[PATCH] x86-64 updates for 2.5.43

This fixes a few files that got lost with the last merge and merges
with 2.5.43/i386.  Only changes architecture specific files.

It depends on one other patch (for linux/ioctl32.h) which I'm sending
separately.

Changes:
- Include missing files (pageattr.c) and Makefile changes
- Update IA32 subsystem. Various small fixes and a big merge
  with sparc64.
- Change HZ to 1000
- Merge some of the 2.5.43/i386 profiling changes. No full oprofile yet.
- Fix many warnings
- Update defconfig
- Various other smaller cleanups and bugfixes.

arch/x86_64/Config.help

@@ -435,6 +435,14 @@ CONFIG_X86_CPUID
   with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to
   /dev/cpu/31/cpuid.
 
+CONFIG_NR_CPUS
+  This allows you to specify the maximum number of CPUs which this
+  kernel will support.  The maximum supported value is 32 and the
+  minimum value which makes sense is 2.
+
+  This is purely to save memory - each supported CPU adds
+  approximately eight kilobytes to the kernel image.
+
 CONFIG_PREEMPT
   This option reduces the latency of the kernel when reacting to
   real-time or interactive events by allowing a low priority process to
@@ -487,10 +495,6 @@ CONFIG_CHECKING
   Enables some internal consistency checks for kernel debugging.
   You should normally say N.
 
-CONFIG_SIMNOW
-  Disable some time consuming optional things for slow CPU simulators.
-  Say N unless you're running on a slow simulator like Simics or SimNow. 
-
 CONFIG_EARLY_PRINTK
   Write kernel log output directly into the VGA buffer. This is useful
   for kernel debugging when your machine crashes very early before
@@ -499,13 +503,26 @@ CONFIG_EARLY_PRINTK
   klogd/syslogd or the X server.You should normally N here, unless
   you want to debug such a crash.
   
-CONFIG_X86_MCE_NONFATAL
-  Enabling this feature starts a timer that triggers every 5 seconds which
-  will look at the machine check registers to see if anything happened.
-  Non-fatal problems automatically get corrected (but still logged).
-  Disable this if you don't want to see these messages.
-  Seeing the messages this option prints out may be indicative of dying hardware,
-  or out-of-spec (ie, overclocked) hardware.
-  This option only does something on hardware with Intel P6 style MCE.
-  (Pentium Pro and above, AMD Athlon/Duron)
+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_INIT_DEBUG
+  Fill __init and __initdata at the end of boot. This is only for debugging.
+
+CONFIG_IA32_EMULATION
+  Include code to run 32bit programs under an 64bit kernel. You should likely
+  turn this on, unless you're 100% sure that you don't have any 32bit programs
+  left.
+
+CONFIG_GART_IOMMU
+  Support the K8 IOMMU. Needed to run systems with more than 4GB of memory
+  properly with 32bit devices. You should probably turn this on.
+  The iommu can be turned off at runtime with the iommu=off parameter.
+
+CONFIG_DUMMY_IOMMU
+  Don't use IOMMU code. This will cause problems when you have more than 4GB 
+  of memory and any 32bit devices. Don't turn on unless you know what you
+  are doing.
 

arch/x86_64/Makefile

@@ -33,7 +33,7 @@ IA32_CPP := $(CROSS_COMPILE)gcc -m32 -E
 export IA32_CC IA32_LD IA32_AS IA32_OBJCOPY IA32_CPP
 
 
-LD=$(CROSS_COMPILE)ld -m elf_x86_64
+LDFLAGS		:= -m elf_x86_64
 OBJCOPYFLAGS	:= -O binary -R .note -R .comment -S
 LDFLAGS_vmlinux := -e stext
 
@@ -47,52 +47,39 @@ CFLAGS += -finline-limit=2000
 
 HEAD := arch/x86_64/kernel/head.o arch/x86_64/kernel/head64.o arch/x86_64/kernel/init_task.o
 
-SUBDIRS += arch/x86_64/kernel arch/x86_64/mm arch/x86_64/lib
-CORE_FILES := arch/x86_64/kernel/kernel.o $(CORE_FILES)
-CORE_FILES +=  arch/x86_64/mm/mm.o
-LIBS := $(TOPDIR)/arch/x86_64/lib/lib.a $(LIBS)
-
-ifdef CONFIG_IA32_EMULATION
-SUBDIRS += arch/x86_64/ia32
-CORE_FILES += arch/x86_64/ia32/ia32.o
-endif
-
-ifdef CONFIG_PCI
-SUBDIRS += arch/x86_64/pci
-DRIVERS += arch/x86_64/pci/pci.o
-endif
-
-CORE_FILES += $(core-y)
+libs-y 					+= arch/x86_64/lib/
+core-y					+= arch/x86_64/kernel/ arch/x86_64/mm/
+core-$(CONFIG_IA32_EMULATION)		+= arch/x86_64/ia32/
+drivers-$(CONFIG_PCI)			+= arch/x86_64/pci/
 
 MAKEBOOT = $(MAKE) -C arch/$(ARCH)/boot
+makeboot = $(call descend,arch/x86_64/boot,$(1))
 
 .PHONY: zImage bzImage compressed zlilo bzlilo zdisk bzdisk install \
 		clean archclean archmrproper
 
-bzImage: vmlinux
-	@$(MAKEBOOT) bzImage
+BOOTIMAGE=arch/x86_64/boot/bzImage
+zImage zlilo zdisk: BOOTIMAGE=arch/x86_64/boot/zImage
 
-bzImage-padded: vmlinux
-	@$(MAKEBOOT) bzImage-padded
+zImage bzImage: vmlinux
+	+@$(call makeboot,$(BOOTIMAGE))
 
-tmp:
-	@$(MAKEBOOT) BOOTIMAGE=bzImage zlilo
+compressed: zImage
 
-bzlilo: vmlinux
-	@$(MAKEBOOT) BOOTIMAGE=bzImage zlilo
+zlilo bzlilo: vmlinux
+	+@$(call makeboot,BOOTIMAGE=$(BOOTIMAGE) zlilo)
 
-bzdisk: vmlinux
-	@$(MAKEBOOT) BOOTIMAGE=bzImage zdisk
+zdisk bzdisk: vmlinux
+	+@$(call makeboot,BOOTIMAGE=$(BOOTIMAGE) zdisk)
 
 install: vmlinux
-	@$(MAKEBOOT) BOOTIMAGE=bzImage install
+	+@$(call makeboot,BOOTIMAGE=$(BOOTIMAGE) install)
 
 archclean:
-	@$(MAKEBOOT) clean
+	+@$(call makeboot,clean)
 
 archmrproper:
 
-
 prepare: include/asm-$(ARCH)/offset.h
 
 arch/$(ARCH)/kernel/asm-offsets.s: include/asm include/linux/version.h \

arch/x86_64/boot/compressed/misc.c

@@ -190,7 +190,7 @@ static void puts(const char *s)
 	outb_p(0xff & (pos >> 1), vidport+1);
 }
 
-void* memset(void* s, int c, size_t n)
+void* memset(void* s, int c, unsigned n)
 {
 	int i;
 	char *ss = (char*)s;
@@ -199,14 +199,13 @@ void* memset(void* s, int c, size_t n)
 	return s;
 }
 
-void* memcpy(void* __dest, __const void* __src,
-			    size_t __n)
+void* memcpy(void* dest, const void* src, unsigned n)
 {
 	int i;
-	char *d = (char *)__dest, *s = (char *)__src;
+	char *d = (char *)dest, *s = (char *)src;
 
-	for (i=0;i<__n;i++) d[i] = s[i];
-	return __dest;
+	for (i=0;i<n;i++) d[i] = s[i];
+	return dest;
 }
 
 /* ===========================================================================

arch/x86_64/config.in

@@ -21,6 +21,7 @@ define_bool CONFIG_RWSEM_XCHGADD_ALGORITHM n
 define_bool CONFIG_X86_CMPXCHG y
 
 define_bool CONFIG_EARLY_PRINTK y
+define_bool CONFIG_GENERIC_ISA_DMA y
 
 source init/Config.in
 
@@ -54,11 +55,17 @@ if [ "$CONFIG_SMP" = "n" ]; then
 fi
 if [ "$CONFIG_SMP" = "y" ]; then
     define_bool CONFIG_HAVE_DEC_LOCK y
+    # actually 64 maximum, but you need to fix the APIC code first
+    # to use clustered mode or whatever your big iron needs
     int  'Maximum number of CPUs (2-8)' CONFIG_NR_CPUS 8
 fi
 
