 			break;
 
-		/*
-		 * all ok, break and return it
-		 */
-		if (!q->prep_rq_fn(q, rq))
+		ret = q->prep_rq_fn(q, rq);
+		if (ret == BLKPREP_OK) {
 			break;
-
-		/*
-		 * prep said no-go, kill it
-		 */
-		blkdev_dequeue_request(rq);
-		if (end_that_request_first(rq, 0, rq->nr_sectors))
-			BUG();
-
-		end_that_request_last(rq);
+		} else if (ret == BLKPREP_DEFER) {
+			rq = NULL;
+			break;
+		} else if (ret == BLKPREP_KILL) {
+			blkdev_dequeue_request(rq);
+			rq->flags |= REQ_QUIET;
+			while (end_that_request_first(rq, 0, rq->nr_sectors))
+				;
+			end_that_request_last(rq);
+		} else {
+			printk("%s: bad return=%d\n", __FUNCTION__, ret);
+			break;
+		}
 	}
 
 	return rq;
@@ -322,6 +344,16 @@ void elv_remove_request(request_queue_t *q, struct request *rq)
 {
 	elevator_t *e = &q->elevator;
 
+	/*
+	 * the main clearing point for q->last_merge is on retrieval of
+	 * request by driver (it calls elv_next_request()), but it _can_
+	 * also happen here if a request is added to the queue but later
+	 * deleted without ever being given to driver (merged with another
+	 * request).
+	 */
+	if (&rq->queuelist == q->last_merge)
+		q->last_merge = NULL;
+
 	if (e->elevator_remove_req_fn)
 		e->elevator_remove_req_fn(q, rq);
 }
@@ -357,6 +389,7 @@ module_init(elevator_global_init);
 
 EXPORT_SYMBOL(elevator_noop);
 
+EXPORT_SYMBOL(elv_add_request);
 EXPORT_SYMBOL(__elv_add_request);
 EXPORT_SYMBOL(elv_next_request);
 EXPORT_SYMBOL(elv_remove_request);

drivers/block/umem.c

@@ -548,12 +548,7 @@ static void process_page(unsigned long data)
 
 		return_bio = bio->bi_next;
 		bio->bi_next = NULL;
-		/* should use bio_endio(), however already cleared
-		 * BIO_UPTODATE. so set bio->bi_size = 0 manually to indicate
-		 * completely done
-		 */
-		bio->bi_size = 0;
-		bio->bi_end_io(bio, bytes, 0);
+		bio_endio(bio, bio->bi_size, 0);
 	}
 }
 
@@ -1041,7 +1036,7 @@ static int __devinit mm_pci_probe(struct pci_dev *dev, const struct pci_device_i
 
         spin_lock_init(&card->lock);
 
-	dev->driver_data = card;
+	pci_set_drvdata(dev, card);
 
 	if (pci_write_cmd != 0x0F) 	/* If not Memory Write & Invalidate */
 		pci_write_cmd = 0x07;	/* then Memory Write command */
@@ -1100,7 +1095,7 @@ static int __devinit mm_pci_probe(struct pci_dev *dev, const struct pci_device_i
 */
 static void mm_pci_remove(struct pci_dev *dev)
 {
-	struct cardinfo *card = dev->driver_data;
+	struct cardinfo *card = pci_get_drvdata(dev);
 
 	tasklet_kill(&card->tasklet);
 	iounmap(card->csr_remap);

drivers/ide/ide-disk.c

@@ -1610,56 +1610,6 @@ static void idedisk_add_settings(ide_drive_t *drive)
 #endif
 }
 
