Merge bk://bk.arm.linux.org.uk/linux-2.5-rmk

into home.transmeta.com:/home/torvalds/v2.5/linux

Documentation/usb/dma.txt

@@ -15,10 +15,12 @@ OR:  they can now be DMA-aware.
   manage dma mappings for existing dma-ready buffers (see below).
 
 - URBs have an additional "transfer_dma" field, as well as a transfer_flags
-  bit saying if it's valid.  (Control requests also needed "setup_dma".) 
+  bit saying if it's valid.  (Control requests also have "setup_dma" and a
+  corresponding transfer_flags bit.)
 
-- "usbcore" will map those DMA addresses, if a DMA-aware driver didn't do it
-  first and set URB_NO_DMA_MAP.  HCDs don't manage dma mappings for urbs.
+- "usbcore" will map those DMA addresses, if a DMA-aware driver didn't do
+  it first and set URB_NO_TRANSFER_DMA_MAP or URB_NO_SETUP_DMA_MAP.  HCDs
+  don't manage dma mappings for URBs.
 
 - There's a new "generic DMA API", parts of which are usable by USB device
   drivers.  Never use dma_set_mask() on any USB interface or device; that
@@ -33,8 +35,9 @@ and effects like cache-trashing can impose subtle penalties.
 - When you're allocating a buffer for DMA purposes anyway, use the buffer
   primitives.  Think of them as kmalloc and kfree that give you the right
   kind of addresses to store in urb->transfer_buffer and urb->transfer_dma,
-  while guaranteeing that hidden copies through DMA "bounce" buffers won't
-  slow things down.  You'd also set URB_NO_DMA_MAP in urb->transfer_flags:
+  while guaranteeing that no hidden copies through DMA "bounce" buffers will
+  slow things down.  You'd also set URB_NO_TRANSFER_DMA_MAP in
+  urb->transfer_flags:
 
 	void *usb_buffer_alloc (struct usb_device *dev, size_t size,
 		int mem_flags, dma_addr_t *dma);
@@ -42,10 +45,18 @@ and effects like cache-trashing can impose subtle penalties.
 	void usb_buffer_free (struct usb_device *dev, size_t size,
 		void *addr, dma_addr_t dma);
 
+  For control transfers you can use the buffer primitives or not for each
+  of the transfer buffer and setup buffer independently.  Set the flag bits
+  URB_NO_TRANSFER_DMA_MAP and URB_NO_SETUP_DMA_MAP to indicate which
+  buffers you have prepared.  For non-control transfers URB_NO_SETUP_DMA_MAP
+  is ignored.
+
   The memory buffer returned is "dma-coherent"; sometimes you might need to
   force a consistent memory access ordering by using memory barriers.  It's
   not using a streaming DMA mapping, so it's good for small transfers on
-  systems where the I/O would otherwise tie up an IOMMU mapping.
+  systems where the I/O would otherwise tie up an IOMMU mapping.  (See
+  Documentation/DMA-mapping.txt for definitions of "coherent" and "streaming"
+  DMA mappings.)
 
   Asking for 1/Nth of a page (as well as asking for N pages) is reasonably
   space-efficient.
@@ -91,7 +102,8 @@ DMA address space of the device.
 
   These calls all work with initialized urbs:  urb->dev, urb->pipe,
   urb->transfer_buffer, and urb->transfer_buffer_length must all be
-  valid when these calls are used:
+  valid when these calls are used (urb->setup_packet must be valid too
+  if urb is a control request):
 
 	struct urb *usb_buffer_map (struct urb *urb);
 
@@ -99,6 +111,6 @@ DMA address space of the device.
 
 	void usb_buffer_unmap (struct urb *urb);
 
-  The calls manage urb->transfer_dma for you, and set URB_NO_DMA_MAP so that
-  usbcore won't map or unmap the buffer.
-
+  The calls manage urb->transfer_dma for you, and set URB_NO_TRANSFER_DMA_MAP
+  so that usbcore won't map or unmap the buffer.  The same goes for
+  urb->setup_dma and URB_NO_SETUP_DMA_MAP for control requests.

arch/ia64/Kconfig

@@ -26,6 +26,10 @@ config RWSEM_XCHGADD_ALGORITHM
 	bool
 	default y
 
+config TIME_INTERPOLATION
+	bool
+	default y
+
 choice
 	prompt "IA-64 processor type"
 	default ITANIUM
@@ -63,7 +67,7 @@ config IA64_GENERIC
 	  HP-simulator   For the HP simulator
 	  (<http://software.hp.com/ia64linux/>).
 	  HP-zx1         For HP zx1-based systems.
-	  SN1-simulator  For the SGI SN1 simulator.
+	  SGI-SN2	 For SGI Altix systems
 	  DIG-compliant  For DIG ("Developer's Interface Guide") compliant
 	  systems.
 
@@ -82,9 +86,6 @@ config IA64_HP_ZX1
 	  for the zx1 I/O MMU and makes root bus bridges appear in PCI config
 	  space (required for zx1 agpgart support).
 
-config IA64_SGI_SN1
-	bool "SGI-SN1"
-
 config IA64_SGI_SN2
 	bool "SGI-SN2"
 
@@ -190,8 +191,8 @@ config ITANIUM_BSTEP_SPECIFIC
 # align cache-sensitive data to 128 bytes
 config IA64_L1_CACHE_SHIFT
 	int
-	default "7" if MCKINLEY || ITANIUM && IA64_SGI_SN1
-	default "6" if ITANIUM && !IA64_SGI_SN1
+	default "7" if MCKINLEY
+	default "6" if ITANIUM
 
 # align cache-sensitive data to 64 bytes
 config MCKINLEY_ASTEP_SPECIFIC
@@ -210,7 +211,7 @@ config MCKINLEY_A0_SPECIFIC
 
 config NUMA
 	bool "Enable NUMA support" if IA64_GENERIC || IA64_DIG || IA64_HP_ZX1
-	default y if IA64_SGI_SN1 || IA64_SGI_SN2
+	default y if IA64_SGI_SN2
 	help
 	  Say Y to compile the kernel to support NUMA (Non-Uniform Memory
 	  Access).  This option is for configuring high-end multiprocessor
@@ -234,7 +235,7 @@ endchoice
 
 config DISCONTIGMEM
 	bool
-	depends on IA64_SGI_SN1 || IA64_SGI_SN2 || (IA64_GENERIC || IA64_DIG || IA64_HP_ZX1) && NUMA
+	depends on IA64_SGI_SN2 || (IA64_GENERIC || IA64_DIG || IA64_HP_ZX1) && NUMA
 	default y
 	help
 	  Say Y to support efficient handling of discontiguous physical memory,
@@ -259,7 +260,7 @@ config VIRTUAL_MEM_MAP
 
 config IA64_MCA
 	bool "Enable IA-64 Machine Check Abort" if IA64_GENERIC || IA64_DIG || IA64_HP_ZX1
-	default y if IA64_SGI_SN1 || IA64_SGI_SN2
+	default y if IA64_SGI_SN2
 	help
 	  Say Y here to enable machine check support for IA-64.  If you're
 	  unsure, answer Y.
@@ -288,17 +289,12 @@ config PM
 
 config IOSAPIC
 	bool
-	depends on IA64_GENERIC || IA64_DIG || IA64_HP_ZX1
-	default y
-
-config IA64_SGI_SN
-	bool
-	depends on IA64_SGI_SN1 || IA64_SGI_SN2
+	depends on IA64_GENERIC || IA64_DIG || IA64_HP_ZX1 || IA64_SGI_SN2
 	default y
 
 config IA64_SGI_SN_DEBUG
 	bool "Enable extra debugging code"
-	depends on IA64_SGI_SN1 || IA64_SGI_SN2
+	depends on IA64_SGI_SN2
 	help
 	  Turns on extra debugging code in the SGI SN (Scalable NUMA) platform
 	  for IA-64.  Unless you are debugging problems on an SGI SN IA-64 box,
@@ -306,14 +302,14 @@ config IA64_SGI_SN_DEBUG
 
 config IA64_SGI_SN_SIM
 	bool "Enable SGI Medusa Simulator Support"
-	depends on IA64_SGI_SN1 || IA64_SGI_SN2
+	depends on IA64_SGI_SN2
 	help
 	  If you are compiling a kernel that will run under SGI's IA-64
 	  simulator (Medusa) then say Y, otherwise say N.
 
 config IA64_SGI_AUTOTEST
 	bool "Enable autotest (llsc). Option to run cache test instead of booting"
-	depends on IA64_SGI_SN1 || IA64_SGI_SN2
+	depends on IA64_SGI_SN2
 	help
 	  Build a kernel used for hardware validation. If you include the
 	  keyword "autotest" on the boot command line, the kernel does NOT boot.
@@ -323,7 +319,7 @@ config IA64_SGI_AUTOTEST
 
 config SERIAL_SGI_L1_PROTOCOL
 	bool "Enable protocol mode for the L1 console"
-	depends on IA64_SGI_SN1 || IA64_SGI_SN2
+	depends on IA64_SGI_SN2
 	help
 	  Uses protocol mode instead of raw mode for the level 1 console on the
 	  SGI SN (Scalable NUMA) platform for IA-64.  If you are compiling for
@@ -331,17 +327,9 @@ config SERIAL_SGI_L1_PROTOCOL
 
 config PERCPU_IRQ
 	bool
-	depends on IA64_SGI_SN1 || IA64_SGI_SN2
+	depends on IA64_SGI_SN2
 	default y
 
-config PCIBA
-	tristate "PCIBA support"
-	depends on IA64_SGI_SN1 || IA64_SGI_SN2
-	help
-	  IRIX PCIBA-inspired user mode PCI interface for the SGI SN (Scalable
-	  NUMA) platform for IA-64.  Unless you are compiling a kernel for an
-	  SGI SN IA-64 box, say N.
-
 # On IA-64, we always want an ELF /proc/kcore.
 config KCORE_ELF
 	bool
@@ -698,20 +686,18 @@ endmenu
 
 source "drivers/usb/Kconfig"
 
-source "lib/Kconfig"
 
 source "net/bluetooth/Kconfig"
 
 endif
 
+source "lib/Kconfig"
+
 source "arch/ia64/hp/sim/Kconfig"
 
 
 menu "Kernel hacking"
 
-config FSYS
-	bool "Light-weight system-call support (via epc)"
-
 choice
 	prompt "Physical memory granularity"
 	default IA64_GRANULE_64MB
@@ -778,7 +764,7 @@ config MAGIC_SYSRQ
 
 config IA64_EARLY_PRINTK
 	bool "Early printk support"
-	depends on DEBUG_KERNEL
+	depends on DEBUG_KERNEL && !IA64_GENERIC
 	help
 	  Selecting this option uses the VGA screen or serial console for
 	  printk() output before the consoles are initialised.  It is useful

arch/ia64/Makefile

@@ -18,8 +18,8 @@ LDFLAGS_MODULE	+= -T arch/ia64/module.lds
 AFLAGS_KERNEL	:= -mconstant-gp
 EXTRA		:=
 
-cflags-y	:= -pipe $(EXTRA) -ffixed-r13 -mfixed-range=f10-f15,f32-f127 \
-		   -falign-functions=32
+cflags-y	:= -pipe $(EXTRA) -ffixed-r13 -mfixed-range=f12-f15,f32-f127 \
+		   -falign-functions=32 -frename-registers
 CFLAGS_KERNEL	:= -mconstant-gp
 
 GCC_VERSION=$(shell $(CC) -v 2>&1 | fgrep 'gcc version' | cut -f3 -d' ' | cut -f1 -d'.')
@@ -27,6 +27,10 @@ GCC_MINOR_VERSION=$(shell $(CC) -v 2>&1 | fgrep 'gcc version' | cut -f3 -d' ' |
 
 GAS_STATUS=$(shell arch/ia64/scripts/check-gas $(CC) $(OBJDUMP))
 
+arch-cppflags	:= $(shell arch/ia64/scripts/toolchain-flags $(CC) $(LD) $(OBJDUMP))
+cflags-y	+= $(arch-cppflags)
+AFLAGS		+= $(arch-cppflags)
+
 ifeq ($(GAS_STATUS),buggy)
 $(error Sorry, you need a newer version of the assember, one that is built from	\
 	a source-tree that post-dates 18-Dec-2002.  You can find a pre-compiled	\
@@ -35,19 +39,18 @@ $(error Sorry, you need a newer version of the assember, one that is built from
 		ftp://ftp.hpl.hp.com/pub/linux-ia64/gas-030124.tar.gz)
 endif
 
-ifneq ($(GCC_VERSION),2)
-	cflags-$(CONFIG_ITANIUM) += -frename-registers
+ifeq ($(GCC_VERSION),2)
+$(error Sorry, your compiler is too old.  GCC v2.96 is known to generate bad code.)
 endif
 
 ifeq ($(GCC_VERSION),3)
  ifeq ($(GCC_MINOR_VERSION),4)
-	cflags-$(CONFIG_ITANIUM) += -mtune=merced
-	cflags-$(CONFIG_MCKINLEY) += -mtune=mckinley
+	cflags-$(CONFIG_ITANIUM)	+= -mtune=merced
+	cflags-$(CONFIG_MCKINLEY)	+= -mtune=mckinley
  endif
 endif
 
 cflags-$(CONFIG_ITANIUM_BSTEP_SPECIFIC)	+= -mb-step
-cflags-$(CONFIG_IA64_SGI_SN)		+= -DBRINGUP
 
 CFLAGS += $(cflags-y)
 head-y := arch/ia64/kernel/head.o arch/ia64/kernel/init_task.o
@@ -58,7 +61,7 @@ core-$(CONFIG_IA32_SUPPORT)	+= arch/ia64/ia32/
 core-$(CONFIG_IA64_DIG) 	+= arch/ia64/dig/
 core-$(CONFIG_IA64_GENERIC) 	+= arch/ia64/dig/
 core-$(CONFIG_IA64_HP_ZX1)	+= arch/ia64/dig/
-core-$(CONFIG_IA64_SGI_SN)	+= arch/ia64/sn/
+core-$(CONFIG_IA64_SGI_SN2)	+= arch/ia64/sn/
 
 drivers-$(CONFIG_PCI)		+= arch/ia64/pci/
 drivers-$(CONFIG_IA64_HP_SIM)	+= arch/ia64/hp/sim/
@@ -66,33 +69,37 @@ drivers-$(CONFIG_IA64_HP_ZX1)	+= arch/ia64/hp/common/ arch/ia64/hp/zx1/
 drivers-$(CONFIG_IA64_GENERIC)	+= arch/ia64/hp/common/ arch/ia64/hp/zx1/ arch/ia64/hp/sim/
 
 boot := arch/ia64/boot
-tools := arch/ia64/tools
-
-.PHONY: boot compressed include/asm-ia64/offsets.h
 