+bool 'IOMMU support' CONFIG_GART_IOMMU
+if [ "$CONFIG_GART_IOMMU" != "y" ]; then
+   define_bool CONFIG_DUMMY_IOMMU y
+fi
+
 define_bool CONFIG_X86_MCE y
-bool 'Check for non-fatal machine check errors' CONFIG_X86_MCE_NONFATAL $CONFIG_X86_MCE
 
 endmenu
 
@@ -108,22 +115,20 @@ tristate 'Kernel support for MISC binaries' CONFIG_BINFMT_MISC
 
 bool 'IA32 Emulation' CONFIG_IA32_EMULATION
 
+
+
 endmenu
 
 source drivers/mtd/Config.in
 
 source drivers/parport/Config.in
 
-#source drivers/pnp/Config.in
-
 source drivers/block/Config.in
 
-source drivers/md/Config.in
-
 mainmenu_option next_comment
 comment 'ATA/ATAPI/MFM/RLL support'
 
-tristate 'ATA/ATAPI/MFM/RLL support' CONFIG_IDE
+tristate 'ATA/ATAPI/MFM/RLL device support' CONFIG_IDE
 
 if [ "$CONFIG_IDE" != "n" ]; then
   source drivers/ide/Config.in
@@ -133,15 +138,19 @@ fi
 endmenu
 
 mainmenu_option next_comment
-comment 'SCSI support'
+comment 'SCSI device support'
 
-tristate 'SCSI support' CONFIG_SCSI
+tristate 'SCSI device support' CONFIG_SCSI
 
 if [ "$CONFIG_SCSI" != "n" ]; then
    source drivers/scsi/Config.in
 fi
 endmenu
 
+source drivers/md/Config.in
+
+source drivers/telephony/Config.in
+
 source drivers/message/fusion/Config.in
 
 source drivers/ieee1394/Config.in
@@ -172,8 +181,6 @@ source net/irda/Config.in
 
 source drivers/isdn/Config.in
 
-source drivers/telephony/Config.in
-
 # no support for non IDE/SCSI cdroms as they were all ISA only
 
 #
@@ -224,6 +231,8 @@ if [ "$CONFIG_DEBUG_KERNEL" != "n" ]; then
    bool '  Spinlock debugging' CONFIG_DEBUG_SPINLOCK
    bool '  Additional run-time checks' CONFIG_CHECKING
    bool '  Debug __init statements' CONFIG_INIT_DEBUG
+   bool '  Spinlock debugging' CONFIG_DEBUG_SPINLOCK
+   bool '  Load all symbols for debugging/kksymoops' CONFIG_KALLSYMS
 fi
 endmenu
 

arch/x86_64/ia32/ia32_binfmt.c

@@ -25,7 +25,7 @@ struct elf_phdr;
 
 #define IA32_EMULATOR 1
 
-#define IA32_PAGE_OFFSET 0xE0000000
+#define IA32_PAGE_OFFSET 0xffff0000
 #define IA32_STACK_TOP IA32_PAGE_OFFSET
 #define ELF_ET_DYN_BASE		(IA32_PAGE_OFFSET/3 + 0x1000000)
 
@@ -216,13 +216,15 @@ static void elf32_init(struct pt_regs *regs)
 extern void put_dirty_page(struct task_struct * tsk, struct page *page, unsigned long address);
  
 
-int ia32_setup_arg_pages(struct linux_binprm *bprm)
+int setup_arg_pages(struct linux_binprm *bprm)
 {
 	unsigned long stack_base;
 	struct vm_area_struct *mpnt;
+	struct mm_struct *mm = current->mm;
 	int i;
 
-	stack_base = IA32_STACK_TOP - MAX_ARG_PAGES*PAGE_SIZE;
+	stack_base = IA32_STACK_TOP - MAX_ARG_PAGES * PAGE_SIZE;
+	mm->arg_start = bprm->p + stack_base;
 
 	bprm->p += stack_base;
 	if (bprm->loader)
@@ -233,9 +235,14 @@ int ia32_setup_arg_pages(struct linux_binprm *bprm)
 	if (!mpnt) 
 		return -ENOMEM; 
 	
-	down_write(&current->mm->mmap_sem);
+	if (!vm_enough_memory((IA32_STACK_TOP - (PAGE_MASK & (unsigned long) bprm->p))>>PAGE_SHIFT)) {
+		kmem_cache_free(vm_area_cachep, mpnt);
+		return -ENOMEM;
+	}
+
+	down_write(&mm->mmap_sem);
 	{
-		mpnt->vm_mm = current->mm;
+		mpnt->vm_mm = mm;
 		mpnt->vm_start = PAGE_MASK & (unsigned long) bprm->p;
 		mpnt->vm_end = IA32_STACK_TOP;
 		mpnt->vm_page_prot = PAGE_COPY;
@@ -243,24 +250,25 @@ int ia32_setup_arg_pages(struct linux_binprm *bprm)
 		mpnt->vm_ops = NULL;
 		mpnt->vm_pgoff = 0;
 		mpnt->vm_file = NULL;
+		INIT_LIST_HEAD(&mpnt->shared);
 		mpnt->vm_private_data = (void *) 0;
-		insert_vm_struct(current->mm, mpnt);
-		current->mm->total_vm = (mpnt->vm_end - mpnt->vm_start) >> PAGE_SHIFT;
+		insert_vm_struct(mm, mpnt);
+		mm->total_vm = (mpnt->vm_end - mpnt->vm_start) >> PAGE_SHIFT;
 	} 
 
 	for (i = 0 ; i < MAX_ARG_PAGES ; i++) {
 		struct page *page = bprm->page[i];
 		if (page) {
 			bprm->page[i] = NULL;
-			current->mm->rss++;
 			put_dirty_page(current,page,stack_base);
 		}
 		stack_base += PAGE_SIZE;
 	}
-	up_write(&current->mm->mmap_sem);
+	up_write(&mm->mmap_sem);
 	
 	return 0;
 }
+
 static unsigned long
 elf32_map (struct file *filep, unsigned long addr, struct elf_phdr *eppnt, int prot, int type)
 {

arch/x86_64/ia32/ia32_signal.c

@@ -7,7 +7,7 @@
  *  2000-06-20  Pentium III FXSR, SSE support by Gareth Hughes
  *  2000-12-*   x86-64 compatibility mode signal handling by Andi Kleen
  * 
- *  $Id: ia32_signal.c,v 1.17 2002/03/21 14:16:32 ak Exp $
+ *  $Id: ia32_signal.c,v 1.22 2002/07/29 10:34:03 ak Exp $
  */
 
 #include <linux/sched.h>
@@ -39,6 +39,7 @@
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
 asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
+void signal_fault(struct pt_regs *regs, void *frame, char *where);
 
 static int ia32_copy_siginfo_to_user(siginfo_t32 *to, siginfo_t *from)
 {
@@ -163,32 +164,38 @@ ia32_restore_sigcontext(struct pt_regs *regs, struct sigcontext_ia32 *sc, unsign
 	regs->r ## x = reg;			\
 }
 
-#define RELOAD_SEG(seg)							\
+#define RELOAD_SEG(seg,mask)						\
 	{ unsigned int cur; 				\
 	  unsigned short pre;				\
 	  err |= __get_user(pre, &sc->seg);				\
     	  asm volatile("movl %%" #seg ",%0" : "=r" (cur));		\
+	  pre |= mask; 							\
 	  if (pre != cur) loadsegment(seg,pre); }
 
-	/* Reload fs and gs if they have changed in the signal handler. */
+	/* Reload fs and gs if they have changed in the signal handler.
+	   This does not handle long fs/gs base changes in the handler, but 
+	   does not clobber them at least in the normal case. */ 
 	
 	{
-		unsigned short gs; 
+		unsigned gs, oldgs; 
 		err |= __get_user(gs, &sc->gs);
+		gs |= 3; 
+		asm("movl %%gs,%0" : "=r" (oldgs));
+		if (gs != oldgs)
 		load_gs_index(gs); 
 	} 
-	RELOAD_SEG(fs);
-	RELOAD_SEG(ds);
-	RELOAD_SEG(es);
+	RELOAD_SEG(fs,3);
+	RELOAD_SEG(ds,3);
+	RELOAD_SEG(es,3);
 