-static int idedisk_suspend(struct device *dev, u32 state, u32 level)
-{
-	ide_drive_t *drive = dev->driver_data;
-
-	printk("Suspending device %p\n", dev->driver_data);
-
-	/* I hope that every freeze operation from the upper levels have
-	 * already been done...
-	 */
-
-	if (level != SUSPEND_SAVE_STATE)
-		return 0;
-	BUG_ON(in_interrupt());
-
-	printk("Waiting for commands to finish\n");
-
-	/* wait until all commands are finished */
-	/* FIXME: waiting for spinlocks should be done instead. */
-	if (!(HWGROUP(drive)))
-		printk("No hwgroup?\n");
-	while (HWGROUP(drive)->handler)
-		yield();
-
-	/* set the drive to standby */
-	printk(KERN_INFO "suspending: %s ", drive->name);
-	if (drive->driver) {
-		if (drive->driver->standby)
-			drive->driver->standby(drive);
-	}
-	drive->blocked = 1;
-
-	while (HWGROUP(drive)->handler)
-		yield();
-
-	return 0;
-}
-
-static int idedisk_resume(struct device *dev, u32 level)
-{
-	ide_drive_t *drive = dev->driver_data;
-
-	if (level != RESUME_RESTORE_STATE)
-		return 0;
-	if (!drive->blocked)
-		panic("ide: Resume but not suspended?\n");
-
-	drive->blocked = 0;
-	return 0;
-}
-
 
 /* This is just a hook for the overall driver tree.
  */

drivers/ide/ide-floppy.c

@@ -1238,6 +1238,21 @@ static void idefloppy_create_rw_cmd (idefloppy_floppy_t *floppy, idefloppy_pc_t
 	set_bit(PC_DMA_RECOMMENDED, &pc->flags);
 }
 
+static int
+idefloppy_blockpc_cmd(idefloppy_floppy_t *floppy, idefloppy_pc_t *pc, struct request *rq)
+{
+	/*
+	 * just support eject for now, it would not be hard to make the
+	 * REQ_BLOCK_PC support fully-featured
+	 */
+	if (rq->cmd[0] != IDEFLOPPY_START_STOP_CMD)
+		return 1;
+
+	idefloppy_init_pc(pc);
+	memcpy(pc->c, rq->cmd, sizeof(pc->c));
+	return 0;
+}
+
 /*
  *	idefloppy_do_request is our request handling function.	
  */
@@ -1280,6 +1295,12 @@ static ide_startstop_t idefloppy_do_request (ide_drive_t *drive, struct request
 		idefloppy_create_rw_cmd(floppy, pc, rq, block);
 	} else if (rq->flags & REQ_SPECIAL) {
 		pc = (idefloppy_pc_t *) rq->buffer;
+	} else if (rq->flags & REQ_BLOCK_PC) {
+		pc = idefloppy_next_pc_storage(drive);
+		if (idefloppy_blockpc_cmd(floppy, pc, rq)) {
+			idefloppy_do_end_request(drive, 0, 0);
+			return ide_stopped;
+		}
 	} else {
 		blk_dump_rq_flags(rq,
 			"ide-floppy: unsupported command in queue");

drivers/ide/ide.c

@@ -878,13 +878,12 @@ ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 {
 	ide_startstop_t startstop;
 	unsigned long block;
-	ide_hwif_t *hwif = HWIF(drive);
 
 	BUG_ON(!(rq->flags & REQ_STARTED));
 
 #ifdef DEBUG
 	printk("%s: start_request: current=0x%08lx\n",
-		hwif->name, (unsigned long) rq);
+		HWIF(drive)->name, (unsigned long) rq);
 #endif
 
 	/* bail early if we've exceeded max_failures */
@@ -910,7 +909,7 @@ ide_startstop_t start_request (ide_drive_t *drive, struct request *rq)
 		block = 1;  /* redirect MBR access to EZ-Drive partn table */
 
 #if (DISK_RECOVERY_TIME > 0)
-	while ((read_timer() - hwif->last_time) < DISK_RECOVERY_TIME);
+	while ((read_timer() - HWIF(drive)->last_time) < DISK_RECOVERY_TIME);
 #endif
 
 	SELECT_DRIVE(drive);
@@ -1128,9 +1127,15 @@ void ide_do_request (ide_hwgroup_t *hwgroup, int masked_irq)
 			break;
 		}
 