-all: prepare vmlinux
+.PHONY: boot compressed check
 
 compressed: vmlinux.gz
 
 vmlinux.gz: vmlinux
-	$(Q)$(MAKE) $(build)=$(boot) vmlinux.gz
+	$(Q)$(MAKE) $(build)=$(boot) $@
 
 check: vmlinux
-	arch/ia64/scripts/unwcheck.sh vmlinux
+	arch/ia64/scripts/unwcheck.sh $<
 
 archclean:
 	$(Q)$(MAKE) $(clean)=$(boot)
-	$(Q)$(MAKE) $(clean)=$(tools)
 
-CLEAN_FILES += include/asm-ia64/offsets.h vmlinux.gz bootloader
+CLEAN_FILES += include/asm-ia64/.offsets.h.stamp include/asm-ia64/offsets.h vmlinux.gz bootloader
 
 prepare: include/asm-ia64/offsets.h
 
+include/asm-$(ARCH)/offsets.h: arch/$(ARCH)/kernel/asm-offsets.s
+	$(call filechk,gen-asm-offsets)
+
+arch/ia64/kernel/asm-offsets.s: include/asm-ia64/.offsets.h.stamp
+
+include/asm-ia64/.offsets.h.stamp:
+	[ -s include/asm-ia64/offsets.h ] \
+	 || echo "#define IA64_TASK_SIZE 0" > include/asm-ia64/offsets.h
+	touch $@
+
 boot:	lib/lib.a vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $@
 
-include/asm-ia64/offsets.h: include/asm include/linux/version.h include/config/MARKER
-	$(Q)$(MAKE) $(build)=$(tools) $@
 
 define archhelp
   echo '  compressed	- Build compressed kernel image'

arch/ia64/boot/bootloader.c

@@ -55,6 +55,9 @@ struct disk_stat {
 
 #include "../kernel/fw-emu.c"
 
+/* This needs to be defined because lib/string.c:strlcat() calls it in case of error... */
+asm (".global printk; printk = 0");
+
 /*
  * Set a break point on this function so that symbols are available to set breakpoints in
  * the kernel being debugged.
@@ -181,10 +184,10 @@ _start (void)
 			continue;
 
 		req.len = elf_phdr->p_filesz;
-		req.addr = __pa(elf_phdr->p_vaddr);
+		req.addr = __pa(elf_phdr->p_paddr);
 		ssc(fd, 1, (long) &req, elf_phdr->p_offset, SSC_READ);
 		ssc((long) &stat, 0, 0, 0, SSC_WAIT_COMPLETION);
-		memset((char *)__pa(elf_phdr->p_vaddr) + elf_phdr->p_filesz, 0,
+		memset((char *)__pa(elf_phdr->p_paddr) + elf_phdr->p_filesz, 0,
 		       elf_phdr->p_memsz - elf_phdr->p_filesz);
 	}
 	ssc(fd, 0, 0, 0, SSC_CLOSE);

arch/ia64/hp/common/sba_iommu.c

@@ -1682,6 +1682,10 @@ ioc_init(u64 hpa, void *handle)
 	ioc_resource_init(ioc);
 	ioc_sac_init(ioc);
 
+	if ((long) ~IOVP_MASK > (long) ia64_max_iommu_merge_mask)
+		ia64_max_iommu_merge_mask = ~IOVP_MASK;
+	MAX_DMA_ADDRESS = ~0UL;
+
 	printk(KERN_INFO PFX
 		"%s %d.%d HPA 0x%lx IOVA space %dMb at 0x%lx\n",
 		ioc->name, (ioc->rev >> 4) & 0xF, ioc->rev & 0xF,
@@ -1898,22 +1902,26 @@ acpi_sba_ioc_add(struct acpi_device *device)
 	struct ioc *ioc;
 	acpi_status status;
 	u64 hpa, length;
-	struct acpi_device_info dev_info;
+	struct acpi_buffer buffer;
+	struct acpi_device_info *dev_info;
 
 	status = hp_acpi_csr_space(device->handle, &hpa, &length);
 	if (ACPI_FAILURE(status))
 		return 1;
 
-	status = acpi_get_object_info(device->handle, &dev_info);
+	buffer.length = ACPI_ALLOCATE_LOCAL_BUFFER;
+	status = acpi_get_object_info(device->handle, &buffer);
 	if (ACPI_FAILURE(status))
 		return 1;
+	dev_info = buffer.pointer;
 
 	/*
 	 * For HWP0001, only SBA appears in ACPI namespace.  It encloses the PCI
 	 * root bridges, and its CSR space includes the IOC function.
 	 */
-	if (strncmp("HWP0001", dev_info.hardware_id, 7) == 0)
+	if (strncmp("HWP0001", dev_info->hardware_id.value, 7) == 0)
 		hpa += ZX1_IOC_OFFSET;
+	ACPI_MEM_FREE(dev_info);
 
 	ioc = ioc_init(hpa, device->handle);
 	if (!ioc)
@@ -1933,8 +1941,6 @@ static struct acpi_driver acpi_sba_ioc_driver = {
 static int __init
 sba_init(void)
 {
-	MAX_DMA_ADDRESS = ~0UL;
-
 	acpi_bus_register_driver(&acpi_sba_ioc_driver);
 
 #ifdef CONFIG_PCI

arch/ia64/hp/sim/Kconfig

@@ -8,6 +8,10 @@ config HP_SIMETH
 config HP_SIMSERIAL
 	bool "Simulated serial driver support"
 
+config HP_SIMSERIAL_CONSOLE
+	bool "Console for HP simulator"
+	depends on HP_SIMSERIAL
+
 config HP_SIMSCSI
 	bool "Simulated SCSI disk"
 	depends on SCSI

arch/ia64/hp/sim/Makefile

@@ -7,9 +7,10 @@
 # Copyright (C) Srinivasa Thirumalachar (sprasad@engr.sgi.com)
 #
 
-obj-y := hpsim_console.o hpsim_irq.o hpsim_setup.o
+obj-y := hpsim_irq.o hpsim_setup.o
 obj-$(CONFIG_IA64_GENERIC) += hpsim_machvec.o
 
 obj-$(CONFIG_HP_SIMETH)	+= simeth.o
 obj-$(CONFIG_HP_SIMSERIAL) += simserial.o
+obj-$(CONFIG_HP_SIMSERIAL_CONSOLE) += hpsim_console.o
 obj-$(CONFIG_HP_SIMSCSI) += simscsi.o

arch/ia64/hp/sim/hpsim_setup.c

@@ -5,6 +5,7 @@
  *	David Mosberger-Tang <davidm@hpl.hp.com>
  * Copyright (C) 1999 Vijay Chander <vijay@engr.sgi.com>
  */
+#include <linux/config.h>
 #include <linux/console.h>
 #include <linux/init.h>
 #include <linux/kdev_t.h>
@@ -24,8 +25,6 @@
 
 #include "hpsim_ssc.h"
 
-extern struct console hpsim_cons;
-
 /*
  * Simulator system call.
  */
@@ -56,5 +55,11 @@ hpsim_setup (char **cmdline_p)
 {
 	ROOT_DEV = Root_SDA1;		/* default to first SCSI drive */
 
-	register_console(&hpsim_cons);
+#ifdef CONFIG_HP_SIMSERIAL_CONSOLE
+	{
+		extern struct console hpsim_cons;
+		if (ia64_platform_is("hpsim"))
+			register_console(&hpsim_cons);
+	}
+#endif
 }

arch/ia64/hp/sim/simserial.c

@@ -1031,6 +1031,9 @@ simrs_init (void)
 	int			i;
 	struct serial_state	*state;
 
+	if (!ia64_platform_is("hpsim"))
+		return -ENODEV;
+
 	hp_simserial_driver = alloc_tty_driver(1);
 	if (!hp_simserial_driver)
 		return -ENOMEM;

arch/ia64/ia32/binfmt_elf32.c

@@ -16,7 +16,8 @@
 
 #include <asm/param.h>
 #include <asm/signal.h>
-#include <asm/ia32.h>
+
+#include "ia32priv.h"
 
 #define CONFIG_BINFMT_ELF32
 

arch/ia64/ia32/ia32_entry.S

@@ -44,14 +44,8 @@ ENTRY(ia32_clone)
 	br.call.sptk.many rp=do_fork
 .ret0:	.restore sp
 	adds sp=IA64_SWITCH_STACK_SIZE,sp	// pop the switch stack
-	mov r2=-1000
-	adds r3=IA64_TASK_PID_OFFSET,r8
-	;;
-	cmp.leu p6,p0=r8,r2
 	mov ar.pfs=loc1
 	mov rp=loc0
-	;;
-(p6)	ld4 r8=[r3]
 	br.ret.sptk.many rp
 END(ia32_clone)
 
@@ -183,14 +177,8 @@ GLOBAL_ENTRY(sys32_fork)
 	br.call.sptk.few rp=do_fork
 .ret5:	.restore sp
 	adds sp=IA64_SWITCH_STACK_SIZE,sp	// pop the switch stack
-	mov r2=-1000
-	adds r3=IA64_TASK_PID_OFFSET,r8
-	;;
-	cmp.leu p6,p0=r8,r2
 	mov ar.pfs=loc1
 	mov rp=loc0
-	;;
-(p6)	ld4 r8=[r3]
 	br.ret.sptk.many rp
 END(sys32_fork)
 
@@ -439,8 +427,8 @@ ia32_syscall_table:
 	data8 sys_ni_syscall
 	data8 sys_ni_syscall
 	data8 compat_sys_futex	/* 240 */
-	data8 compat_sys_setaffinity
-	data8 compat_sys_getaffinity
+	data8 compat_sys_sched_setaffinity
+	data8 compat_sys_sched_getaffinity
 	data8 sys_ni_syscall
 	data8 sys_ni_syscall
 	data8 sys_ni_syscall	/* 245 */

arch/ia64/ia32/ia32_ioctl.c

@@ -11,36 +11,107 @@
 #include <linux/dirent.h>
 #include <linux/fs.h>		/* argh, msdos_fs.h isn't self-contained... */
 #include <linux/signal.h>	/* argh, msdos_fs.h isn't self-contained... */
-
-#include <asm/ia32.h>
-
-#include <linux/msdos_fs.h>
-#include <linux/mtio.h>
-#include <linux/ncp_fs.h>
-#include <linux/capi.h>
-#include <linux/videodev.h>
-#include <linux/synclink.h>
-#include <linux/atmdev.h>
-#include <linux/atm_eni.h>
-#include <linux/atm_nicstar.h>
-#include <linux/atm_zatm.h>
-#include <linux/atm_idt77105.h>
+#include <linux/compat.h>
+
+#include "ia32priv.h"
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+#include <linux/smp_lock.h>
+#include <linux/ioctl.h>
+#include <linux/if.h>
+#include <linux/slab.h>
+#include <linux/hdreg.h>
+#include <linux/raid/md.h>
+#include <linux/kd.h>
+#include <linux/route.h>
+#include <linux/in6.h>
+#include <linux/ipv6_route.h>
+#include <linux/skbuff.h>
+#include <linux/netlink.h>
+#include <linux/vt.h>
+#include <linux/file.h>
+#include <linux/fd.h>
 #include <linux/ppp_defs.h>
 #include <linux/if_ppp.h>
-#include <linux/ixjuser.h>
-#include <linux/i2o-dev.h>
+#include <linux/if_pppox.h>
+#include <linux/mtio.h>
+#include <linux/cdrom.h>
+#include <linux/loop.h>
+#include <linux/auto_fs.h>
+#include <linux/auto_fs4.h>
+#include <linux/devfs_fs.h>
+#include <linux/tty.h>
+#include <linux/vt_kern.h>
+#include <linux/fb.h>
+#include <linux/ext2_fs.h>
+#include <linux/videodev.h>
+#include <linux/netdevice.h>
+#include <linux/raw.h>
+#include <linux/smb_fs.h>
+#include <linux/blkpg.h>
+#include <linux/blk.h>
+#include <linux/elevator.h>
+#include <linux/rtc.h>
+#include <linux/pci.h>
+#include <linux/rtc.h>
+#include <linux/module.h>
+#include <linux/serial.h>
+#include <linux/reiserfs_fs.h>
+#include <linux/if_tun.h>
+#include <linux/dirent.h>
+#include <linux/ctype.h>
+#include <linux/ncp_fs.h>
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/rfcomm.h>
+
 #include <scsi/scsi.h>
 /* Ugly hack. */
-#undef	__KERNEL__
+#undef __KERNEL__
 #include <scsi/scsi_ioctl.h>
-#define	__KERNEL__
+#define __KERNEL__
 #include <scsi/sg.h>
 
+#include <asm/types.h>
+#include <asm/uaccess.h>
+#include <linux/ethtool.h>
+#include <linux/mii.h>
+#include <linux/if_bonding.h>
+#include <linux/watchdog.h>
+
+#include <asm/module.h>
+#include <asm/ioctl32.h>
+#include <linux/soundcard.h>
+#include <linux/lp.h>
+
+#include <linux/atm.h>
+#include <linux/atmarp.h>
+#include <linux/atmclip.h>
+#include <linux/atmdev.h>
+#include <linux/atmioc.h>
+#include <linux/atmlec.h>
+#include <linux/atmmpc.h>
+#include <linux/atmsvc.h>
+#include <linux/atm_tcp.h>
+#include <linux/sonet.h>
+#include <linux/atm_suni.h>
+#include <linux/mtd/mtd.h>
+
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/hci.h>
+
+#include <linux/usb.h>
+#include <linux/usbdevice_fs.h>
+#include <linux/nbd.h>
+#include <linux/random.h>
+#include <linux/filter.h>
+
 #include <../drivers/char/drm/drm.h>
 #include <../drivers/char/drm/mga_drm.h>
 #include <../drivers/char/drm/i810_drm.h>
 