 	COPY(di); COPY(si); COPY(bp); COPY(sp); COPY(bx);
 	COPY(dx); COPY(cx); COPY(ip);
 	/* Don't touch extended registers */ 
 	
 	err |= __get_user(regs->cs, &sc->cs); 
-	regs->cs |= 2;  
+	regs->cs |= 3;  
 	err |= __get_user(regs->ss, &sc->ss); 
-	regs->ss |= 2; 
+	regs->ss |= 3; 
 
 	{
 		unsigned int tmpflags;
@@ -284,7 +291,7 @@ asmlinkage int sys32_rt_sigreturn(struct pt_regs regs)
 	return eax;
 
 badframe:
-	signal_fault(&regs, frame, "32bit rt sigreturn"); 
+	signal_fault(&regs,frame,"32bit rt sigreturn");
 	return 0;
 }	
 
@@ -377,8 +384,9 @@ void ia32_setup_frame(int sig, struct k_sigaction *ka,
 
 	{
 		struct exec_domain *ed = current_thread_info()->exec_domain;
-		
-		err |= __put_user((ed && ed->signal_invmap && sig < 32
+	err |= __put_user((ed
+		           && ed->signal_invmap
+		           && sig < 32
 		           ? ed->signal_invmap[sig]
 		           : sig),
 		          &frame->sig);
@@ -435,7 +443,7 @@ void ia32_setup_frame(int sig, struct k_sigaction *ka,
 give_sigsegv:
 	if (sig == SIGSEGV)
 		ka->sa.sa_handler = SIG_DFL;
-	signal_fault(regs,frame,"32bit signal setup"); 
+	signal_fault(regs,frame,"32bit signal deliver");
 }
 
 void ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
@@ -449,9 +457,12 @@ void ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
 		goto give_sigsegv;
 
+
 	{
 		struct exec_domain *ed = current_thread_info()->exec_domain;
-		err |= __put_user((ed && ed->signal_invmap && sig < 32
+	err |= __put_user((ed
+		    	   && ed->signal_invmap
+		    	   && sig < 32
 		    	   ? ed->signal_invmap[sig]
 			   : sig),
 			  &frame->sig);

arch/x86_64/ia32/ia32entry.S

@@ -27,7 +27,7 @@
  */ 	
 ENTRY(ia32_cstar_target)
 	movq $-ENOSYS,%rax
-	sysret
+	sysretl
 	
 /* 
  * Emulated IA32 system calls via int 0x80. 
@@ -77,6 +77,7 @@ ia32_tracesys:
 	jmp ia32_do_syscall
 
 ia32_badsys:
+	movq $0,ORIG_RAX-ARGOFFSET(%rsp)
 	movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
 	jmp int_ret_from_sys_call
 
@@ -84,6 +85,10 @@ ni_syscall:
 	movq %rax,%rdi
 	jmp  sys32_ni_syscall			
 
+quiet_ni_syscall:
+	movq $-ENOSYS,%rax
+	ret
+	
 	.macro PTREGSCALL label, func
 	.globl \label
 \label:
@@ -229,7 +234,7 @@ ia32_sys_call_table:
 	.quad stub32_iopl		/* 110 */
 	.quad sys_vhangup
 	.quad ni_syscall	/* old "idle" system call */
-	.quad ni_syscall	/* vm86old */ 
+	.quad sys32_vm86_warning	/* vm86old */ 
 	.quad sys32_wait4
 	.quad sys_swapoff		/* 115 */
 	.quad sys32_sysinfo
@@ -238,7 +243,7 @@ ia32_sys_call_table:
 	.quad stub32_sigreturn
 	.quad stub32_clone		/* 120 */
 	.quad sys_setdomainname
-	.quad sys32_newuname
+	.quad sys_uname
 	.quad sys_modify_ldt
 	.quad sys32_adjtimex
 	.quad sys32_mprotect		/* 125 */
@@ -282,8 +287,8 @@ ia32_sys_call_table:
 	.quad sys_mremap
 	.quad sys_setresuid16
 	.quad sys_getresuid16	/* 165 */
-	.quad ni_syscall	/* vm86 */ 
-	.quad ni_syscall	/* query_module */
+	.quad sys32_vm86_warning	/* vm86 */ 
+	.quad quiet_ni_syscall	/* query_module */
 	.quad sys_poll
 	.quad sys32_nfsservctl
 	.quad sys_setresgid16	/* 170 */
@@ -368,7 +373,7 @@ ia32_sys_call_table:
 	.quad sys_io_cancel
 	.quad sys_ni_syscall /* 250 alloc_huge_pages */
 	.quad sys_ni_syscall /* free_huge_pages */
-	.quad sys_ni_syscall /* exit_group */
+	.quad sys_exit_group /* exit_group */
 ia32_syscall_end:		
 	.rept IA32_NR_syscalls-(ia32_syscall_end-ia32_sys_call_table)/8
 		.quad ni_syscall

arch/x86_64/ia32/ipc32.c

@@ -264,6 +264,9 @@ semctl32 (int first, int second, int third, void *uptr)
 		if (err2)
 		    err = -EFAULT;
 		break;
+	default:
+		err = -EINVAL;
+		break;
 	}
 	return err;
 }
@@ -284,7 +287,7 @@ do_sys32_msgsnd (int first, int second, int third, void *uptr)
 	if (!p)
 		return -ENOMEM;
 	err = get_user(p->mtype, &up->mtype);
-	err |= copy_from_user(p->mtext, &up->mtext, second);
+	err |= (copy_from_user(p->mtext, &up->mtext, second) ? -EFAULT : 0);
 	if (err)
 		goto out;
 	old_fs = get_fs();

arch/x86_64/ia32/sys_ia32.c

@@ -35,7 +35,6 @@
 #include <linux/smp_lock.h>
 #include <linux/sem.h>
 #include <linux/msg.h>
-#include <linux/binfmts.h>
 #include <linux/mm.h>
 #include <linux/shm.h>
 #include <linux/slab.h>
@@ -56,6 +55,7 @@
 #include <linux/stat.h>
 #include <linux/ipc.h>
 #include <linux/rwsem.h>
+#include <linux/binfmts.h>
 #include <linux/init.h>
 #include <linux/aio_abi.h>
 #include <asm/mman.h>
@@ -264,8 +264,11 @@ sys32_mmap(struct mmap_arg_struct *arg)
 		if (!file)
 			return -EBADF;
 	}
+	
+#if 0 /* reenable when noexec works */
 	if (a.prot & PROT_READ) 
 		a.prot |= PROT_EXEC; 
+#endif
 
 	a.flags |= MAP_32BIT;
 
@@ -275,8 +278,8 @@ sys32_mmap(struct mmap_arg_struct *arg)
 	if (file)
 		fput(file);
 
-	/* Should not happen */
-	if (retval >= 0xFFFFFFFF && (long)retval > 0) { 
+	/* Cannot wrap */
+	if (retval+a.len >= 0xFFFFFFFF && (long)retval > 0) { 
 		do_munmap(mm, retval, a.len); 
 		retval = -ENOMEM; 
 	} 
@@ -964,22 +967,24 @@ sys32_nanosleep(struct timespec32 *rqtp, struct timespec32 *rmtp)
 asmlinkage ssize_t sys_readv(unsigned long,const struct iovec *,unsigned long);
 asmlinkage ssize_t sys_writev(unsigned long,const struct iovec *,unsigned long);
 
-struct iovec *
-get_iovec32(struct iovec32 *iov32, struct iovec *iov_buf, u32 count, int type)
+static struct iovec *
+get_iovec32(struct iovec32 *iov32, struct iovec *iov_buf, u32 count, int type, int *errp)
 {
 	int i;
 	u32 buf, len;
 	struct iovec *ivp, *iov;
+	unsigned long totlen; 
 