+		/*
+		 * we know that the queue isn't empty, but this can happen
+		 * if the q->prep_rq_fn() decides to kill a request
+		 */
 		rq = elv_next_request(&drive->queue);
-		if (!rq)
+		if (!rq) {
+			hwgroup->busy = !!ata_pending_commands(drive);
 			break;
+		}
 
 		if (!rq->bio && ata_pending_commands(drive))
 			break;
@@ -1515,10 +1520,8 @@ int ide_do_drive_cmd (ide_drive_t *drive, struct request *rq, ide_action_t actio
 {
 	unsigned long flags;
 	ide_hwgroup_t *hwgroup = HWGROUP(drive);
-	unsigned int major = HWIF(drive)->major;
-	request_queue_t *q = &drive->queue;
-	struct list_head *queue_head = &q->queue_head;
 	DECLARE_COMPLETION(wait);
+	int insert_end = 1, err;
 
 #ifdef CONFIG_BLK_DEV_PDC4030
 	if (HWIF(drive)->chipset == ide_pdc4030 && rq->buffer != NULL)
@@ -1540,29 +1543,35 @@ int ide_do_drive_cmd (ide_drive_t *drive, struct request *rq, ide_action_t actio
 	}
 
 	rq->rq_disk = drive->disk;
-	if (action == ide_wait)
+
+	/*
+	 * we need to hold an extra reference to request for safe inspection
+	 * after completion
+	 */
+	if (action == ide_wait) {
+		rq->ref_count++;
 		rq->waiting = &wait;
+	}
+
 	spin_lock_irqsave(&ide_lock, flags);
-	if (blk_queue_empty(q) || action == ide_preempt) {
-		if (action == ide_preempt)
-			hwgroup->rq = NULL;
-	} else {
-		if (action == ide_wait || action == ide_end) {
-			queue_head = queue_head->prev;
-		} else
-			queue_head = queue_head->next;
+	if (action == ide_preempt) {
+		hwgroup->rq = NULL;
+		insert_end = 0;
 	}
-	q->elevator.elevator_add_req_fn(q, rq, queue_head);
+	__elv_add_request(&drive->queue, rq, insert_end, 0);
 	ide_do_request(hwgroup, 0);
 	spin_unlock_irqrestore(&ide_lock, flags);
+
+	err = 0;
 	if (action == ide_wait) {
-		/* wait for it to be serviced */
 		wait_for_completion(&wait);
-		/* return -EIO if errors */
-		return rq->errors ? -EIO : 0;
+		if (rq->errors)
+			err = -EIO;
+
+		blk_put_request(rq);
 	}
-	return 0;
 
+	return err;
 }
 
 EXPORT_SYMBOL(ide_do_drive_cmd);
@@ -3369,7 +3378,7 @@ int ide_register_driver(ide_driver_t *driver)
 		list_del_init(&drive->list);
 		ata_attach(drive);
 	}
-	driver->gen_driver.name = driver->name;
+	driver->gen_driver.name = (char *) driver->name;
 	driver->gen_driver.bus = &ide_bus_type;
 	driver->gen_driver.remove = ide_drive_remove;
 	return driver_register(&driver->gen_driver);

drivers/md/raid0.c

@@ -168,8 +168,7 @@ static int create_strip_zones (mddev_t *mddev)
  *	@bio: the buffer head that's been built up so far
  *	@biovec: the request that could be merged to it.
  *
- *	Return 1 if the merge is not permitted (because the
- *	result would cross a chunk boundary), 0 otherwise.
+ *	Return amount of bytes we can accept at this offset
  */
 static int raid0_mergeable_bvec(request_queue_t *q, struct bio *bio, struct bio_vec *biovec)
 {
@@ -182,7 +181,7 @@ static int raid0_mergeable_bvec(request_queue_t *q, struct bio *bio, struct bio_
 	block = bio->bi_sector >> 1;
 	bio_sz = (bio->bi_size + biovec->bv_len) >> 10;
 
-	return chunk_size < ((block & (chunk_size - 1)) + bio_sz);
+	return (chunk_size - ((block & (chunk_size - 1)) + bio_sz)) << 10;
 }
 
 static int raid0_run (mddev_t *mddev)