-
 #define IOCTL_NR(a)	((a) & ~(_IOC_SIZEMASK << _IOC_SIZESHIFT))
 
 #define DO_IOCTL(fd, cmd, arg) ({			\
@@ -57,6 +128,9 @@
 
 asmlinkage long sys_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg);
 
+#define	VFAT_IOCTL_READDIR_BOTH32	_IOR('r', 1, struct linux32_dirent[2])
+#define	VFAT_IOCTL_READDIR_SHORT32	_IOR('r', 2, struct linux32_dirent[2])
+
 static long
 put_dirent32 (struct dirent *d, struct linux32_dirent *d32)
 {
@@ -67,6 +141,23 @@ 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));
 }
+
+static int vfat_ioctl32(unsigned fd, unsigned cmd,  void *ptr) 
+{
+	int ret;
+	mm_segment_t oldfs = get_fs();
+	struct dirent d[2]; 
+
+	set_fs(KERNEL_DS);
+	ret = sys_ioctl(fd,cmd,(unsigned long)&d); 
+	set_fs(oldfs); 
+	if (!ret) { 
+		ret |= put_dirent32(&d[0], (struct linux32_dirent *)ptr); 
+		ret |= put_dirent32(&d[1], ((struct linux32_dirent *)ptr) + 1); 
+	}
+	return ret; 
+} 
+
 /*
  *  The transform code for the SG_IO ioctl was brazenly lifted from
  *  the Sparc64 port in the file `arch/sparc64/kernel/ioctl32.c'.
@@ -294,3 +385,83 @@ static int sg_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg)
 	}
 	return err;
 }
+
+static __inline__ void *alloc_user_space(long len)
+{
+	struct pt_regs	*regs = ((struct pt_regs *)((unsigned long) current +
+						    IA64_STK_OFFSET)) - 1;
+	return (void *)regs->r12 - len; 
+}
+
+struct ifmap32 {
+	u32 mem_start;
+	u32 mem_end;
+	unsigned short base_addr;
+	unsigned char irq;
+	unsigned char dma;
+	unsigned char port;
+};
+
+struct ifreq32 {
+#define IFHWADDRLEN     6
+#define IFNAMSIZ        16
+        union {
+                char    ifrn_name[IFNAMSIZ];            /* if name, e.g. "en0" */
+        } ifr_ifrn;
+        union {
+                struct  sockaddr ifru_addr;
+                struct  sockaddr ifru_dstaddr;
+                struct  sockaddr ifru_broadaddr;
+                struct  sockaddr ifru_netmask;
+                struct  sockaddr ifru_hwaddr;
+                short   ifru_flags;
+                int     ifru_ivalue;
+                int     ifru_mtu;
+                struct  ifmap32 ifru_map;
+                char    ifru_slave[IFNAMSIZ];   /* Just fits the size */
+		char	ifru_newname[IFNAMSIZ];
+                compat_caddr_t ifru_data;
+        } ifr_ifru;
+};
+
+int siocdevprivate_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg)
+{
+	struct ifreq *u_ifreq64;
+	struct ifreq32 *u_ifreq32 = (struct ifreq32 *) arg;
+	char tmp_buf[IFNAMSIZ];
+	void *data64;
+	u32 data32;
+
+	if (copy_from_user(&tmp_buf[0], &(u_ifreq32->ifr_ifrn.ifrn_name[0]),
+			   IFNAMSIZ))
+		return -EFAULT;
+	if (__get_user(data32, &u_ifreq32->ifr_ifru.ifru_data))
+		return -EFAULT;
+	data64 = (void *) P(data32);
+
+	u_ifreq64 = alloc_user_space(sizeof(*u_ifreq64));
+
+	/* Don't check these user accesses, just let that get trapped
+	 * in the ioctl handler instead.
+	 */
+	copy_to_user(&u_ifreq64->ifr_ifrn.ifrn_name[0], &tmp_buf[0], IFNAMSIZ);
+	__put_user(data64, &u_ifreq64->ifr_ifru.ifru_data);
+
+	return sys_ioctl(fd, cmd, (unsigned long) u_ifreq64);
+}
+
+typedef int (* ioctl32_handler_t)(unsigned int, unsigned int, unsigned long, struct file *);
+
+#define COMPATIBLE_IOCTL(cmd)		HANDLE_IOCTL((cmd),sys_ioctl)
+#define HANDLE_IOCTL(cmd,handler)	{ (cmd), (ioctl32_handler_t)(handler), NULL },
+#define IOCTL_TABLE_START \
+	struct ioctl_trans ioctl_start[] = {
+#define IOCTL_TABLE_END \
+	}; struct ioctl_trans ioctl_end[0];
+
+IOCTL_TABLE_START
+#include <linux/compat_ioctl.h>
+HANDLE_IOCTL(VFAT_IOCTL_READDIR_BOTH32, vfat_ioctl32)
+HANDLE_IOCTL(VFAT_IOCTL_READDIR_SHORT32, vfat_ioctl32)
+HANDLE_IOCTL(SG_IO,sg_ioctl_trans)
+IOCTL_TABLE_END

arch/ia64/ia32/ia32_ldt.c

@@ -14,7 +14,8 @@
 #include <linux/vmalloc.h>
 
 #include <asm/uaccess.h>
-#include <asm/ia32.h>
+
+#include "ia32priv.h"
 
 #define P(p)	((void *) (unsigned long) (p))
 

arch/ia64/ia32/ia32_signal.c

@@ -28,7 +28,8 @@
 #include <asm/rse.h>
 #include <asm/sigcontext.h>
 #include <asm/segment.h>
-#include <asm/ia32.h>
+
+#include "ia32priv.h"
 
 #include "../kernel/sigframe.h"
 
@@ -179,8 +180,10 @@ copy_siginfo_to_user32 (siginfo_t32 *to, siginfo_t *from)
  *    datasel    ar.fdr(32:47)
  *
  *    _st[(0+TOS)%8]   f8
- *    _st[(1+TOS)%8]   f9                    (f8, f9 from ptregs)
- *      : :            :                     (f10..f15 from live reg)
+ *    _st[(1+TOS)%8]   f9
+ *    _st[(2+TOS)%8]   f10
+ *    _st[(3+TOS)%8]   f11                   (f8..f11 from ptregs)
+ *      : :            :                     (f12..f15 from live reg)
  *      : :            :
  *    _st[(7+TOS)%8]   f15                   TOS=sw.top(bits11:13)
  *
@@ -262,8 +265,8 @@ save_ia32_fpstate_live (struct _fpstate_ia32 *save)
 	__put_user( 0, &save->magic); //#define X86_FXSR_MAGIC   0x0000
 
 	/*
-	 * save f8 and f9  from pt_regs
-	 * save f10..f15 from live register set
+	 * save f8..f11  from pt_regs
+	 * save f12..f15 from live register set
 	 */
 	/*
 	 *  Find the location where f8 has to go in fp reg stack.  This depends on
@@ -278,11 +281,11 @@ save_ia32_fpstate_live (struct _fpstate_ia32 *save)
 	copy_to_user(&save->_st[(0+fr8_st_map)&0x7], fpregp, sizeof(struct _fpreg_ia32));
 	ia64f2ia32f(fpregp, &ptp->f9);
 	copy_to_user(&save->_st[(1+fr8_st_map)&0x7], fpregp, sizeof(struct _fpreg_ia32));
-
-	__stfe(fpregp, 10);
+	ia64f2ia32f(fpregp, &ptp->f10);
 	copy_to_user(&save->_st[(2+fr8_st_map)&0x7], fpregp, sizeof(struct _fpreg_ia32));
-	__stfe(fpregp, 11);
+	ia64f2ia32f(fpregp, &ptp->f11);
 	copy_to_user(&save->_st[(3+fr8_st_map)&0x7], fpregp, sizeof(struct _fpreg_ia32));
+
 	__stfe(fpregp, 12);
 	copy_to_user(&save->_st[(4+fr8_st_map)&0x7], fpregp, sizeof(struct _fpreg_ia32));
 	__stfe(fpregp, 13);
@@ -394,8 +397,8 @@ restore_ia32_fpstate_live (struct _fpstate_ia32 *save)
 	asm volatile ( "mov ar.fdr=%0;" :: "r"(fdr));
 
 	/*
-	 * restore f8, f9 onto pt_regs
-	 * restore f10..f15 onto live registers
+	 * restore f8..f11 onto pt_regs
+	 * restore f12..f15 onto live registers
 	 */
 	/*
 	 *  Find the location where f8 has to go in fp reg stack.  This depends on
@@ -411,11 +414,11 @@ restore_ia32_fpstate_live (struct _fpstate_ia32 *save)
 	ia32f2ia64f(&ptp->f8, fpregp);
 	copy_from_user(fpregp, &save->_st[(1+fr8_st_map)&0x7], sizeof(struct _fpreg_ia32));
 	ia32f2ia64f(&ptp->f9, fpregp);
-
 	copy_from_user(fpregp, &save->_st[(2+fr8_st_map)&0x7], sizeof(struct _fpreg_ia32));
-	__ldfe(10, fpregp);
+	ia32f2ia64f(&ptp->f10, fpregp);
 	copy_from_user(fpregp, &save->_st[(3+fr8_st_map)&0x7], sizeof(struct _fpreg_ia32));
-	__ldfe(11, fpregp);
+	ia32f2ia64f(&ptp->f11, fpregp);
+
 	copy_from_user(fpregp, &save->_st[(4+fr8_st_map)&0x7], sizeof(struct _fpreg_ia32));
 	__ldfe(12, fpregp);
 	copy_from_user(fpregp, &save->_st[(5+fr8_st_map)&0x7], sizeof(struct _fpreg_ia32));
@@ -738,11 +741,11 @@ restore_sigcontext_ia32 (struct pt_regs *regs, struct sigcontext_ia32 *sc, int *
 
 #define COPY(ia64x, ia32x)	err |= __get_user(regs->ia64x, &sc->ia32x)
 
-#define copyseg_gs(tmp)		(regs->r16 |= (unsigned long) tmp << 48)
-#define copyseg_fs(tmp)		(regs->r16 |= (unsigned long) tmp << 32)
+#define copyseg_gs(tmp)		(regs->r16 |= (unsigned long) (tmp) << 48)
+#define copyseg_fs(tmp)		(regs->r16 |= (unsigned long) (tmp) << 32)
 #define copyseg_cs(tmp)		(regs->r17 |= tmp)
-#define copyseg_ss(tmp)		(regs->r17 |= (unsigned long) tmp << 16)
-#define copyseg_es(tmp)		(regs->r16 |= (unsigned long) tmp << 16)
+#define copyseg_ss(tmp)		(regs->r17 |= (unsigned long) (tmp) << 16)
+#define copyseg_es(tmp)		(regs->r16 |= (unsigned long) (tmp) << 16)
 #define copyseg_ds(tmp)		(regs->r16 |= tmp)
 
 #define COPY_SEG(seg)					\

arch/ia64/ia32/ia32_support.c

@@ -22,7 +22,8 @@
 #include <asm/pgtable.h>
 #include <asm/system.h>
 #include <asm/processor.h>
-#include <asm/ia32.h>
+
+#include "ia32priv.h"
 
 extern void die_if_kernel (char *str, struct pt_regs *regs, long err);
 
@@ -60,30 +61,26 @@ ia32_load_segment_descriptors (struct task_struct *task)
 	regs->r27 = load_desc(regs->r16 >>  0);		/* DSD */
 	regs->r28 = load_desc(regs->r16 >> 32);		/* FSD */
 	regs->r29 = load_desc(regs->r16 >> 48);		/* GSD */
-	task->thread.csd = load_desc(regs->r17 >>  0);	/* CSD */
-	task->thread.ssd = load_desc(regs->r17 >> 16);	/* SSD */
+	regs->ar_csd = load_desc(regs->r17 >>  0);	/* CSD */
+	regs->ar_ssd = load_desc(regs->r17 >> 16);	/* SSD */
 }
 
 void
 ia32_save_state (struct task_struct *t)
 {
-	unsigned long eflag, fsr, fcr, fir, fdr, csd, ssd;
+	unsigned long eflag, fsr, fcr, fir, fdr;
 
 	asm ("mov %0=ar.eflag;"
 	     "mov %1=ar.fsr;"
 	     "mov %2=ar.fcr;"
 	     "mov %3=ar.fir;"
 	     "mov %4=ar.fdr;"
-	     "mov %5=ar.csd;"
-	     "mov %6=ar.ssd;"
-	     : "=r"(eflag), "=r"(fsr), "=r"(fcr), "=r"(fir), "=r"(fdr), "=r"(csd), "=r"(ssd));
+	     : "=r"(eflag), "=r"(fsr), "=r"(fcr), "=r"(fir), "=r"(fdr));
 	t->thread.eflag = eflag;
 	t->thread.fsr = fsr;
 	t->thread.fcr = fcr;
 	t->thread.fir = fir;
 	t->thread.fdr = fdr;
-	t->thread.csd = csd;
-	t->thread.ssd = ssd;
 	ia64_set_kr(IA64_KR_IO_BASE, t->thread.old_iob);
 	ia64_set_kr(IA64_KR_TSSD, t->thread.old_k1);
 }
@@ -91,7 +88,7 @@ ia32_save_state (struct task_struct *t)
 void
 ia32_load_state (struct task_struct *t)
 {
-	unsigned long eflag, fsr, fcr, fir, fdr, csd, ssd, tssd;
+	unsigned long eflag, fsr, fcr, fir, fdr, tssd;
 	struct pt_regs *regs = ia64_task_regs(t);
 	int nr = get_cpu();	/* LDT and TSS depend on CPU number: */
 
@@ -100,8 +97,6 @@ ia32_load_state (struct task_struct *t)
 	fcr = t->thread.fcr;
 	fir = t->thread.fir;
 	fdr = t->thread.fdr;
-	csd = t->thread.csd;
-	ssd = t->thread.ssd;
 	tssd = load_desc(_TSS(nr));					/* TSSD */
 
 	asm volatile ("mov ar.eflag=%0;"
@@ -109,9 +104,7 @@ ia32_load_state (struct task_struct *t)
 		      "mov ar.fcr=%2;"
 		      "mov ar.fir=%3;"
 		      "mov ar.fdr=%4;"
-		      "mov ar.csd=%5;"
-		      "mov ar.ssd=%6;"
-		      :: "r"(eflag), "r"(fsr), "r"(fcr), "r"(fir), "r"(fdr), "r"(csd), "r"(ssd));
+		      :: "r"(eflag), "r"(fsr), "r"(fcr), "r"(fir), "r"(fdr));
 	current->thread.old_iob = ia64_get_kr(IA64_KR_IO_BASE);
 	current->thread.old_k1 = ia64_get_kr(IA64_KR_TSSD);
 	ia64_set_kr(IA64_KR_IO_BASE, IA32_IOBASE);
@@ -181,6 +174,13 @@ ia32_bad_interrupt (unsigned long int_num, struct pt_regs *regs)
 	force_sig_info(SIGTRAP, &siginfo, current);
 }
 