 	/* Get the "struct iovec" from user memory */
 
 	if (!count)
 		return 0;
-	if(verify_area(VERIFY_READ, iov32, sizeof(struct iovec32)*count))
-		return(struct iovec *)0;
 	if (count > UIO_MAXIOV)
 		return(struct iovec *)0;
+	if(verify_area(VERIFY_READ, iov32, sizeof(struct iovec32)*count))
+		return(struct iovec *)0;
 	if (count > UIO_FASTIOV) {
+		*errp = -ENOMEM; 
 		iov = kmalloc(count*sizeof(struct iovec), GFP_KERNEL);
 		if (!iov)
 			return((struct iovec *)0);
@@ -987,24 +992,33 @@ get_iovec32(struct iovec32 *iov32, struct iovec *iov_buf, u32 count, int type)
 		iov = iov_buf;
 
 	ivp = iov;
+	totlen = 0;
 	for (i = 0; i < count; i++) {
-		if (__get_user(len, &iov32->iov_len) ||
-		    __get_user(buf, &iov32->iov_base)) {
-			if (iov != iov_buf)
-				kfree(iov);
-			return((struct iovec *)0);
-		}
-		if (verify_area(type, (void *)A(buf), len)) {
-			if (iov != iov_buf)
-				kfree(iov);
-			return((struct iovec *)0);
-		}
+		*errp = __get_user(len, &iov32->iov_len) |
+		  	__get_user(buf, &iov32->iov_base);	
+		if (*errp)
+			goto error;
+		*errp = verify_area(type, (void *)A(buf), len);
+		if (*errp) 
+			goto error;
+		/* SuS checks: */
+		*errp = -EINVAL; 
+		if ((int)len < 0)
+			goto error;
+		if ((totlen += len) >= 0x7fffffff)
+			goto error;			
 		ivp->iov_base = (void *)A(buf);
 		ivp->iov_len = (__kernel_size_t)len;
 		iov32++;
 		ivp++;
 	}
+	*errp = 0;
 	return(iov);
+
+error:
+	if (iov != iov_buf)
+		kfree(iov);
+	return NULL;
 }
 
 asmlinkage long
@@ -1015,8 +1029,8 @@ sys32_readv(int fd, struct iovec32 *vector, u32 count)
 	int ret;
 	mm_segment_t old_fs = get_fs();
 
-	if ((iov = get_iovec32(vector, iovstack, count, VERIFY_WRITE)) == (struct iovec *)0)
-		return -EFAULT;
+	if ((iov = get_iovec32(vector, iovstack, count, VERIFY_WRITE, &ret)) == NULL)
+		return ret;
 	set_fs(KERNEL_DS);
 	ret = sys_readv(fd, iov, count);
 	set_fs(old_fs);
@@ -1033,8 +1047,8 @@ sys32_writev(int fd, struct iovec32 *vector, u32 count)
 	int ret;
 	mm_segment_t old_fs = get_fs();
 
-	if ((iov = get_iovec32(vector, iovstack, count, VERIFY_READ)) == (struct iovec *)0)
-		return -EFAULT;
+	if ((iov = get_iovec32(vector, iovstack, count, VERIFY_READ, &ret)) == NULL)
+		return ret;
 	set_fs(KERNEL_DS);
 	ret = sys_writev(fd, iov, count);
 	set_fs(old_fs);
@@ -1121,7 +1135,7 @@ sys32_setrlimit(unsigned int resource, struct rlimit32 *rlim)
 
 /*
  * sys_time() can be implemented in user-level using
- * sys_gettimeofday().  IA64 did this but i386 Linux did not
+ * sys_gettimeofday().  x86-64 did this but i386 Linux did not
  * so we have to implement this system call here.
  */
 asmlinkage long sys32_time(int * tloc)
@@ -1288,6 +1302,39 @@ static inline int put_flock(struct flock *kfl, struct flock32 *ufl)
 }
 
 extern asmlinkage long sys_fcntl(unsigned int fd, unsigned int cmd, unsigned long arg);
+asmlinkage long sys32_fcntl64(unsigned int fd, unsigned int cmd, unsigned long arg);
+
+
+asmlinkage long sys32_fcntl(unsigned int fd, unsigned int cmd, unsigned long arg)
+{
+	switch (cmd) {
+	case F_GETLK:
+	case F_SETLK:
+	case F_SETLKW:
+		{
+			struct flock f;
+			mm_segment_t old_fs;
+			long ret;
+			
+			if (get_flock(&f, (struct flock32 *)arg))
+				return -EFAULT;
+			old_fs = get_fs(); set_fs (KERNEL_DS);
+			ret = sys_fcntl(fd, cmd, (unsigned long)&f);
+			set_fs (old_fs);
+			if (ret) return ret;
+			if (put_flock(&f, (struct flock32 *)arg))
+				return -EFAULT;
+			return 0;
+		}
+	case F_GETLK64: 
+	case F_SETLK64: 
+	case F_SETLKW64: 
+		return sys32_fcntl64(fd,cmd,arg); 
+
+	default:
+		return sys_fcntl(fd, cmd, (unsigned long)arg);
+	}
+}
 
 static inline int get_flock64(struct ia32_flock64 *fl32, struct flock *fl64)
 {
@@ -1319,40 +1366,34 @@ asmlinkage long sys32_fcntl64(unsigned int fd, unsigned int cmd, unsigned long a
 {
 	struct flock fl64;  
 	mm_segment_t oldfs = get_fs(); 
-	int ret = 0, origcmd; 
-	unsigned long origarg;
+	int ret = 0; 
+	int oldcmd = cmd;
+	unsigned long oldarg = arg;
 
-	origcmd = cmd; 
-	origarg = arg;
 	switch (cmd) {
-	case F_GETLK: 
-	case F_SETLK:
-	case F_SETLKW:
-		ret = get_flock(&fl64, (struct flock32 *)arg); 
-		arg = (unsigned long) &fl64; 
-		set_fs(KERNEL_DS); 
-		break;
 	case F_GETLK64: 
 		cmd = F_GETLK; 
-		goto cnv64;
+		goto cnv;
 	case F_SETLK64: 
 		cmd = F_SETLK; 
-		goto cnv64; 
+		goto cnv; 
 	case F_SETLKW64:
 		cmd = F_SETLKW; 
-	cnv64:
+	cnv:
 		ret = get_flock64((struct ia32_flock64 *)arg, &fl64); 
 		arg = (unsigned long)&fl64; 
 		set_fs(KERNEL_DS); 
 		break; 
+	case F_GETLK:
+	case F_SETLK:
+	case F_SETLKW:
+		return sys32_fcntl(fd,cmd,arg); 
 	}
 	if (!ret)
 		ret = sys_fcntl(fd, cmd, arg);
 	set_fs(oldfs); 
-	if (origcmd == F_GETLK && !ret)
-		ret = put_flock(&fl64, (struct flock32 *)origarg); 
-	else if (cmd == F_GETLK && !ret)
-		ret = put_flock64((struct ia32_flock64 *)origarg, &fl64); 
+	if (oldcmd == F_GETLK64 && !ret)
+		ret = put_flock64((struct ia32_flock64 *)oldarg, &fl64); 
 	return ret; 
 }
 
@@ -1804,19 +1845,6 @@ sys32_sysctl(struct sysctl_ia32 *args32)
 #endif
 }
 
-extern asmlinkage long sys_newuname(struct new_utsname * name);
-
-asmlinkage long
-sys32_newuname(struct new_utsname * name)
-{
-	int ret = sys_newuname(name);
-	
-	if (current->personality == PER_LINUX32 && !ret) {
-		ret = copy_to_user(name->machine, "i686\0\0", 6);
-	}
-	return ret;
-}
-
 extern asmlinkage ssize_t sys_pread64(unsigned int fd, char * buf,
 				    size_t count, loff_t pos);
 
@@ -1825,7 +1853,8 @@ extern asmlinkage ssize_t sys_pwrite64(unsigned int fd, const char * buf,
 
 typedef __kernel_ssize_t32 ssize_t32;
 
-/* warning. next two assume LE */
+
+/* warning: next two assume little endian */ 
 asmlinkage ssize_t32
 sys32_pread(unsigned int fd, char *ubuf, __kernel_size_t32 count,
 	    u32 poslo, u32 poshi)
@@ -1849,7 +1878,8 @@ asmlinkage long
 sys32_personality(unsigned long personality)
 {
 	int ret;
-	if (current->personality == PER_LINUX32 && personality == PER_LINUX)
+	if (personality(current->personality) == PER_LINUX32 && 
+		personality == PER_LINUX)
 		personality = PER_LINUX32;
 	ret = sys_personality(personality);
 	if (ret == PER_LINUX32)
@@ -1956,42 +1986,39 @@ sys32_adjtimex(struct timex32 *utp)
 	return ret;
 }
 
-
-/* common code for old and new mmaps */
-static inline long do_mmap2(
-	unsigned long addr, unsigned long len,
+asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
 	unsigned long prot, unsigned long flags,
 	unsigned long fd, unsigned long pgoff)
 {
-	int error = -EBADF;
+	struct mm_struct *mm = current->mm;
+	unsigned long error;
 	struct file * file = NULL;
 