+void
+ia32_cpu_init (void)
+{
+	/* initialize global ia32 state - CR0 and CR4 */
+	asm volatile ("mov ar.cflg = %0" :: "r" (((ulong) IA32_CR4 << 32) | IA32_CR0));
+}
+
 static int __init
 ia32_init (void)
 {

arch/ia64/ia32/ia32_traps.c

@@ -12,7 +12,8 @@
 #include <linux/kernel.h>
 #include <linux/sched.h>
 
-#include <asm/ia32.h>
+#include "ia32priv.h"
+
 #include <asm/ptrace.h>
 
 int

arch/ia64/ia32/sys_ia32.c

@@ -53,7 +53,8 @@
 #include <asm/types.h>
 #include <asm/uaccess.h>
 #include <asm/semaphore.h>
-#include <asm/ia32.h>
+
+#include "ia32priv.h"
 
 #include <net/scm.h>
 #include <net/sock.h>
@@ -206,9 +207,8 @@ int cp_compat_stat(struct kstat *stat, struct compat_stat *ubuf)
 
 
 static int
-get_page_prot (unsigned long addr)
+get_page_prot (struct vm_area_struct *vma, unsigned long addr)
 {
-	struct vm_area_struct *vma = find_vma(current->mm, addr);
 	int prot = 0;
 
 	if (!vma || vma->vm_start > addr)
@@ -231,14 +231,26 @@ static unsigned long
 mmap_subpage (struct file *file, unsigned long start, unsigned long end, int prot, int flags,
 	      loff_t off)
 {
-	void *page = (void *) get_zeroed_page(GFP_KERNEL);
+	void *page = NULL;
 	struct inode *inode;
-	unsigned long ret;
-	int old_prot = get_page_prot(start);
+	unsigned long ret = 0;
+	struct vm_area_struct *vma = find_vma(current->mm, start);
+	int old_prot = get_page_prot(vma, start);
 
 	DBG("mmap_subpage(file=%p,start=0x%lx,end=0x%lx,prot=%x,flags=%x,off=0x%llx)\n",
 	    file, start, end, prot, flags, off);
 
+
+	/* Optimize the case where the old mmap and the new mmap are both anonymous */
+	if ((old_prot & PROT_WRITE) && (flags & MAP_ANONYMOUS) && !vma->vm_file) {
+		if (clear_user((void *) start, end - start)) {
+			ret = -EFAULT;
+			goto out;
+		}
+		goto skip_mmap;
+	}
+
+	page = (void *) get_zeroed_page(GFP_KERNEL);
 	if (!page)
 		return -ENOMEM;
 
@@ -263,6 +275,7 @@ mmap_subpage (struct file *file, unsigned long start, unsigned long end, int pro
 			copy_to_user((void *) end, page + PAGE_OFF(end),
 				     PAGE_SIZE - PAGE_OFF(end));
 	}
+
 	if (!(flags & MAP_ANONYMOUS)) {
 		/* read the file contents */
 		inode = file->f_dentry->d_inode;
@@ -273,10 +286,13 @@ mmap_subpage (struct file *file, unsigned long start, unsigned long end, int pro
 			goto out;
 		}
 	}
+
+ skip_mmap:
 	if (!(prot & PROT_WRITE))
 		ret = sys_mprotect(PAGE_START(start), PAGE_SIZE, prot | old_prot);
   out:
-	free_page((unsigned long) page);
+	if (page)
+		free_page((unsigned long) page);
 	return ret;
 }
 
@@ -532,11 +548,12 @@ static long
 mprotect_subpage (unsigned long address, int new_prot)
 {
 	int old_prot;
+	struct vm_area_struct *vma;
 
 	if (new_prot == PROT_NONE)
 		return 0;		/* optimize case where nothing changes... */
-
-	old_prot = get_page_prot(address);
+	vma = find_vma(current->mm, address);
+	old_prot = get_page_prot(vma, address);
 	return sys_mprotect(address, PAGE_SIZE, new_prot | old_prot);
 }
 
@@ -642,7 +659,6 @@ sys32_alarm (unsigned int seconds)
    sorts of things, like timeval and itimerval.  */
 
 extern struct timezone sys_tz;
-extern int do_sys_settimeofday (struct timeval *tv, struct timezone *tz);
 
 asmlinkage long
 sys32_gettimeofday (struct compat_timeval *tv, struct timezone *tz)
@@ -664,18 +680,21 @@ asmlinkage long
 sys32_settimeofday (struct compat_timeval *tv, struct timezone *tz)
 {
 	struct timeval ktv;
+	struct timespec kts;
 	struct timezone ktz;
 
 	if (tv) {
 		if (get_tv32(&ktv, tv))
 			return -EFAULT;
+		kts.tv_sec = ktv.tv_sec;
+		kts.tv_nsec = ktv.tv_usec * 1000;
 	}
 	if (tz) {
 		if (copy_from_user(&ktz, tz, sizeof(ktz)))
 			return -EFAULT;
 	}
 
-	return do_sys_settimeofday(tv ? &ktv : NULL, tz ? &ktz : NULL);
+	return do_sys_settimeofday(tv ? &kts : NULL, tz ? &ktz : NULL);
 }
 
 struct getdents32_callback {
@@ -836,9 +855,8 @@ sys32_select (int n, fd_set *inp, fd_set *outp, fd_set *exp, struct compat_timev
 		}
 	}
 
-	size = FDS_BYTES(n);
 	ret = -EINVAL;
-	if (n < 0 || size < n)
+	if (n < 0)
 		goto out_nofds;
 
 	if (n > current->files->max_fdset)
@@ -850,6 +868,7 @@ sys32_select (int n, fd_set *inp, fd_set *outp, fd_set *exp, struct compat_timev
 	 * long-words.
 	 */
 	ret = -ENOMEM;
+	size = FDS_BYTES(n);
 	bits = kmalloc(6 * size, GFP_KERNEL);
 	if (!bits)
 		goto out_nofds;
@@ -1102,7 +1121,7 @@ struct shmid_ds32 {
 };
 
 struct shmid64_ds32 {
-	struct ipc64_perm shm_perm;
+	struct ipc64_perm32 shm_perm;
 	compat_size_t shm_segsz;
 	compat_time_t   shm_atime;
 	unsigned int __unused1;
@@ -1320,7 +1339,6 @@ static int
 msgctl32 (int first, int second, void *uptr)
 {
 	int err = -EINVAL, err2;
-	struct msqid_ds m;
 	struct msqid64_ds m64;
 	struct msqid_ds32 *up32 = (struct msqid_ds32 *)uptr;
 	struct msqid64_ds32 *up64 = (struct msqid64_ds32 *)uptr;
@@ -1336,21 +1354,21 @@ msgctl32 (int first, int second, void *uptr)
 
 	      case IPC_SET:
 		if (version == IPC_64) {
-			err = get_user(m.msg_perm.uid, &up64->msg_perm.uid);
-			err |= get_user(m.msg_perm.gid, &up64->msg_perm.gid);
-			err |= get_user(m.msg_perm.mode, &up64->msg_perm.mode);
-			err |= get_user(m.msg_qbytes, &up64->msg_qbytes);
+			err = get_user(m64.msg_perm.uid, &up64->msg_perm.uid);
+			err |= get_user(m64.msg_perm.gid, &up64->msg_perm.gid);
+			err |= get_user(m64.msg_perm.mode, &up64->msg_perm.mode);
+			err |= get_user(m64.msg_qbytes, &up64->msg_qbytes);
 		} else {
-			err = get_user(m.msg_perm.uid, &up32->msg_perm.uid);
-			err |= get_user(m.msg_perm.gid, &up32->msg_perm.gid);
-			err |= get_user(m.msg_perm.mode, &up32->msg_perm.mode);
-			err |= get_user(m.msg_qbytes, &up32->msg_qbytes);
+			err = get_user(m64.msg_perm.uid, &up32->msg_perm.uid);
+			err |= get_user(m64.msg_perm.gid, &up32->msg_perm.gid);
+			err |= get_user(m64.msg_perm.mode, &up32->msg_perm.mode);
+			err |= get_user(m64.msg_qbytes, &up32->msg_qbytes);
 		}
 		if (err)
 			break;
 		old_fs = get_fs();
 		set_fs(KERNEL_DS);
-		err = sys_msgctl(first, second, &m);
+		err = sys_msgctl(first, second, &m64);
 		set_fs(old_fs);
 		break;
 
@@ -1430,7 +1448,7 @@ static int
 shmctl32 (int first, int second, void *uptr)
 {
 	int err = -EFAULT, err2;
-	struct shmid_ds s;
+
 	struct shmid64_ds s64;
 	struct shmid_ds32 *up32 = (struct shmid_ds32 *)uptr;
 	struct shmid64_ds32 *up64 = (struct shmid64_ds32 *)uptr;
@@ -1482,19 +1500,19 @@ shmctl32 (int first, int second, void *uptr)
 
 	      case IPC_SET:
 		if (version == IPC_64) {
-			err = get_user(s.shm_perm.uid, &up64->shm_perm.uid);
-			err |= get_user(s.shm_perm.gid, &up64->shm_perm.gid);
-			err |= get_user(s.shm_perm.mode, &up64->shm_perm.mode);
+			err = get_user(s64.shm_perm.uid, &up64->shm_perm.uid);
+			err |= get_user(s64.shm_perm.gid, &up64->shm_perm.gid);
+			err |= get_user(s64.shm_perm.mode, &up64->shm_perm.mode);
 		} else {
-			err = get_user(s.shm_perm.uid, &up32->shm_perm.uid);
-			err |= get_user(s.shm_perm.gid, &up32->shm_perm.gid);
-			err |= get_user(s.shm_perm.mode, &up32->shm_perm.mode);
+			err = get_user(s64.shm_perm.uid, &up32->shm_perm.uid);
+			err |= get_user(s64.shm_perm.gid, &up32->shm_perm.gid);
+			err |= get_user(s64.shm_perm.mode, &up32->shm_perm.mode);
 		}
 		if (err)
 			break;
 		old_fs = get_fs();
 		set_fs(KERNEL_DS);
-		err = sys_shmctl(first, second, &s);
+		err = sys_shmctl(first, second, &s64);
 		set_fs(old_fs);
 		break;
 
@@ -1798,12 +1816,16 @@ put_fpreg (int regno, struct _fpreg_ia32 *reg, struct pt_regs *ptp, struct switc
 		ia64f2ia32f(f, &ptp->f9);
 		break;
 	      case 2:
+		ia64f2ia32f(f, &ptp->f10);
+		break;
 	      case 3:
+		ia64f2ia32f(f, &ptp->f11);
+		break;
 	      case 4:
 	      case 5:
 	      case 6:
 	      case 7:
-		ia64f2ia32f(f, &swp->f10 + (regno - 2));
+		ia64f2ia32f(f, &swp->f12 + (regno - 4));
 		break;
 	}
 	copy_to_user(reg, f, sizeof(*reg));
@@ -1824,12 +1846,16 @@ get_fpreg (int regno, struct _fpreg_ia32 *reg, struct pt_regs *ptp, struct switc
 		copy_from_user(&ptp->f9, reg, sizeof(*reg));
 		break;
 	      case 2:
+		copy_from_user(&ptp->f10, reg, sizeof(*reg));
+		break;
 	      case 3:
+		copy_from_user(&ptp->f11, reg, sizeof(*reg));
+		break;
 	      case 4:
 	      case 5:
 	      case 6:
 	      case 7:
-		copy_from_user(&swp->f10 + (regno - 2), reg, sizeof(*reg));
+		copy_from_user(&swp->f12 + (regno - 4), reg, sizeof(*reg));
 		break;
 	}
 	return;
@@ -1860,7 +1886,7 @@ save_ia32_fpstate (struct task_struct *tsk, struct ia32_user_i387_struct *save)
 	ptp = ia64_task_regs(tsk);
 	tos = (tsk->thread.fsr >> 11) & 7;
 	for (i = 0; i < 8; i++)
-		put_fpreg(i, (struct _fpreg_ia32 *)&save->st_space[4*i], ptp, swp, tos);
+		put_fpreg(i, &save->st_space[i], ptp, swp, tos);
 	return 0;
 }
 
@@ -1893,7 +1919,7 @@ restore_ia32_fpstate (struct task_struct *tsk, struct ia32_user_i387_struct *sav
 	ptp = ia64_task_regs(tsk);
 	tos = (tsk->thread.fsr >> 11) & 7;
 	for (i = 0; i < 8; i++)
-		get_fpreg(i, (struct _fpreg_ia32 *)&save->st_space[4*i], ptp, swp, tos);
+		get_fpreg(i, &save->st_space[i], ptp, swp, tos);
 	return 0;
 }
 

arch/ia64/kernel/Makefile

@@ -4,12 +4,11 @@
 
 extra-y	:= head.o init_task.o
 
-obj-y := acpi.o entry.o efi.o efi_stub.o gate.o ia64_ksyms.o irq.o irq_ia64.o irq_lsapic.o	\
-	 ivt.o machvec.o pal.o perfmon.o process.o ptrace.o sal.o semaphore.o setup.o signal.o	\
-	 sys_ia64.o time.o traps.o unaligned.o unwind.o
+obj-y := acpi.o entry.o efi.o efi_stub.o gate-data.o fsys.o ia64_ksyms.o irq.o irq_ia64.o	\
+	 irq_lsapic.o ivt.o machvec.o pal.o patch.o process.o perfmon.o ptrace.o sal.o		\
+	 semaphore.o setup.o signal.o sys_ia64.o time.o traps.o unaligned.o unwind.o
 
 obj-$(CONFIG_EFI_VARS)		+= efivars.o
-obj-$(CONFIG_FSYS)		+= fsys.o
 obj-$(CONFIG_IA64_BRL_EMU)	+= brl_emu.o
 obj-$(CONFIG_IA64_GENERIC)	+= acpi-ext.o
 obj-$(CONFIG_IA64_HP_ZX1)	+= acpi-ext.o
@@ -18,3 +17,30 @@ obj-$(CONFIG_IA64_PALINFO)	+= palinfo.o
 obj-$(CONFIG_IOSAPIC)		+= iosapic.o
 obj-$(CONFIG_MODULES)		+= module.o
 obj-$(CONFIG_SMP)		+= smp.o smpboot.o
+obj-$(CONFIG_PERFMON)		+= perfmon_default_smpl.o
+
+# The gate DSO image is built using a special linker script.
+targets += gate.so gate-syms.o
+
+AFLAGS_gate.lds.o += -P -C -U$(ARCH)
+arch/ia64/kernel/gate.lds.s: %.s: %.S scripts FORCE
+	$(call if_changed_dep,as_s_S)
+
+quiet_cmd_gate = GATE $@
+      cmd_gate = $(CC) -nostdlib $(GATECFLAGS_$(@F)) -Wl,-T,$(filter-out FORCE,$^) -o $@
+
+GATECFLAGS_gate.so = -shared -s -Wl,-soname=linux-gate.so.1
+$(obj)/gate.so: $(src)/gate.lds.s $(obj)/gate.o FORCE
+	$(call if_changed,gate)
+
+$(obj)/built-in.o: $(obj)/gate-syms.o
+$(obj)/built-in.o: ld_flags += -R $(obj)/gate-syms.o
+
+GATECFLAGS_gate-syms.o = -r
+$(obj)/gate-syms.o: $(src)/gate.lds.s $(obj)/gate.o FORCE
+	$(call if_changed,gate)
+
+# gate-data.o contains the gate DSO image as data in section .data.gate.
+# We must build gate.so before we can assemble it.
+# Note: kbuild does not track this dependency due to usage of .incbin
+$(obj)/gate-data.o: $(obj)/gate.so

arch/ia64/kernel/acpi-ext.c

@@ -84,7 +84,6 @@ hp_acpi_csr_space(acpi_handle obj, u64 *csr_base, u64 *csr_length)
 	acpi_status status;
 	u8 *data;
 	u32 length;
-	int i;
 
 	status = acpi_find_vendor_resource(obj, &hp_ccsr_descriptor, &data, &length);
 

arch/ia64/kernel/acpi.c

@@ -96,6 +96,9 @@ acpi_get_sysname (void)
 	if (!strcmp(hdr->oem_id, "HP")) {
 		return "hpzx1";
 	}
+	else if (!strcmp(hdr->oem_id, "SGI")) {
+		return "sn2";
+	}
 
 	return "dig";
 #else
@@ -103,8 +106,6 @@ acpi_get_sysname (void)
 	return "hpsim";
 # elif defined (CONFIG_IA64_HP_ZX1)
 	return "hpzx1";
-# elif defined (CONFIG_IA64_SGI_SN1)
-	return "sn1";
 # elif defined (CONFIG_IA64_SGI_SN2)
 	return "sn2";
 # elif defined (CONFIG_IA64_DIG)
@@ -191,21 +192,19 @@ acpi_parse_lsapic (acpi_table_entry_header *header)
 
 	printk(KERN_INFO "CPU %d (0x%04x)", total_cpus, (lsapic->id << 8) | lsapic->eid);
 
-	if (lsapic->flags.enabled) {
-		available_cpus++;
+	if (!lsapic->flags.enabled)
+		printk(" disabled");
+	else if (available_cpus >= NR_CPUS)
+		printk(" ignored (increase NR_CPUS)");
+	else {
 		printk(" enabled");
 #ifdef CONFIG_SMP
-		smp_boot_data.cpu_phys_id[total_cpus] = (lsapic->id << 8) | lsapic->eid;
+		smp_boot_data.cpu_phys_id[available_cpus] = (lsapic->id << 8) | lsapic->eid;
 		if (hard_smp_processor_id()
-		    == (unsigned int) smp_boot_data.cpu_phys_id[total_cpus])
+		    == (unsigned int) smp_boot_data.cpu_phys_id[available_cpus])
 			printk(" (BSP)");
 #endif
-	}
-	else {
-		printk(" disabled");
-#ifdef CONFIG_SMP
-		smp_boot_data.cpu_phys_id[total_cpus] = -1;
-#endif
+		++available_cpus;
 	}
 
 	printk("\n");
@@ -694,11 +693,11 @@ acpi_boot_init (void)
 #endif
 
 #ifdef CONFIG_SMP
+	smp_boot_data.cpu_count = available_cpus;
 	if (available_cpus == 0) {
 		printk(KERN_INFO "ACPI: Found 0 CPUS; assuming 1\n");
 		available_cpus = 1; /* We've got at least one of these, no? */
 	}
-	smp_boot_data.cpu_count = total_cpus;
 
 	smp_build_cpu_map();
 # ifdef CONFIG_NUMA

arch/ia64/kernel/efi_stub.S

@@ -62,7 +62,7 @@ GLOBAL_ENTRY(efi_call_phys)
 	mov b6=r2
 	;;
 	andcm r16=loc3,r16		// get psr with IT, DT, and RT bits cleared
-	br.call.sptk.many rp=ia64_switch_mode
+	br.call.sptk.many rp=ia64_switch_mode_phys
 .ret0:	mov out4=in5
 	mov out0=in1
 	mov out1=in2
@@ -73,7 +73,7 @@ GLOBAL_ENTRY(efi_call_phys)
 	br.call.sptk.many rp=b6		// call the EFI function
 .ret1:	mov ar.rsc=0			// put RSE in enforced lazy, LE mode
 	mov r16=loc3
-	br.call.sptk.many rp=ia64_switch_mode // return to virtual mode
+	br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode
 .ret2:	mov ar.rsc=loc4			// restore RSE configuration
 	mov ar.pfs=loc1
 	mov rp=loc0

arch/ia64/kernel/entry.h

@@ -4,8 +4,9 @@
  * Preserved registers that are shared between code in ivt.S and entry.S.  Be
  * careful not to step on these!
  */
-#define pKStk		p2	/* will leave_kernel return to kernel-stacks? */
-#define pUStk		p3	/* will leave_kernel return to user-stacks? */
+#define pLvSys		p1	/* set 1 if leave from syscall; otherwise, set 0*/
+#define pKStk		p2	/* will leave_{kernel,syscall} return to kernel-stacks? */
+#define pUStk		p3	/* will leave_{kernel,syscall} return to user-stacks? */
 #define pSys		p4	/* are we processing a (synchronous) system call? */
 #define pNonSys		p5	/* complement of pSys */
 
@@ -13,6 +14,7 @@
 #define SW(f)		(IA64_SWITCH_STACK_##f##_OFFSET)
 
 #define PT_REGS_SAVES(off)			\
+	.unwabi 3, 'i';				\
 	.unwabi @svr4, 'i';			\
 	.fframe IA64_PT_REGS_SIZE+16+(off);	\
 	.spillsp rp, PT(CR_IIP)+16+(off);	\

arch/ia64/kernel/gate-data.S

+	.section .data.gate, "ax"
+
+	.incbin "arch/ia64/kernel/gate.so"

arch/ia64/kernel/gate.S

@@ -6,20 +6,47 @@
  * 	David Mosberger-Tang <davidm@hpl.hp.com>
  */
 
+#include <linux/config.h>
+
 #include <asm/asmmacro.h>
+#include <asm/errno.h>
 #include <asm/offsets.h>
 #include <asm/sigcontext.h>
 #include <asm/system.h>
 #include <asm/unistd.h>
-#include <asm/page.h>
-
-	.section .text.gate, "ax"
-.start_gate:
 
+/*
+ * We can't easily refer to symbols inside the kernel.  To avoid full runtime relocation,
+ * complications with the linker (which likes to create PLT stubs for branches
+ * to targets outside the shared object) and to avoid multi-phase kernel builds, we
+ * simply create minimalistic "patch lists" in special ELF sections.
+ */
+	.section ".data.patch.fsyscall_table", "a"
+	.previous
+#define LOAD_FSYSCALL_TABLE(reg)			\
+[1:]	movl reg=0;					\
+	.xdata4 ".data.patch.fsyscall_table", 1b-.
 
-#ifdef CONFIG_FSYS
+	.section ".data.patch.brl_fsys_bubble_down", "a"
+	.previous
+#define BRL_COND_FSYS_BUBBLE_DOWN(pr)			\
+[1:](pr)brl.cond.sptk 0;				\
+	.xdata4 ".data.patch.brl_fsys_bubble_down", 1b-.
 
-#include <asm/errno.h>
+GLOBAL_ENTRY(__kernel_syscall_via_break)
+	.prologue
+	.altrp b6
+	.body
+	/*
+	 * Note: for (fast) syscall restart to work, the break instruction must be
+	 *	 the first one in the bundle addressed by syscall_via_break.
+	 */
+{ .mib
+	break 0x100000
+	nop.i 0
+	br.ret.sptk.many b6
+}
+END(__kernel_syscall_via_break)
 
 /*
  * On entry:
@@ -34,7 +61,8 @@
  *	all other "scratch" registers:	undefined
  *	all "preserved" registers:	same as on entry
  */
-GLOBAL_ENTRY(syscall_via_epc)
+
+GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	.prologue
 	.altrp b6
 	.body
@@ -49,52 +77,50 @@ GLOBAL_ENTRY(syscall_via_epc)
 	epc
 }
 	;;
-	rsm psr.be
-	movl r18=fsyscall_table
+	rsm psr.be // note: on McKinley "rsm psr.be/srlz.d" is slightly faster than "rum psr.be"
+	LOAD_FSYSCALL_TABLE(r14)
 
-	mov r16=IA64_KR(CURRENT)
-	mov r19=255
+	mov r16=IA64_KR(CURRENT)		// 12 cycle read latency
+	mov r19=NR_syscalls-1
 	;;
-	shladd r18=r17,3,r18
-	cmp.geu p6,p0=r19,r17			// (syscall > 0 && syscall <= 1024+255)?
+	shladd r18=r17,3,r14
+
+	srlz.d
+	cmp.ne p8,p0=r0,r0			// p8 <- FALSE
+	/* Note: if r17 is a NaT, p6 will be set to zero.  */
+	cmp.geu p6,p7=r19,r17			// (syscall > 0 && syscall < 1024+NR_syscalls)?
 	;;
-	srlz.d					// ensure little-endian byteorder is in effect
 (p6)	ld8 r18=[r18]
+	mov r29=psr				// read psr (12 cyc load latency)
+	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
 	;;
 (p6)	mov b7=r18
+(p6)	tbit.z p8,p0=r18,0
+(p8)	br.dptk.many b7
+
+	mov r27=ar.rsc
+	mov r21=ar.fpsr
+	mov r26=ar.pfs
+/*
+ * brl.cond doesn't work as intended because the linker would convert this branch
+ * into a branch to a PLT.  Perhaps there will be a way to avoid this with some
+ * future version of the linker.  In the meantime, we just use an indirect branch
+ * instead.
+ */
+#ifdef CONFIG_ITANIUM
+(p6)	ld8 r14=[r14]				// r14 <- fsys_bubble_down
+	;;
+(p6)	mov b7=r14
 (p6)	br.sptk.many b7
+#else
+	BRL_COND_FSYS_BUBBLE_DOWN(p6)
+#endif
 
 	mov r10=-1
 	mov r8=ENOSYS
 	MCKINLEY_E9_WORKAROUND
 	br.ret.sptk.many b6
-END(syscall_via_epc)
-
-GLOBAL_ENTRY(syscall_via_break)
-	.prologue
-	.altrp b6
-	.body
-	break 0x100000
-	br.ret.sptk.many b6
-END(syscall_via_break)
-
-GLOBAL_ENTRY(fsys_fallback_syscall)
-	/*
-	 * It would be better/fsyser to do the SAVE_MIN magic directly here, but for now
-	 * we simply fall back on doing a system-call via break.  Good enough
-	 * to get started.  (Note: we have to do this through the gate page again, since
-	 * the br.ret will switch us back to user-level privilege.)
-	 *
-	 * XXX Move this back to fsys.S after changing it over to avoid break 0x100000.
-	 */
-	movl r2=(syscall_via_break - .start_gate) + GATE_ADDR
-	;;
-	MCKINLEY_E9_WORKAROUND
-	mov b7=r2
-	br.ret.sptk.many b7
-END(fsys_fallback_syscall)
-
-#endif /* CONFIG_FSYS */
+END(__kernel_syscall_via_epc)
 
 #	define ARG0_OFF		(16 + IA64_SIGFRAME_ARG0_OFFSET)
 #	define ARG1_OFF		(16 + IA64_SIGFRAME_ARG1_OFFSET)
@@ -145,7 +171,8 @@ END(fsys_fallback_syscall)
 	 */
 
 #define SIGTRAMP_SAVES										\
-	.unwabi @svr4, 's';	/* mark this as a sigtramp handler (saves scratch regs) */	\
+	.unwabi 3, 's';		/* mark this as a sigtramp handler (saves scratch regs) */	\
+	.unwabi @svr4, 's'; /* backwards compatibility with old unwinders (remove in v2.7) */	\
 	.savesp ar.unat, UNAT_OFF+SIGCONTEXT_OFF;						\
 	.savesp ar.fpsr, FPSR_OFF+SIGCONTEXT_OFF;						\
 	.savesp pr, PR_OFF+SIGCONTEXT_OFF;     							\
@@ -153,7 +180,7 @@ END(fsys_fallback_syscall)
 	.savesp ar.pfs, CFM_OFF+SIGCONTEXT_OFF;							\
 	.vframesp SP_OFF+SIGCONTEXT_OFF
 