 	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
 	if (!(flags & MAP_ANONYMOUS)) {
 		file = fget(fd);
 		if (!file)
-			goto out;
+			return -EBADF;
 	}
 
-	down_write(&current->mm->mmap_sem);
-	error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
-	up_write(&current->mm->mmap_sem);
+	/* later add PROT_EXEC for PROT_READ here */
+
+	down_write(&mm->mmap_sem);
+	error = do_mmap_pgoff(file, addr, len, prot, flags|MAP_32BIT, pgoff);
+	up_write(&mm->mmap_sem);
+
+	/* cannot wrap */
+	if (error+len >= 0xFFFFFFFF && (long)error >= 0) { 
+		do_munmap(mm, error, len); 
+		error = -ENOMEM; 
+	} 
 
 	if (file)
 		fput(file);
-out:
 	return error;
 }
 
-asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
-	unsigned long prot, unsigned long flags,
-	unsigned long fd, unsigned long pgoff)
-{
-	return do_mmap2(addr, len, prot, flags, fd, pgoff);
-}
-
-
-asmlinkage int sys32_olduname(struct oldold_utsname * name)
+asmlinkage long sys32_olduname(struct oldold_utsname * name)
 {
 	int error;
 
@@ -2581,6 +2608,19 @@ long sys32_io_setup(unsigned nr_reqs, u32 *ctx32p)
 	return ret;
 } 
 
+
+int sys32_vm86_warning(void)
+{ 
+	static long warn_time = -(60*HZ); 
+	if (time_before(warn_time + 60*HZ,jiffies)) { 
+		printk(KERN_INFO "%s: vm86 mode not supported on 64 bit kernel\n",
+		       current->comm);
+		warn_time = jiffies;
+	} 
+	return -ENOSYS ;
+} 
+
+
 struct exec_domain ia32_exec_domain = { 
 	.name = "linux/x86",
 	.pers_low = PER_LINUX32,

arch/x86_64/kernel/Makefile

@@ -9,7 +9,7 @@ export-objs     := mtrr.o x8664_ksyms.o pci-gart.o
 obj-y	:= process.o semaphore.o signal.o entry.o traps.o irq.o \
 		ptrace.o i8259.o ioport.o ldt.o setup.o time.o sys_x86_64.o \
 		pci-dma.o x8664_ksyms.o i387.o syscall.o vsyscall.o \
-		setup64.o bluesmoke.o bootflag.o e820.o reboot.o
+		setup64.o bluesmoke.o bootflag.o e820.o reboot.o profile.o
 
 obj-$(CONFIG_MTRR)	+= mtrr.o
 obj-$(CONFIG_X86_MSR)	+= msr.o

arch/x86_64/kernel/apic.c

@@ -897,17 +897,9 @@ int setup_profiling_timer(unsigned int multiplier)
 
 inline void smp_local_timer_interrupt(struct pt_regs *regs)
 {
-	int user = user_mode(regs);
 	int cpu = smp_processor_id();
 
-	/*
-	 * The profiling function is SMP safe. (nothing can mess
-	 * around with "current", and the profiling counters are
-	 * updated with atomic operations). This is especially
-	 * useful with a profiling multiplier != 1
-	 */
-	if (!user)
-		x86_do_profile(regs->rip);
+	x86_do_profile(regs);
 
 	if (--prof_counter[cpu] <= 0) {
 		/*
@@ -925,7 +917,7 @@ inline void smp_local_timer_interrupt(struct pt_regs *regs)
 		}
 
 #ifdef CONFIG_SMP
-		update_process_times(user);
+		update_process_times(user_mode(regs));
 #endif
 	}
 
@@ -951,8 +943,6 @@ inline void smp_local_timer_interrupt(struct pt_regs *regs)
  */
 void smp_apic_timer_interrupt(struct pt_regs *regs)
 {
-	int cpu = smp_processor_id();
-
 	/*
 	 * the NMI deadlock-detector uses this.
 	 */
@@ -1084,11 +1074,13 @@ int __init APIC_init_uniprocessor (void)
 static __init int setup_disableapic(char *str) 
 { 
 	disable_apic = 1;
+	return 0;
 } 
 
 static __init int setup_noapictimer(char *str) 
 { 
 	disable_apic_timer = 1;
+	return 0;
 } 
 
 __setup("disableapic", setup_disableapic); 

arch/x86_64/kernel/mpparse.c

@@ -785,7 +785,7 @@ void __init mp_register_ioapic (
 	mp_ioapic_routing[idx].irq_end = irq_base + 
 		io_apic_get_redir_entries(idx);
 
-	printk("IOAPIC[%d]: apic_id %d, version %d, address 0x%lx, "
+	printk("IOAPIC[%d]: apic_id %d, version %d, address 0x%x, "
 		"IRQ %d-%d\n", idx, mp_ioapics[idx].mpc_apicid, 
 		mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
 		mp_ioapic_routing[idx].irq_start,

arch/x86_64/kernel/mtrr.c

@@ -646,7 +646,7 @@ int mtrr_add_page (u64 base, u32 size, unsigned int type, char increment)
     }
 
 	if (base & (size_or_mask>>PAGE_SHIFT)) {
-		printk (KERN_WARNING "mtrr: base(%lx) exceeds the MTRR width(%lx)\n",
+		printk (KERN_WARNING "mtrr: base(%Lx) exceeds the MTRR width(%Lx)\n",
 				base, (size_or_mask>>PAGE_SHIFT));
 		return -EINVAL;
 	}

arch/x86_64/kernel/nmi.c

@@ -23,6 +23,7 @@
 #include <asm/smp.h>
 #include <asm/mtrr.h>
 #include <asm/mpspec.h>
+#include <asm/nmi.h>
 
 unsigned int nmi_watchdog = NMI_LOCAL_APIC;
 static unsigned int nmi_hz = HZ;
@@ -137,6 +138,18 @@ static int nmi_pm_callback(struct pm_dev *dev, pm_request_t rqst, void *data)
 	return 0;
 }
 
+struct pm_dev * set_nmi_pm_callback(pm_callback callback)
+{
+	apic_pm_unregister(nmi_pmdev);
+	return apic_pm_register(PM_SYS_DEV, 0, callback);
+}
+
+void unset_nmi_pm_callback(struct pm_dev * dev)
+{
+	apic_pm_unregister(dev);
+	nmi_pmdev = apic_pm_register(PM_SYS_DEV, 0, nmi_pm_callback);
+}
+
 static void nmi_pm_init(void)
 {
 	if (!nmi_pmdev)
@@ -178,7 +191,7 @@ static void __pminit setup_k7_watchdog(void)
 		| K7_NMI_EVENT;
 
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
-	printk(KERN_INFO "watchdog: setting K7_PERFCTR0 to %08lx\n", -(cpu_khz/nmi_hz*1000));
+	printk(KERN_INFO "watchdog: setting K7_PERFCTR0 to %08x\n", -(cpu_khz/nmi_hz*1000));
 	wrmsr(MSR_K7_PERFCTR0, -(cpu_khz/nmi_hz*1000), -1);
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
@@ -275,3 +288,30 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 	if (nmi_perfctr_msr)
 		wrmsr(nmi_perfctr_msr, -(cpu_khz/nmi_hz*1000), -1);
 }
+
+static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
+{
+	return 0;
+}
+ 
+static nmi_callback_t nmi_callback = dummy_nmi_callback;
+ 
+asmlinkage void do_nmi(struct pt_regs * regs, long error_code)
+{
+	int cpu = smp_processor_id();
+
+	add_pda(__nmi_count,1);
+
+	if (!nmi_callback(regs, cpu))
+		default_do_nmi(regs);
+}
+
+void set_nmi_callback(nmi_callback_t callback)
+{
+	nmi_callback = callback;
+}
+
+void unset_nmi_callback(void)
+{
+	nmi_callback = dummy_nmi_callback;
+}

arch/x86_64/kernel/process.c

@@ -32,6 +32,7 @@
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
 #include <linux/user.h>
+#include <linux/module.h>
 #include <linux/a.out.h>
 #include <linux/interrupt.h>
 #include <linux/config.h>

arch/x86_64/kernel/profile.c

+/*
+ *	linux/arch/i386/kernel/profile.c
+ *
+ *	(C) 2002 John Levon <levon@movementarian.org>
+ *
+ */
+
+#include <linux/profile.h>
+#include <linux/spinlock.h>
+#include <linux/notifier.h>
+#include <linux/irq.h>
+#include <asm/hw_irq.h> 
+ 
+static struct notifier_block * profile_listeners;
+static rwlock_t profile_lock = RW_LOCK_UNLOCKED;
+ 
+int register_profile_notifier(struct notifier_block * nb)
+{
+	int err;
+	write_lock_irq(&profile_lock);
+	err = notifier_chain_register(&profile_listeners, nb);
+	write_unlock_irq(&profile_lock);
+	return err;
+}
+
+
+int unregister_profile_notifier(struct notifier_block * nb)
+{
+	int err;
+	write_lock_irq(&profile_lock);
+	err = notifier_chain_unregister(&profile_listeners, nb);
+	write_unlock_irq(&profile_lock);
+	return err;
+}
+
+
+void x86_profile_hook(struct pt_regs * regs)
+{
+	/* we would not even need this lock if
+	 * we had a global cli() on register/unregister
+	 */ 
+	read_lock(&profile_lock);
+	notifier_call_chain(&profile_listeners, 0, regs);
+	read_unlock(&profile_lock);
+}

arch/x86_64/kernel/reboot.c

@@ -8,6 +8,7 @@
 #include <asm/io.h>
 #include <asm/kdebug.h>
 #include <asm/delay.h>
+#include <asm/hw_irq.h>
 