-GLOBAL_ENTRY(ia64_sigtramp)
+GLOBAL_ENTRY(__kernel_sigtramp)
 	// describe the state that is active when we get here:
 	.prologue
 	SIGTRAMP_SAVES
@@ -335,4 +362,4 @@ restore_rbs:
 	mov ar.rsc=0xf				// (will be restored later on from sc_ar_rsc)
 	// invala not necessary as that will happen when returning to user-mode
 	br.cond.sptk back_from_restore_rbs
-END(ia64_sigtramp)
+END(__kernel_sigtramp)

arch/ia64/kernel/gate.lds.S

+/*
+ * Linker script for gate DSO.  The gate pages are an ELF shared object prelinked to its
+ * virtual address, with only one read-only segment and one execute-only segment (both fit
+ * in one page).  This script controls its layout.
+ */
+
+#include <linux/config.h>
+
+#include <asm/system.h>
+
+SECTIONS
+{
+  . = GATE_ADDR + SIZEOF_HEADERS;
+
+  .hash				: { *(.hash) }				:readable
+  .dynsym			: { *(.dynsym) }
+  .dynstr			: { *(.dynstr) }
+  .gnu.version			: { *(.gnu.version) }
+  .gnu.version_d		: { *(.gnu.version_d) }
+  .gnu.version_r		: { *(.gnu.version_r) }
+  .dynamic			: { *(.dynamic) }			:readable :dynamic
+
+  /*
+   * This linker script is used both with -r and with -shared.  For the layouts to match,
+   * we need to skip more than enough space for the dynamic symbol table et al.  If this
+   * amount is insufficient, ld -shared will barf.  Just increase it here.
+   */
+  . = GATE_ADDR + 0x500;
+
+  .data.patch			: {
+				    __start_gate_mckinley_e9_patchlist = .;
+				    *(.data.patch.mckinley_e9)
+				    __end_gate_mckinley_e9_patchlist = .;
+
+				    __start_gate_vtop_patchlist = .;
+				    *(.data.patch.vtop)
+				    __end_gate_vtop_patchlist = .;
+
+				    __start_gate_fsyscall_patchlist = .;
+				    *(.data.patch.fsyscall_table)
+				    __end_gate_fsyscall_patchlist = .;
+
+				    __start_gate_brl_fsys_bubble_down_patchlist = .;
+				    *(.data.patch.brl_fsys_bubble_down)
+				    __end_gate_brl_fsys_bubble_down_patchlist = .;
+  }									:readable
+  .IA_64.unwind_info		: { *(.IA_64.unwind_info*) }
+  .IA_64.unwind			: { *(.IA_64.unwind*) }			:readable :unwind
+#ifdef HAVE_BUGGY_SEGREL
+  .text (GATE_ADDR + PAGE_SIZE)	: { *(.text) *(.text.*) }		:readable
+#else
+  . = ALIGN (PERCPU_PAGE_SIZE) + (. & (PERCPU_PAGE_SIZE - 1));
+  .text				: { *(.text) *(.text.*) }		:epc
+#endif
+
+  /DISCARD/			: {
+  	*(.got.plt) *(.got)
+	*(.data .data.* .gnu.linkonce.d.*)
+	*(.dynbss)
+	*(.bss .bss.* .gnu.linkonce.b.*)
+	*(__ex_table)
+  }
+}
+
+/*
+ * We must supply the ELF program headers explicitly to get just one
+ * PT_LOAD segment, and set the flags explicitly to make segments read-only.
+ */
+PHDRS
+{
+  readable  PT_LOAD	FILEHDR	PHDRS	FLAGS(4);	/* PF_R */
+#ifndef HAVE_BUGGY_SEGREL
+  epc	    PT_LOAD	FILEHDR PHDRS	FLAGS(1);	/* PF_X */
+#endif
+  dynamic   PT_DYNAMIC			FLAGS(4);	/* PF_R */
+  unwind    0x70000001; /* PT_IA_64_UNWIND, but ld doesn't match the name */
+}
+
+/*
+ * This controls what symbols we export from the DSO.
+ */
+VERSION
+{
+  LINUX_2.5 {
+    global:
+	__kernel_syscall_via_break;
+	__kernel_syscall_via_epc;
+	__kernel_sigtramp;
+
+    local: *;
+  };
+}
+
+/* The ELF entry point can be used to set the AT_SYSINFO value.  */
+ENTRY(__kernel_syscall_via_epc)

arch/ia64/kernel/head.S

@@ -60,22 +60,42 @@ start_ap:
 	mov r4=r0
 	.body
 
-	/*
-	 * Initialize the region register for region 7 and install a translation register
-	 * that maps the kernel's text and data:
-	 */
 	rsm psr.i | psr.ic
-	mov r16=((ia64_rid(IA64_REGION_ID_KERNEL, PAGE_OFFSET) << 8) | (IA64_GRANULE_SHIFT << 2))
 	;;
 	srlz.i
+	;;
+	/*
+	 * Initialize kernel region registers:
+	 *	rr[5]: VHPT enabled, page size = PAGE_SHIFT
+	 *	rr[6]: VHPT disabled, page size = IA64_GRANULE_SHIFT
+	 *	rr[5]: VHPT disabled, page size = IA64_GRANULE_SHIFT
+	 */
+	mov r16=((ia64_rid(IA64_REGION_ID_KERNEL, (5<<61)) << 8) | (PAGE_SHIFT << 2) | 1)
+	movl r17=(5<<61)
+	mov r18=((ia64_rid(IA64_REGION_ID_KERNEL, (6<<61)) << 8) | (IA64_GRANULE_SHIFT << 2))
+	movl r19=(6<<61)
+	mov r20=((ia64_rid(IA64_REGION_ID_KERNEL, (7<<61)) << 8) | (IA64_GRANULE_SHIFT << 2))
+	movl r21=(7<<61)
+	;;
+	mov rr[r17]=r16
+	mov rr[r19]=r18
+	mov rr[r21]=r20
+	;;
+	/*
+	 * Now pin mappings into the TLB for kernel text and data
+	 */
 	mov r18=KERNEL_TR_PAGE_SHIFT<<2
 	movl r17=KERNEL_START
 	;;
-	mov rr[r17]=r16
 	mov cr.itir=r18
 	mov cr.ifa=r17
 	mov r16=IA64_TR_KERNEL
-	movl r18=((1 << KERNEL_TR_PAGE_SHIFT) | PAGE_KERNEL)
+	mov r3=ip
+	movl r18=PAGE_KERNEL
+	;;
+	dep r2=0,r3,0,KERNEL_TR_PAGE_SHIFT
+	;;
+	or r18=r2,r18
 	;;
 	srlz.i
 	;;
@@ -113,16 +133,6 @@ start_ap:
 	mov ar.fpsr=r2
 	;;
 
-#ifdef CONFIG_IA64_EARLY_PRINTK
-	mov r3=(6<<8) | (IA64_GRANULE_SHIFT<<2)
-	movl r2=6<<61
-	;;
-	mov rr[r2]=r3
-	;;
-	srlz.i
-	;;
-#endif
-
 #define isAP	p2	// are we an Application Processor?
 #define isBP	p3	// are we the Bootstrap Processor?
 
@@ -143,12 +153,36 @@ start_ap:
 	movl r2=init_thread_union
 	cmp.eq isBP,isAP=r0,r0
 #endif
-	mov r16=KERNEL_TR_PAGE_NUM
 	;;
+	tpa r3=r2		// r3 == phys addr of task struct
+	// load mapping for stack (virtaddr in r2, physaddr in r3)
+	rsm psr.ic
+	movl r17=PAGE_KERNEL
+	;;
+	srlz.d
+	dep r18=0,r3,0,12
+	;;
+	or r18=r17,r18
+	dep r2=-1,r3,61,3	// IMVA of task
+	;;
+	mov r17=rr[r2]
+	shr.u r16=r3,IA64_GRANULE_SHIFT
+	;;
+	dep r17=0,r17,8,24
+	;;
+	mov cr.itir=r17
+	mov cr.ifa=r2
+
+	mov r19=IA64_TR_CURRENT_STACK
+	;;
+	itr.d dtr[r19]=r18
+	;;
+	ssm psr.ic
+	srlz.d
+  	;;
 
 	// load the "current" pointer (r13) and ar.k6 with the current task
 	mov IA64_KR(CURRENT)=r2		// virtual address
-	// initialize k4 to a safe value (64-128MB is mapped by TR_KERNEL)
 	mov IA64_KR(CURRENT_STACK)=r16
 	mov r13=r2
 	/*
@@ -665,14 +699,14 @@ GLOBAL_ENTRY(__ia64_init_fpu)
 END(__ia64_init_fpu)
 
 /*
- * Switch execution mode from virtual to physical or vice versa.
+ * Switch execution mode from virtual to physical
  *
  * Inputs:
  *	r16 = new psr to establish
  *
  * Note: RSE must already be in enforced lazy mode
  */
-GLOBAL_ENTRY(ia64_switch_mode)
+GLOBAL_ENTRY(ia64_switch_mode_phys)
  {
 	alloc r2=ar.pfs,0,0,0,0
 	rsm psr.i | psr.ic		// disable interrupts and interrupt collection
@@ -682,35 +716,86 @@ GLOBAL_ENTRY(ia64_switch_mode)
  {
 	flushrs				// must be first insn in group
 	srlz.i
-	shr.u r19=r15,61		// r19 <- top 3 bits of current IP
  }
 	;;
 	mov cr.ipsr=r16			// set new PSR
-	add r3=1f-ia64_switch_mode,r15
-	xor r15=0x7,r19			// flip the region bits
+	add r3=1f-ia64_switch_mode_phys,r15
 
 	mov r17=ar.bsp
 	mov r14=rp			// get return address into a general register
+	;;
 
-	// switch RSE backing store:
+	// going to physical mode, use tpa to translate virt->phys
+	tpa r17=r17
+	tpa r3=r3
+	tpa sp=sp
+	tpa r14=r14
 	;;
-	dep r17=r15,r17,61,3		// make ar.bsp physical or virtual
+
 	mov r18=ar.rnat			// save ar.rnat
-	;;
 	mov ar.bspstore=r17		// this steps on ar.rnat
-	dep r3=r15,r3,61,3		// make rfi return address physical or virtual
+	mov cr.iip=r3
+	mov cr.ifs=r0
 	;;
+	mov ar.rnat=r18			// restore ar.rnat
+	rfi				// must be last insn in group
+	;;
+1:	mov rp=r14
+	br.ret.sptk.many rp
+END(ia64_switch_mode_phys)
+
+/*
+ * Switch execution mode from physical to virtual
+ *
+ * Inputs:
+ *	r16 = new psr to establish
+ *
+ * Note: RSE must already be in enforced lazy mode
+ */
+GLOBAL_ENTRY(ia64_switch_mode_virt)
+ {
+	alloc r2=ar.pfs,0,0,0,0
+	rsm psr.i | psr.ic		// disable interrupts and interrupt collection
+	mov r15=ip
+ }
+	;;
+ {
+	flushrs				// must be first insn in group
+	srlz.i
+ }
+	;;
+	mov cr.ipsr=r16			// set new PSR
+	add r3=1f-ia64_switch_mode_virt,r15
+
+	mov r17=ar.bsp
+	mov r14=rp			// get return address into a general register
+	;;
+
+	// going to virtual
+	//   - for code addresses, set upper bits of addr to KERNEL_START
+	//   - for stack addresses, set upper 3 bits to 0xe.... Dont change any of the
+	//     lower bits since we want it to stay identity mapped
+	movl r18=KERNEL_START
+	dep r3=0,r3,KERNEL_TR_PAGE_SHIFT,64-KERNEL_TR_PAGE_SHIFT
+	dep r14=0,r14,KERNEL_TR_PAGE_SHIFT,64-KERNEL_TR_PAGE_SHIFT
+	dep r17=-1,r17,61,3
+	dep sp=-1,sp,61,3
+	;;
+	or r3=r3,r18
+	or r14=r14,r18
+	;;
+
+	mov r18=ar.rnat			// save ar.rnat
+	mov ar.bspstore=r17		// this steps on ar.rnat
 	mov cr.iip=r3
 	mov cr.ifs=r0
-	dep sp=r15,sp,61,3		// make stack pointer physical or virtual
 	;;
 	mov ar.rnat=r18			// restore ar.rnat
-	dep r14=r15,r14,61,3		// make function return address physical or virtual
 	rfi				// must be last insn in group
 	;;
 1:	mov rp=r14
 	br.ret.sptk.many rp
-END(ia64_switch_mode)
+END(ia64_switch_mode_virt)
 
 #ifdef CONFIG_IA64_BRL_EMU
 
@@ -753,7 +838,7 @@ SET_REG(b5);
 	 *   r29    - available for use.
 	 *   r30    - available for use.
 	 *   r31    - address of lock, available for use.
-	 *   b7     - return address
+	 *   b6     - return address
 	 *   p14    - available for use.
 	 *
 	 * If you patch this code to use more registers, do not forget to update

arch/ia64/kernel/ia64_ksyms.c

@@ -65,6 +65,9 @@ EXPORT_SYMBOL(ia64_pfn_valid);
 
 #include <asm/processor.h>
 EXPORT_SYMBOL(cpu_info__per_cpu);
+#ifdef CONFIG_SMP
+EXPORT_SYMBOL(__per_cpu_offset);
+#endif
 EXPORT_SYMBOL(kernel_thread);
 
 #include <asm/system.h>
@@ -88,6 +91,7 @@ EXPORT_SYMBOL(synchronize_irq);
 EXPORT_SYMBOL(smp_call_function);
 EXPORT_SYMBOL(smp_call_function_single);
 EXPORT_SYMBOL(cpu_online_map);
+EXPORT_SYMBOL(phys_cpu_present_map);
 EXPORT_SYMBOL(ia64_cpu_to_sapicid);
 #else /* !CONFIG_SMP */
 
@@ -124,6 +128,18 @@ EXPORT_SYMBOL_NOVERS(__udivdi3);
 EXPORT_SYMBOL_NOVERS(__moddi3);
 EXPORT_SYMBOL_NOVERS(__umoddi3);
 
+#if defined(CONFIG_MD_RAID5) || defined(CONFIG_MD_RAID5_MODULE)
+extern void xor_ia64_2(void);
+extern void xor_ia64_3(void);
+extern void xor_ia64_4(void);
+extern void xor_ia64_5(void);
+
+EXPORT_SYMBOL_NOVERS(xor_ia64_2);
+EXPORT_SYMBOL_NOVERS(xor_ia64_3);
+EXPORT_SYMBOL_NOVERS(xor_ia64_4);
+EXPORT_SYMBOL_NOVERS(xor_ia64_5);
+#endif
+
 extern unsigned long ia64_iobase;
 EXPORT_SYMBOL(ia64_iobase);
 
@@ -147,10 +163,15 @@ EXPORT_SYMBOL(efi_dir);
 EXPORT_SYMBOL(ia64_mv);
 #endif
 EXPORT_SYMBOL(machvec_noop);
+EXPORT_SYMBOL(machvec_memory_fence);
+EXPORT_SYMBOL(zero_page_memmap_ptr);
 #ifdef CONFIG_PERFMON
 #include <asm/perfmon.h>
-EXPORT_SYMBOL(pfm_install_alternate_syswide_subsystem);
-EXPORT_SYMBOL(pfm_remove_alternate_syswide_subsystem);
+EXPORT_SYMBOL(pfm_register_buffer_fmt);
+EXPORT_SYMBOL(pfm_unregister_buffer_fmt);
+EXPORT_SYMBOL(pfm_mod_fast_read_pmds);
+EXPORT_SYMBOL(pfm_mod_read_pmds);
+EXPORT_SYMBOL(pfm_mod_write_pmcs);
 #endif
 
 #ifdef CONFIG_NUMA
@@ -169,10 +190,25 @@ EXPORT_SYMBOL(unw_access_fr);
 EXPORT_SYMBOL(unw_access_ar);
 EXPORT_SYMBOL(unw_access_pr);
 
-#if __GNUC__ < 3 || (__GNUC__ == 3 && __GNUC_MINOR__ < 4)
-extern void ia64_spinlock_contention_pre3_4 (void);
+#ifdef CONFIG_SMP
+# if __GNUC__ < 3 || (__GNUC__ == 3 && __GNUC_MINOR__ < 4)
+/*
+ * This is not a normal routine and we don't want a function descriptor for it, so we use
+ * a fake declaration here.
+ */
+extern char ia64_spinlock_contention_pre3_4;
 EXPORT_SYMBOL(ia64_spinlock_contention_pre3_4);
-#else
-extern void ia64_spinlock_contention (void);
+# else
+/*
+ * This is not a normal routine and we don't want a function descriptor for it, so we use
+ * a fake declaration here.
+ */
+extern char ia64_spinlock_contention;
 EXPORT_SYMBOL(ia64_spinlock_contention);
+# endif
 #endif
+
+EXPORT_SYMBOL(ia64_max_iommu_merge_mask);
+
+#include <linux/pm.h>
+EXPORT_SYMBOL(pm_idle);

arch/ia64/kernel/init_task.c

@@ -36,7 +36,7 @@ union init_thread {
 	unsigned long stack[KERNEL_STACK_SIZE/sizeof (unsigned long)];
 } init_thread_union __attribute__((section(".data.init_task"))) = {{
 	.task =		INIT_TASK(init_thread_union.s.task),
-	.thread_info =	INIT_THREAD_INFO(init_thread_union.s.thread_info)
+	.thread_info =	INIT_THREAD_INFO(init_thread_union.s.task)
 }};
 
 asm (".global init_task; init_task = init_thread_union");

arch/ia64/kernel/irq.c

@@ -65,7 +65,7 @@
 /*
  * Controller mappings for all interrupt sources:
  */
-irq_desc_t irq_desc[NR_IRQS] __cacheline_aligned = {
+irq_desc_t _irq_desc[NR_IRQS] __cacheline_aligned = {
 	[0 ... NR_IRQS-1] = {
 		.status = IRQ_DISABLED,
 		.handler = &no_irq_type,
@@ -235,7 +235,6 @@ int handle_IRQ_event(unsigned int irq,
 {
 	int status = 1;	/* Force the "do bottom halves" bit */
 	int retval = 0;
-	struct irqaction *first_action = action;
 
 	if (!(action->flags & SA_INTERRUPT))
 		local_irq_enable();
@@ -248,30 +247,88 @@ int handle_IRQ_event(unsigned int irq,
 	if (status & SA_SAMPLE_RANDOM)
 		add_interrupt_randomness(irq);
 	local_irq_disable();
-	if (retval != 1) {
-		static int count = 100;
-		if (count) {
-			count--;
-			if (retval) {
-				printk("irq event %d: bogus retval mask %x\n",
-					irq, retval);
-			} else {
-				printk("irq %d: nobody cared!\n", irq);
-			}
-			dump_stack();
-			printk("handlers:\n");
-			action = first_action;
-			do {
-				printk("[<%p>]", action->handler);
-				print_symbol(" (%s)",
-					(unsigned long)action->handler);
-				printk("\n");
-				action = action->next;
-			} while (action);
-		}
+	return retval;
+}
+
+static void __report_bad_irq(int irq, irq_desc_t *desc, irqreturn_t action_ret)
+{
+	struct irqaction *action;
+
+	if (action_ret != IRQ_HANDLED && action_ret != IRQ_NONE) {
+		printk(KERN_ERR "irq event %d: bogus return value %x\n",
+				irq, action_ret);
+	} else {
+		printk(KERN_ERR "irq %d: nobody cared!\n", irq);
+	}
+	dump_stack();
+	printk(KERN_ERR "handlers:\n");
+	action = desc->action;
+	do {
+		printk(KERN_ERR "[<%p>]", action->handler);
+		print_symbol(" (%s)",
+			(unsigned long)action->handler);
+		printk("\n");
+		action = action->next;
+	} while (action);
+}
+
+static void report_bad_irq(int irq, irq_desc_t *desc, irqreturn_t action_ret)
+{
+	static int count = 100;
+
+	if (count) {
+		count--;
+		__report_bad_irq(irq, desc, action_ret);
+	}
+}
+
+static int noirqdebug;
+
+static int __init noirqdebug_setup(char *str)
+{
+	noirqdebug = 1;
+	printk("IRQ lockup detection disabled\n");
+	return 1;
+}
+
+__setup("noirqdebug", noirqdebug_setup);
+
+/*
+ * If 99,900 of the previous 100,000 interrupts have not been handled then
+ * assume that the IRQ is stuck in some manner.  Drop a diagnostic and try to
+ * turn the IRQ off.
+ *
+ * (The other 100-of-100,000 interrupts may have been a correctly-functioning
+ *  device sharing an IRQ with the failing one)
+ *
+ * Called under desc->lock
+ */
+static void note_interrupt(int irq, irq_desc_t *desc, irqreturn_t action_ret)
+{
+	if (action_ret != IRQ_HANDLED) {
+		desc->irqs_unhandled++;
+		if (action_ret != IRQ_NONE)
+			report_bad_irq(irq, desc, action_ret);
 	}
 
-	return status;
+	desc->irq_count++;
+	if (desc->irq_count < 100000)
+		return;
+
+	desc->irq_count = 0;
+	if (desc->irqs_unhandled > 99900) {
+		/*
+		 * The interrupt is stuck
+		 */
+		__report_bad_irq(irq, desc, action_ret);
+		/*
+		 * Now kill the IRQ
+		 */
+		printk(KERN_EMERG "Disabling IRQ #%d\n", irq);
+		desc->status |= IRQ_DISABLED;
+		desc->handler->disable(irq);
+	}
+	desc->irqs_unhandled = 0;
 }
 
 /*
@@ -380,21 +437,24 @@ unsigned int do_IRQ(unsigned long irq, struct pt_regs *regs)
 	 * 0 return value means that this irq is already being
 	 * handled by some other CPU. (or is disabled)
 	 */
-	int cpu;
 	irq_desc_t *desc = irq_desc(irq);
 	struct irqaction * action;
+	irqreturn_t action_ret;
 	unsigned int status;
+	int cpu;
 
 	irq_enter();
-	cpu = smp_processor_id();
+	cpu = smp_processor_id(); /* for CONFIG_PREEMPT, this must come after irq_enter()! */
 
 	kstat_cpu(cpu).irqs[irq]++;
 
 	if (desc->status & IRQ_PER_CPU) {
 		/* no locking required for CPU-local interrupts: */
 		desc->handler->ack(irq);
-		handle_IRQ_event(irq, regs, desc->action);
+		action_ret = handle_IRQ_event(irq, regs, desc->action);
 		desc->handler->end(irq);
+		if (!noirqdebug)
+			note_interrupt(irq, desc, action_ret);
 	} else {
 		spin_lock(&desc->lock);
 		desc->handler->ack(irq);
@@ -438,9 +498,10 @@ unsigned int do_IRQ(unsigned long irq, struct pt_regs *regs)
 		 */
 		for (;;) {
 			spin_unlock(&desc->lock);
-			handle_IRQ_event(irq, regs, action);
+			action_ret = handle_IRQ_event(irq, regs, action);
 			spin_lock(&desc->lock);
-
+			if (!noirqdebug)
+				note_interrupt(irq, desc, action_ret);
 			if (!(desc->status & IRQ_PENDING))
 				break;
 			desc->status &= ~IRQ_PENDING;

arch/ia64/kernel/mca.c

@@ -322,7 +322,7 @@ fetch_min_state (pal_min_state_area_t *ms, struct pt_regs *pt, struct switch_sta
 }
 
 void
-init_handler_platform (sal_log_processor_info_t *proc_ptr,
+init_handler_platform (pal_min_state_area_t *ms,
 		       struct pt_regs *pt, struct switch_stack *sw)
 {
 	struct unw_frame_info info;
@@ -337,15 +337,18 @@ init_handler_platform (sal_log_processor_info_t *proc_ptr,
 	 */
 	printk("Delaying for 5 seconds...\n");
 	udelay(5*1000000);
-	show_min_state(&SAL_LPI_PSI_INFO(proc_ptr)->min_state_area);
+	show_min_state(ms);
 
 	printk("Backtrace of current task (pid %d, %s)\n", current->pid, current->comm);
-	fetch_min_state(&SAL_LPI_PSI_INFO(proc_ptr)->min_state_area, pt, sw);
+	fetch_min_state(ms, pt, sw);
 	unw_init_from_interruption(&info, current, pt, sw);
 	ia64_do_show_stack(&info, NULL);
 
+#ifdef CONFIG_SMP
+	/* read_trylock() would be handy... */
 	if (!tasklist_lock.write_lock)
 		read_lock(&tasklist_lock);
+#endif
 	{
 		struct task_struct *g, *t;
 		do_each_thread (g, t) {
@@ -353,11 +356,13 @@ init_handler_platform (sal_log_processor_info_t *proc_ptr,
 				continue;
 
 			printk("\nBacktrace of pid %d (%s)\n", t->pid, t->comm);
-			show_stack(t);
+			show_stack(t, NULL);
 		} while_each_thread (g, t);
 	}
+#ifdef CONFIG_SMP
 	if (!tasklist_lock.write_lock)
 		read_unlock(&tasklist_lock);
+#endif
 
 	printk("\nINIT dump complete.  Please reboot now.\n");
 	while (1);			/* hang city if no debugger */
@@ -657,17 +662,17 @@ ia64_mca_init(void)
 
 	IA64_MCA_DEBUG("ia64_mca_init: registered mca rendezvous spinloop and wakeup mech.\n");
 
-	ia64_mc_info.imi_mca_handler        = __pa(mca_hldlr_ptr->fp);
+	ia64_mc_info.imi_mca_handler        = ia64_tpa(mca_hldlr_ptr->fp);
 	/*
 	 * XXX - disable SAL checksum by setting size to 0; should be
-	 *	__pa(ia64_os_mca_dispatch_end) - __pa(ia64_os_mca_dispatch);
+	 *	ia64_tpa(ia64_os_mca_dispatch_end) - ia64_tpa(ia64_os_mca_dispatch);
 	 */
 	ia64_mc_info.imi_mca_handler_size	= 0;
 
 	/* Register the os mca handler with SAL */
 	if ((rc = ia64_sal_set_vectors(SAL_VECTOR_OS_MCA,
 				       ia64_mc_info.imi_mca_handler,
-				       mca_hldlr_ptr->gp,
+				       ia64_tpa(mca_hldlr_ptr->gp),
 				       ia64_mc_info.imi_mca_handler_size,
 				       0, 0, 0)))
 	{
@@ -677,15 +682,15 @@ ia64_mca_init(void)
 	}
 
 	IA64_MCA_DEBUG("ia64_mca_init: registered os mca handler with SAL at 0x%lx, gp = 0x%lx\n",
-		       ia64_mc_info.imi_mca_handler, mca_hldlr_ptr->gp);
+		       ia64_mc_info.imi_mca_handler, ia64_tpa(mca_hldlr_ptr->gp));
 
 	/*
 	 * XXX - disable SAL checksum by setting size to 0, should be
 	 * IA64_INIT_HANDLER_SIZE
 	 */
-	ia64_mc_info.imi_monarch_init_handler		= __pa(mon_init_ptr->fp);
+	ia64_mc_info.imi_monarch_init_handler		= ia64_tpa(mon_init_ptr->fp);
 	ia64_mc_info.imi_monarch_init_handler_size	= 0;
-	ia64_mc_info.imi_slave_init_handler		= __pa(slave_init_ptr->fp);
+	ia64_mc_info.imi_slave_init_handler		= ia64_tpa(slave_init_ptr->fp);
 	ia64_mc_info.imi_slave_init_handler_size	= 0;
 
 	IA64_MCA_DEBUG("ia64_mca_init: os init handler at %lx\n",
@@ -694,10 +699,10 @@ ia64_mca_init(void)
 	/* Register the os init handler with SAL */
 	if ((rc = ia64_sal_set_vectors(SAL_VECTOR_OS_INIT,
 				       ia64_mc_info.imi_monarch_init_handler,
-				       __pa(ia64_get_gp()),
+				       ia64_tpa(ia64_get_gp()),
 				       ia64_mc_info.imi_monarch_init_handler_size,
 				       ia64_mc_info.imi_slave_init_handler,
-				       __pa(ia64_get_gp()),
+				       ia64_tpa(ia64_get_gp()),
 				       ia64_mc_info.imi_slave_init_handler_size)))
 	{
 		printk(KERN_ERR "ia64_mca_init: Failed to register m/s init handlers with SAL. "
@@ -1235,32 +1240,19 @@ device_initcall(ia64_mca_late_init);
 void
 ia64_init_handler (struct pt_regs *pt, struct switch_stack *sw)
 {
-	sal_log_processor_info_t *proc_ptr;
-	ia64_err_rec_t *plog_ptr;
+	pal_min_state_area_t *ms;
 