 
 /*

arch/x86_64/kernel/setup.c

@@ -234,7 +234,6 @@ static void __init contig_initmem_init(void)
 
 void __init setup_arch(char **cmdline_p)
 {
-	unsigned long bootmap_size, low_mem_size;
 	int i;
 
  	ROOT_DEV = ORIG_ROOT_DEV;
@@ -284,14 +283,8 @@ void __init setup_arch(char **cmdline_p)
 
 	contig_initmem_init(); 
 
-	/*
-	 * Reserve the bootmem bitmap itself as well. We do this in two
-	 * steps (first step was init_bootmem()) because this catches
-	 * the (very unlikely) case of us accidentally initializing the
-	 * bootmem allocator with an invalid RAM area.
-	 */
-	reserve_bootmem(HIGH_MEMORY, (PFN_PHYS(start_pfn) +
-			 bootmap_size + PAGE_SIZE-1) - (HIGH_MEMORY));
+	/* reserve kernel */
+	reserve_bootmem(HIGH_MEMORY, PFN_PHYS(start_pfn) - HIGH_MEMORY); 
 
 	/*
 	 * reserve physical page 0 - it's a special BIOS page on many boxes,

arch/x86_64/kernel/traps.c

@@ -559,11 +559,10 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 	printk("Do you have a strange power saving mode enabled?\n");
 }
 
-asmlinkage void do_nmi(struct pt_regs * regs)
+asmlinkage void default_do_nmi(struct pt_regs * regs)
 {
 	unsigned char reason = inb(0x61);
 
-	add_pda(__nmi_count,1);
 	if (!(reason & 0xc0)) {
 #if CONFIG_X86_LOCAL_APIC
 		/*

arch/x86_64/mm/Makefile

 #
 # Makefile for the linux i386-specific parts of the memory manager.
 #
-# Note! Dependencies are done automagically by 'make dep', which also
-# removes any old dependencies. DON'T put your own dependencies here
-# unless it's something special (ie not a .c file).
-#
-# Note 2! The CFLAGS definition is now in the main makefile...
 
-O_TARGET := mm.o
-obj-y	 := init.o fault.o ioremap.o extable.o modutil.o
+export-objs := pageattr.o
+
+obj-y	 := init.o fault.o ioremap.o extable.o modutil.o pageattr.o
 
 include $(TOPDIR)/Rules.make

arch/x86_64/mm/fault.c

@@ -32,18 +32,14 @@
 
 extern void die(const char *,struct pt_regs *,long);
 
-extern spinlock_t console_lock, timerlist_lock;
+extern spinlock_t console_lock;
 
 void bust_spinlocks(int yes)
 {
- 	spin_lock_init(&timerlist_lock);
+	int loglevel_save = console_loglevel;
 	if (yes) {
 		oops_in_progress = 1;
-#ifdef CONFIG_SMP
-		global_irq_lock = 0;	/* Many serial drivers do __global_cli() */
-#endif
 	} else {
-	int loglevel_save = console_loglevel;
 #ifdef CONFIG_VT
 		unblank_screen();
 #endif
@@ -108,6 +104,18 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	int write;
 	siginfo_t info;
 
+#ifdef CONFIG_CHECKING
+	{ 
+		unsigned long gs; 
+		struct x8664_pda *pda = cpu_pda + stack_smp_processor_id(); 
+		rdmsrl(MSR_GS_BASE, gs); 
+		if (gs != (unsigned long)pda) { 
+			wrmsrl(MSR_GS_BASE, pda); 
+			printk("page_fault: wrong gs %lx expected %p\n", gs, pda);
+		}
+	}
+#endif
+
 	/* get the address */
 	__asm__("movq %%cr2,%0":"=r" (address));
 
@@ -129,7 +137,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	 * If we're in an interrupt or have no user
 	 * context, we must not take the fault..
 	 */
-	if (in_interrupt() || !mm)
+	if (in_atomic() || !mm)
 		goto no_context;
 
  again:
@@ -223,7 +231,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code)
 	/* Are we prepared to handle this kernel fault?  */
 	if ((fixup = search_exception_table(regs->rip)) != 0) {
 		regs->rip = fixup;
-		if (exception_trace) 
+		if (0 && exception_trace) 
 		printk(KERN_ERR 
 		       "%s: fixed kernel exception at %lx address %lx err:%ld\n", 
 		       current->comm, regs->rip, address, error_code);

arch/x86_64/mm/init.c

@@ -37,6 +37,8 @@
 #include <asm/tlb.h>
 #include <asm/mmu_context.h>
 
+unsigned long start_pfn, end_pfn;
+
 mmu_gather_t mmu_gathers[NR_CPUS];
 
 /*
@@ -49,19 +51,24 @@ void show_mem(void)
 {
 	int i, total = 0, reserved = 0;
 	int shared = 0, cached = 0;
+	pg_data_t *pgdat;
+	struct page *page;
 
 	printk("Mem-info:\n");
 	show_free_areas();
 	printk("Free swap:       %6dkB\n",nr_swap_pages<<(PAGE_SHIFT-10));
-	i = max_mapnr;
-	while (i-- > 0) {
+
+	for_each_pgdat(pgdat) {
+               for (i = 0; i < pgdat->node_size; ++i) {
+                       page = pgdat->node_mem_map + i;
 		total++;
-		if (PageReserved(mem_map+i))
+                       if (PageReserved(page))
 			reserved++;
-		else if (PageSwapCache(mem_map+i))
+                       else if (PageSwapCache(page))
 			cached++;
-		else if (page_count(mem_map+i))
-			shared += page_count(mem_map+i) - 1;
+                       else if (page_count(page))
+                               shared += page_count(page) - 1;
+               }
 	}
 	printk("%d pages of RAM\n", total);
 	printk("%d reserved pages\n",reserved);
@@ -264,16 +271,15 @@ void __init paging_init(void)
 {
 	{
 		unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
-		unsigned int max_dma, low;
+		unsigned int max_dma;
 
 		max_dma = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
-		low = max_low_pfn;
 
-		if (low < max_dma)
-			zones_size[ZONE_DMA] = low;
+		if (end_pfn < max_dma)
+			zones_size[ZONE_DMA] = end_pfn;
 		else {
 			zones_size[ZONE_DMA] = max_dma;
-			zones_size[ZONE_NORMAL] = low - max_dma;
+			zones_size[ZONE_NORMAL] = end_pfn - max_dma;
 		}
 		free_area_init(zones_size);
 	}
@@ -308,11 +314,15 @@ void __init mem_init(void)
 	int codesize, reservedpages, datasize, initsize;
 	int tmp;
 
+	printk("mem_init\n");
+
 	if (!mem_map)
 		BUG();
 
-	max_mapnr = num_physpages = max_low_pfn;
-	high_memory = (void *) __va(max_low_pfn * PAGE_SIZE);
+	max_low_pfn = end_pfn;
+	max_pfn = end_pfn;
+	max_mapnr = num_physpages = end_pfn;
+	high_memory = (void *) __va(end_pfn * PAGE_SIZE);
 