-	printk(KERN_INFO "Entered OS INIT handler\n");
-
-	/* Get the INIT processor log */
-	if (!ia64_log_get(SAL_INFO_TYPE_INIT, (prfunc_t)printk))
-		return;                 // no record retrieved
-
-#ifdef IA64_DUMP_ALL_PROC_INFO
-	ia64_log_print(SAL_INFO_TYPE_INIT, (prfunc_t)printk);
-#endif
+	printk(KERN_INFO "Entered OS INIT handler. PSP=%lx\n",
+		ia64_sal_to_os_handoff_state.proc_state_param);
 
 	/*
-	 * get pointer to min state save area
-	 *
+	 * Address of minstate area provided by PAL is physical,
+	 * uncacheable (bit 63 set). Convert to Linux virtual
+	 * address in region 6.
 	 */
-	plog_ptr=(ia64_err_rec_t *)IA64_LOG_CURR_BUFFER(SAL_INFO_TYPE_INIT);
-	proc_ptr = &plog_ptr->proc_err;
-
-	ia64_process_min_state_save(&SAL_LPI_PSI_INFO(proc_ptr)->min_state_area);
-
-	/* Clear the INIT SAL logs now that they have been saved in the OS buffer */
-	ia64_sal_clear_state_info(SAL_INFO_TYPE_INIT);
+	ms = (pal_min_state_area_t *)(ia64_sal_to_os_handoff_state.pal_min_state | (6ul<<61));
 
-	init_handler_platform(proc_ptr, pt, sw);	/* call platform specific routines */
+	init_handler_platform(ms, pt, sw);	/* call platform specific routines */
 }
 
 /*

arch/ia64/kernel/mca_asm.S

@@ -50,14 +50,15 @@
  *		6. GR12 = Return address to location within SAL_CHECK
  */
 #define SAL_TO_OS_MCA_HANDOFF_STATE_SAVE(_tmp)		\
-	movl	_tmp=ia64_sal_to_os_handoff_state;;	\
-	DATA_VA_TO_PA(_tmp);;				\
+	LOAD_PHYSICAL(p0, _tmp, ia64_sal_to_os_handoff_state);; \
 	st8	[_tmp]=r1,0x08;;			\
 	st8	[_tmp]=r8,0x08;;			\
 	st8	[_tmp]=r9,0x08;;			\
 	st8	[_tmp]=r10,0x08;;			\
 	st8	[_tmp]=r11,0x08;;			\
-	st8	[_tmp]=r12,0x08
+	st8	[_tmp]=r12,0x08;;			\
+	st8	[_tmp]=r17,0x08;;			\
+	st8	[_tmp]=r18,0x08
 
 /*
  * OS_MCA_TO_SAL_HANDOFF_STATE (SAL 3.0 spec)
@@ -70,9 +71,8 @@
  *	returns ptr to SAL rtn save loc in _tmp
  */
 #define OS_MCA_TO_SAL_HANDOFF_STATE_RESTORE(_tmp)	\
-(p6)	movl	_tmp=ia64_sal_to_os_handoff_state;;	\
-(p7)	movl	_tmp=ia64_os_to_sal_handoff_state;;	\
-	DATA_VA_TO_PA(_tmp);;				\
+	LOAD_PHYSICAL(p6, _tmp, ia64_sal_to_os_handoff_state);; \
+	LOAD_PHYSICAL(p7, _tmp, ia64_os_to_sal_handoff_state);; \
 (p6)	movl	r8=IA64_MCA_COLD_BOOT;			\
 (p6)	movl	r10=IA64_MCA_SAME_CONTEXT;		\
 (p6)	add     _tmp=0x18,_tmp;;			\

arch/ia64/kernel/module.c

@@ -18,7 +18,8 @@
    LTOFF22X
    LTOFF22X
    LTOFF_FPTR22
-   PCREL21B
+   PCREL21B	(for br.call only; br.cond is not supported out of modules!)
+   PCREL60B	(for brl.cond only; brl.call is not supported for modules!)
    PCREL64LSB
    SECREL32LSB
    SEGREL64LSB
@@ -33,6 +34,7 @@
 #include <linux/string.h>
 #include <linux/vmalloc.h>
 
+#include <asm/patch.h>
 #include <asm/unaligned.h>
 
 #define ARCH_MODULE_DEBUG 0
@@ -158,27 +160,6 @@ slot (const struct insn *insn)
 	return (uint64_t) insn & 0x3;
 }
 
-/* Patch instruction with "val" where "mask" has 1 bits. */
-static void
-apply (struct insn *insn, uint64_t mask, uint64_t val)
-{
-	uint64_t m0, m1, v0, v1, b0, b1, *b = (uint64_t *) bundle(insn);
-#	define insn_mask ((1UL << 41) - 1)
-	unsigned long shift;
-
-	b0 = b[0]; b1 = b[1];
-	shift = 5 + 41 * slot(insn); /* 5 bits of template, then 3 x 41-bit instructions */
-	if (shift >= 64) {
-		m1 = mask << (shift - 64);
-		v1 = val << (shift - 64);
-	} else {
-		m0 = mask << shift; m1 = mask >> (64 - shift);
-		v0 = val  << shift; v1 = val >> (64 - shift);
-		b[0] = (b0 & ~m0) | (v0 & m0);
-	}
-	b[1] = (b1 & ~m1) | (v1 & m1);
-}
-
 static int
 apply_imm64 (struct module *mod, struct insn *insn, uint64_t val)
 {
@@ -187,12 +168,7 @@ apply_imm64 (struct module *mod, struct insn *insn, uint64_t val)
 		       mod->name, slot(insn));
 		return 0;
 	}
-	apply(insn, 0x01fffefe000, (  ((val & 0x8000000000000000) >> 27) /* bit 63 -> 36 */
-				    | ((val & 0x0000000000200000) <<  0) /* bit 21 -> 21 */
-				    | ((val & 0x00000000001f0000) <<  6) /* bit 16 -> 22 */
-				    | ((val & 0x000000000000ff80) << 20) /* bit  7 -> 27 */
-				    | ((val & 0x000000000000007f) << 13) /* bit  0 -> 13 */));
-	apply((void *) insn - 1, 0x1ffffffffff, val >> 22);
+	ia64_patch_imm64((u64) insn, val);
 	return 1;
 }
 
@@ -208,9 +184,7 @@ apply_imm60 (struct module *mod, struct insn *insn, uint64_t val)
 		printk(KERN_ERR "%s: value %ld out of IMM60 range\n", mod->name, (int64_t) val);
 		return 0;
 	}
-	apply(insn, 0x011ffffe000, (  ((val & 0x1000000000000000) >> 24) /* bit 60 -> 36 */
-				    | ((val & 0x00000000000fffff) << 13) /* bit  0 -> 13 */));
-	apply((void *) insn - 1, 0x1fffffffffc, val >> 18);
+	ia64_patch_imm60((u64) insn, val);
 	return 1;
 }
 
@@ -221,10 +195,10 @@ apply_imm22 (struct module *mod, struct insn *insn, uint64_t val)
 		printk(KERN_ERR "%s: value %li out of IMM22 range\n", mod->name, (int64_t)val);
 		return 0;
 	}
-	apply(insn, 0x01fffcfe000, (  ((val & 0x200000) << 15) /* bit 21 -> 36 */
-				    | ((val & 0x1f0000)	<<  6) /* bit 16 -> 22 */
-				    | ((val & 0x00ff80) << 20) /* bit  7 -> 27 */
-				    | ((val & 0x00007f)	<< 13) /* bit  0 -> 13 */));
+	ia64_patch((u64) insn, 0x01fffcfe000, (  ((val & 0x200000) << 15) /* bit 21 -> 36 */
+					       | ((val & 0x1f0000) <<  6) /* bit 16 -> 22 */
+					       | ((val & 0x00ff80) << 20) /* bit  7 -> 27 */
+					       | ((val & 0x00007f) << 13) /* bit  0 -> 13 */));
 	return 1;
 }
 
@@ -235,8 +209,8 @@ apply_imm21b (struct module *mod, struct insn *insn, uint64_t val)
 		printk(KERN_ERR "%s: value %li out of IMM21b range\n", mod->name, (int64_t)val);
 		return 0;
 	}
-	apply(insn, 0x11ffffe000, (  ((val & 0x100000) << 16) /* bit 20 -> 36 */
-				   | ((val & 0x0fffff) << 13) /* bit  0 -> 13 */));
+	ia64_patch((u64) insn, 0x11ffffe000, (  ((val & 0x100000) << 16) /* bit 20 -> 36 */
+					      | ((val & 0x0fffff) << 13) /* bit  0 -> 13 */));
 	return 1;
 }
 
@@ -281,7 +255,7 @@ plt_target (struct plt_entry *plt)
 	b0 = b[0]; b1 = b[1];
 	off = (  ((b1 & 0x00fffff000000000) >> 36)		/* imm20b -> bit 0 */
 	       | ((b0 >> 48) << 20) | ((b1 & 0x7fffff) << 36)	/* imm39 -> bit 20 */
-	       | ((b1 & 0x0800000000000000) << 1));		/* i -> bit 60 */
+	       | ((b1 & 0x0800000000000000) << 0));		/* i -> bit 59 */
 	return (long) plt->bundle[1] + 16*off;
 }
 
@@ -751,7 +725,7 @@ do_reloc (struct module *mod, uint8_t r_type, Elf64_Sym *sym, uint64_t addend,
 			if (gp_addressable(mod, val)) {
 				/* turn "ld8" into "mov": */
 				DEBUGP("%s: patching ld8 at %p to mov\n", __FUNCTION__, location);
-				apply(location, 0x1fff80fe000, 0x10000000000);
+				ia64_patch((u64) location, 0x1fff80fe000, 0x10000000000);
 			}
 			return 0;
 
@@ -889,7 +863,8 @@ module_arch_cleanup (struct module *mod)
 }
 
 #ifdef CONFIG_SMP
-void percpu_modcopy(void *pcpudst, const void *src, unsigned long size)
+void
+percpu_modcopy (void  *pcpudst, const void *src, unsigned long size)
 {
 	unsigned int i;
 	for (i = 0; i < NR_CPUS; i++)

arch/ia64/kernel/pal.S

@@ -164,7 +164,7 @@ GLOBAL_ENTRY(ia64_pal_call_phys_static)
 	;;
 	mov loc4=ar.rsc			// save RSE configuration
 	dep.z loc2=loc2,0,61		// convert pal entry point to physical
-	dep.z r8=r8,0,61		// convert rp to physical
+	tpa r8=r8			// convert rp to physical
 	;;
 	mov b7 = loc2			// install target to branch reg
 	mov ar.rsc=0			// put RSE in enforced lazy, LE mode
@@ -174,13 +174,13 @@ GLOBAL_ENTRY(ia64_pal_call_phys_static)
 	or loc3=loc3,r17		// add in psr the bits to set
 	;;
 	andcm r16=loc3,r16		// removes bits to clear from psr
-	br.call.sptk.many rp=ia64_switch_mode
+	br.call.sptk.many rp=ia64_switch_mode_phys
 .ret1:	mov rp = r8			// install return address (physical)
 	br.cond.sptk.many b7
 1:
 	mov ar.rsc=0			// put RSE in enforced lazy, LE mode
 	mov r16=loc3			// r16= original psr
-	br.call.sptk.many rp=ia64_switch_mode // return to virtual mode
+	br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode
 .ret2:
 	mov psr.l = loc3		// restore init PSR
 
@@ -228,13 +228,13 @@ GLOBAL_ENTRY(ia64_pal_call_phys_stacked)
 	mov b7 = loc2			// install target to branch reg
 	;;
 	andcm r16=loc3,r16		// removes bits to clear from psr
-	br.call.sptk.many rp=ia64_switch_mode
+	br.call.sptk.many rp=ia64_switch_mode_phys
 .ret6:
 	br.call.sptk.many rp=b7		// now make the call
 .ret7:
 	mov ar.rsc=0			// put RSE in enforced lazy, LE mode
 	mov r16=loc3			// r16= original psr
-	br.call.sptk.many rp=ia64_switch_mode	// return to virtual mode
+	br.call.sptk.many rp=ia64_switch_mode_virt	// return to virtual mode
 
 .ret8:	mov psr.l  = loc3		// restore init PSR
 	mov ar.pfs = loc1

arch/ia64/kernel/smpboot.c

@@ -352,10 +352,10 @@ static struct task_struct * __init
 fork_by_hand (void)
 {
 	/*
-	 * don't care about the eip and regs settings since we'll never reschedule the
+	 * Don't care about the IP and regs settings since we'll never reschedule the
 	 * forked task.
 	 */
-	return do_fork(CLONE_VM|CLONE_IDLETASK, 0, 0, 0, NULL, NULL);
+	return copy_process(CLONE_VM|CLONE_IDLETASK, 0, 0, 0, NULL, NULL);
 }
 
 static int __init
@@ -370,6 +370,7 @@ do_boot_cpu (int sapicid, int cpu)
 	idle = fork_by_hand();
 	if (IS_ERR(idle))
 		panic("failed fork for CPU %d", cpu);
+	wake_up_forked_process(idle);
 
 	/*
 	 * We remove it from the pidhash and the runqueue
@@ -449,7 +450,7 @@ smp_build_cpu_map (void)
 
 	for (cpu = 1, i = 0; i < smp_boot_data.cpu_count; i++) {
 		sapicid = smp_boot_data.cpu_phys_id[i];
-		if (sapicid == -1 || sapicid == boot_cpu_id)
+		if (sapicid == boot_cpu_id)
 			continue;
 		phys_cpu_present_map |= (1 << cpu);
 		ia64_cpu_to_sapicid[cpu] = sapicid;
@@ -598,7 +599,7 @@ init_smp_config(void)
 	/* Tell SAL where to drop the AP's.  */
 	ap_startup = (struct fptr *) start_ap;
 	sal_ret = ia64_sal_set_vectors(SAL_VECTOR_OS_BOOT_RENDEZ,
-				       __pa(ap_startup->fp), __pa(ap_startup->gp), 0, 0, 0, 0);
+				       ia64_tpa(ap_startup->fp), ia64_tpa(ap_startup->gp), 0, 0, 0, 0);
 	if (sal_ret < 0)
 		printk(KERN_ERR "SMP: Can't set SAL AP Boot Rendezvous: %s\n",
 		       ia64_sal_strerror(sal_ret));