 	/* clear the zero-page */
 	memset(empty_zero_page, 0, PAGE_SIZE);
@@ -323,7 +333,7 @@ void __init mem_init(void)
 	after_bootmem = 1;
 
 	reservedpages = 0;
-	for (tmp = 0; tmp < max_low_pfn; tmp++)
+	for (tmp = 0; tmp < end_pfn; tmp++)
 		/*
 		 * Only count reserved RAM pages
 		 */

arch/x86_64/mm/ioremap.c

@@ -10,6 +10,7 @@
 
 #include <linux/vmalloc.h>
 #include <linux/init.h>
+#include <linux/slab.h>
 #include <asm/io.h>
 #include <asm/pgalloc.h>
 #include <asm/fixmap.h>
@@ -157,14 +158,72 @@ void * __ioremap(unsigned long phys_addr, unsigned long size, unsigned long flag
 		return NULL;
 	addr = area->addr;
 	if (remap_area_pages(VMALLOC_VMADDR(addr), phys_addr, size, flags)) {
-		vfree(addr);
+		vunmap(addr);
 		return NULL;
 	}
 	return (void *) (offset + (char *)addr);
 }
 
+/**
+ * ioremap_nocache     -   map bus memory into CPU space
+ * @offset:    bus address of the memory
+ * @size:      size of the resource to map
+ *
+ * ioremap_nocache performs a platform specific sequence of operations to
+ * make bus memory CPU accessible via the readb/readw/readl/writeb/
+ * writew/writel functions and the other mmio helpers. The returned
+ * address is not guaranteed to be usable directly as a virtual
+ * address. 
+ *
+ * This version of ioremap ensures that the memory is marked uncachable
+ * on the CPU as well as honouring existing caching rules from things like
+ * the PCI bus. Note that there are other caches and buffers on many 
+ * busses. In particular driver authors should read up on PCI writes
+ *
+ * It's useful if some control registers are in such an area and
+ * write combining or read caching is not desirable:
+ * 
+ * Must be freed with iounmap.
+ */
+
+void *ioremap_nocache (unsigned long phys_addr, unsigned long size)
+{
+	void *p = __ioremap(phys_addr, size, _PAGE_PCD);
+	if (!p) 
+		return p; 
+
+	if (phys_addr + size < virt_to_phys(high_memory)) { 
+		struct page *ppage = virt_to_page(__va(phys_addr));		
+		unsigned long npages = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;
+
+		BUG_ON(phys_addr+size > (unsigned long)high_memory);
+		BUG_ON(phys_addr + size < phys_addr);
+
+		if (change_page_attr(ppage, npages, PAGE_KERNEL_NOCACHE) < 0) { 
+			iounmap(p); 
+			p = NULL;
+		}
+	} 
+
+	return p;					
+}
+
 void iounmap(void *addr)
 {
-	if (addr > high_memory)
-		vfree((void *) (PAGE_MASK & (unsigned long) addr));
+	struct vm_struct *p;
+	if (addr <= high_memory) 
+		return; 
+	p = remove_vm_area((void *)(PAGE_MASK & (unsigned long) addr)); 
+	if (!p) { 
+		printk("__iounmap: bad address %p\n", addr);
+		return;
+	} 
+
+	unmap_vm_area(p);
+	if (p->flags && p->phys_addr < virt_to_phys(high_memory)) { 
+		change_page_attr(virt_to_page(__va(p->phys_addr)),
+				 p->size >> PAGE_SHIFT,
+				 PAGE_KERNEL); 				 
+	} 
+	kfree(p); 
 }

arch/x86_64/mm/modutil.c

@@ -13,12 +13,15 @@
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
 
 static struct vm_struct * modvmlist = NULL;
 
 void module_unmap (void * addr)
 {
 	struct vm_struct **p, *tmp;
+	int i;
 
 	if (!addr)
 		return;
@@ -29,21 +32,33 @@ void module_unmap (void * addr)
 	for (p = &modvmlist ; (tmp = *p) ; p = &tmp->next) {
 		if (tmp->addr == addr) {
 			*p = tmp->next;
-			vmfree_area_pages(VMALLOC_VMADDR(tmp->addr), tmp->size);
-			kfree(tmp);
-			return;
+			goto found;
 		}
 	}
 	printk("Trying to unmap nonexistent module vm area (%p)\n", addr);
+	return;
+ found:
+	unmap_vm_area(tmp);
+	for (i = 0; i < tmp->nr_pages; i++) {
+		if (unlikely(!tmp->pages[i]))
+			BUG();
+		__free_page(tmp->pages[i]);
+	}
+	
+	kfree(tmp->pages);
+	kfree(tmp);					
 }
 
 void * module_map (unsigned long size)
 {
-	void * addr;
 	struct vm_struct **p, *tmp, *area;
+	struct page **pages;
+	void * addr;
+	unsigned int nr_pages, array_size, i;
 
 	size = PAGE_ALIGN(size);
-	if (!size || size > MODULES_LEN) return NULL;
+	if (!size || size > MODULES_LEN)
+		return NULL;
 		
 	addr = (void *) MODULES_VADDR;
 	for (p = &modvmlist; (tmp = *p) ; p = &tmp->next) {
@@ -51,18 +66,53 @@ void * module_map (unsigned long size)
 			break;
 		addr = (void *) (tmp->size + (unsigned long) tmp->addr);
 	}
-	if ((unsigned long) addr + size >= MODULES_END) return NULL;
+	if ((unsigned long) addr + size >= MODULES_END)
+		return NULL;
 	
 	area = (struct vm_struct *) kmalloc(sizeof(*area), GFP_KERNEL);
-	if (!area) return NULL;
+	if (!area)
+		return NULL;
 	area->size = size + PAGE_SIZE;
 	area->addr = addr;
 	area->next = *p;
+	area->pages = NULL;
+	area->nr_pages = 0;
+	area->phys_addr = 0;
 	*p = area;
 
-	if (vmalloc_area_pages(VMALLOC_VMADDR(addr), size, GFP_KERNEL, PAGE_KERNEL)) {
-		module_unmap(addr);
-		return NULL;
+	nr_pages = size >> PAGE_SHIFT;
+	array_size = (nr_pages * sizeof(struct page *));
+
+	area->nr_pages = nr_pages;
+	area->pages = pages = kmalloc(array_size, GFP_KERNEL);
+	if (!area->pages)
+		goto fail;
+
+	memset(area->pages, 0, array_size);
+
+	for (i = 0; i < area->nr_pages; i++) {
+		area->pages[i] = alloc_page(GFP_KERNEL);
+		if (unlikely(!area->pages[i]))
+			goto fail;
+	}
+	
+	if (map_vm_area(area, PAGE_KERNEL, &pages)) {
+		unmap_vm_area(area);
+		goto fail;
 	}
-	return addr;
+
+	return area->addr;
+
+fail:
+	if (area->pages) {
+		for (i = 0; i < area->nr_pages; i++) {
+			if (area->pages[i])
+				__free_page(area->pages[i]);
+		}
+		kfree(area->pages);
+	}
+	kfree(area);
+
+	return NULL;
 }
+

arch/x86_64/mm/pageattr.c

+/* 
+ * Copyright 2002 Andi Kleen, SuSE Labs. 
+ * Thanks to Ben LaHaise for precious feedback.
+ */ 
+
+#include <linux/config.h>
+#include <linux/mm.h>
+#include <linux/sched.h>
+#include <linux/highmem.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+#include <asm/tlbflush.h>
+#include <asm/io.h>
+
+static inline pte_t *lookup_address(unsigned long address) 
+{ 
+	pgd_t *pgd = pgd_offset_k(address); 
+	pmd_t *pmd;
+	pte_t *pte;
+	if (!pgd || !pgd_present(*pgd))
+		return NULL; 
+	pmd = pmd_offset(pgd, address); 	       
+	if (!pmd_present(*pmd))
+		return NULL; 
+	if (pmd_large(*pmd))
+		return (pte_t *)pmd;
+	pte = pte_offset_kernel(pmd, address);
+	if (pte && !pte_present(*pte))
+		pte = NULL; 
+	return pte;
+} 
+
+static struct page *split_large_page(unsigned long address, pgprot_t prot)
+{ 
+	int i; 
+	unsigned long addr;
+	struct page *base = alloc_pages(GFP_KERNEL, 0);
+	pte_t *pbase;
+	if (!base) 
+		return NULL;
+	address = __pa(address);
+	addr = address & LARGE_PAGE_MASK; 
+	pbase = (pte_t *)page_address(base);
+	for (i = 0; i < PTRS_PER_PTE; i++, addr += PAGE_SIZE) {
+		pbase[i] = pfn_pte(addr >> PAGE_SHIFT, 
+				   addr == address ? prot : PAGE_KERNEL);
+	}
+	return base;
+} 
+
+
+static void flush_kernel_map(void *address) 
+{
+	if (address && cpu_has_clflush) {
+		/* is this worth it? */ 
+		int i;
+		for (i = 0; i < PAGE_SIZE; i += boot_cpu_data.x86_clflush_size) 
+			asm volatile("clflush %0" :: "m" (address + i)); 
+	} else
+		asm volatile("wbinvd":::"memory"); 
+	__flush_tlb_one(address);
+}
+
+/* 
+ * No more special protections in this 2/4MB area - revert to a
+ * large page again. 
+ */
+static inline void revert_page(struct page *kpte_page, unsigned long address)
+{
+       pgd_t *pgd;
+       pmd_t *pmd; 
+       pte_t large_pte; 
+       
+       pgd = pgd_offset_k(address); 
+       if (!pgd) BUG(); 
+       pmd = pmd_offset(pgd, address);
+       if (!pmd) BUG(); 
+       if ((pmd_val(*pmd) & _PAGE_GLOBAL) == 0) BUG(); 
+       
+       large_pte = mk_pte_phys(__pa(address) & LARGE_PAGE_MASK, PAGE_KERNEL_LARGE);
+       set_pte((pte_t *)pmd, large_pte);
+}      
+
+static int
+__change_page_attr(unsigned long address, struct page *page, pgprot_t prot, 
+		   struct page **oldpage) 
+{ 
+	pte_t *kpte; 
+	struct page *kpte_page;
+	unsigned kpte_flags;
+
+	kpte = lookup_address(address);
+	if (!kpte) return 0;
+	kpte_page = virt_to_page(((unsigned long)kpte) & PAGE_MASK);
+	kpte_flags = pte_val(*kpte); 
+	if (pgprot_val(prot) != pgprot_val(PAGE_KERNEL)) { 
+		if ((kpte_flags & _PAGE_PSE) == 0) { 
+			pte_t old = *kpte;
+			pte_t standard = mk_pte(page, PAGE_KERNEL); 
+
+			set_pte(kpte, mk_pte(page, prot)); 
+			if (pte_same(old,standard))
+				atomic_inc(&kpte_page->count);
+		} else {
+			struct page *split = split_large_page(address, prot); 
+			if (!split)
+				return -ENOMEM;
+			set_pte(kpte,mk_pte(split, PAGE_KERNEL));
+		}	
+	} else if ((kpte_flags & _PAGE_PSE) == 0) { 
+		set_pte(kpte, mk_pte(page, PAGE_KERNEL));
+		atomic_dec(&kpte_page->count); 
+	}
+
+	if (atomic_read(&kpte_page->count) == 1) { 
+		*oldpage = kpte_page;
+		revert_page(kpte_page, address);
+	} 
+	return 0;
+} 
+
+static inline void flush_map(unsigned long address)
+{	
+#ifdef CONFIG_SMP 
+	smp_call_function(flush_kernel_map, (void *)address, 1, 1);
+#endif	
+	flush_kernel_map((void *)address);
+}
+
+struct deferred_page { 
+	struct deferred_page *next; 
+	struct page *fpage;
+	unsigned long address;
+}; 
+static struct deferred_page *df_list; /* protected by init_mm.mmap_sem */
+
+static inline void save_page(unsigned long address, struct page *fpage)
+{
+	struct deferred_page *df;
+	df = kmalloc(sizeof(struct deferred_page), GFP_KERNEL); 
+	if (!df) {
+		flush_map(address);
+		__free_page(fpage);
+	} else { 
+		df->next = df_list;
+		df->fpage = fpage;
+		df->address = address;
+		df_list = df;
+	} 			
+}
+
+/*
+ * Change the page attributes of an page in the linear mapping.
+ *
+ * This should be used when a page is mapped with a different caching policy
+ * than write-back somewhere - some CPUs do not like it when mappings with
+ * different caching policies exist. This changes the page attributes of the
+ * in kernel linear mapping too.
+ * 
+ * The caller needs to ensure that there are no conflicting mappings elsewhere.
+ * This function only deals with the kernel linear map.
+ * 
+ * Caller must call global_flush_tlb() after this.
+ */
+int change_page_attr(struct page *page, int numpages, pgprot_t prot)
+{
+	int err = 0; 
+	struct page *fpage, *fpage2; 
+	int i; 
+
+	down_write(&init_mm.mmap_sem);
+	for (i = 0; i < numpages; i++, page++) { 
+		unsigned long address = (unsigned long)page_address(page); 
+		fpage = NULL;
+		err = __change_page_attr(address, page, prot, &fpage); 
+		/* Handle kernel mapping too which aliases part of the lowmem */
+		if (!err && page_to_phys(page) < KERNEL_TEXT_SIZE) {		
+			unsigned long addr2 = __START_KERNEL_map + page_to_phys(page);
+			fpage2 = NULL;
+			err = __change_page_attr(addr2, page, prot, &fpage2); 
+			if (fpage2) 
+				save_page(addr2, fpage2); 		     
+		} 
+		if (fpage)
+			save_page(address, fpage);
+	} 	
+	up_write(&init_mm.mmap_sem); 
+	return err;
+}
+
+void global_flush_tlb(void)
+{ 
+	struct deferred_page *df, *next_df;
+
+	down_read(&init_mm.mmap_sem);
+	df = xchg(&df_list, NULL);
+	up_read(&init_mm.mmap_sem);
+	flush_map((df && !df->next) ? df->address : 0);
+	for (; df; df = next_df) { 
+		next_df = df->next;
+		if (df->fpage) 
+			__free_page(df->fpage);
+		kfree(df);
+	} 
+} 
+
+EXPORT_SYMBOL(change_page_attr);
+EXPORT_SYMBOL(global_flush_tlb);

include/asm-x86_64/hw_irq.h

@@ -20,6 +20,10 @@
 #include <linux/config.h>
 #include <asm/atomic.h>
 #include <asm/irq.h>
+#include <linux/profile.h>
+#include <linux/smp.h>
+
+struct hw_interrupt_type;
 #endif
 
 /*
@@ -126,20 +130,23 @@ __asm__( \
 	"push $" #nr "-256 ; " \
 	"jmp common_interrupt");
 
-extern unsigned long prof_cpu_mask;
-extern unsigned int * prof_buffer;
-extern unsigned long prof_len;
-extern unsigned long prof_shift;
-
-/*
- * x86 profiling function, SMP safe. We might want to do this in
- * assembly totally?
- */
-static inline void x86_do_profile (unsigned long rip)
+static inline void x86_do_profile (struct pt_regs *regs) 
 {
+	unsigned long rip;
+	extern unsigned long prof_cpu_mask;
+	extern char _stext;
+#ifdef CONFIG_PROFILING
+	extern void x86_profile_hook(struct pt_regs *);
+ 
+	x86_profile_hook(regs);
+#endif
+	if (user_mode(regs))
+		return;
 	if (!prof_buffer)
 		return;
 
+	rip = regs->rip;
+
 	/*
 	 * Only measure the CPUs specified by /proc/irq/prof_cpu_mask.
 	 * (default is all CPUs.)
@@ -159,6 +166,11 @@ static inline void x86_do_profile (unsigned long rip)
 	atomic_inc((atomic_t *)&prof_buffer[rip]);
 }
 
+struct notifier_block;
+ 
+int register_profile_notifier(struct notifier_block * nb);
+int unregister_profile_notifier(struct notifier_block * nb);
+
 #ifdef CONFIG_SMP /*more of this file should probably be ifdefed SMP */
 static inline void hw_resend_irq(struct hw_interrupt_type *h, unsigned int i) {
 	if (IO_APIC_IRQ(i))

include/asm-x86_64/ia32.h

@@ -235,11 +235,6 @@ struct iovec32 {
 };
 
 
-#ifdef __KERNEL__
-struct iovec *get_iovec32(struct iovec32 *iov32, struct iovec *iov_buf, u32 count, int type);
-#endif
-
-
 #endif /* !CONFIG_IA32_SUPPORT */
  
 #endif 

include/asm-x86_64/ide.h

@@ -5,7 +5,7 @@
  */
 
 /*
- *  This file contains the i386 architecture specific IDE code.
+ *  This file contains the x86_64 architecture specific IDE code.
  */
 
 #ifndef __ASMx86_64_IDE_H

include/asm-x86_64/io_apic.h

@@ -148,9 +148,18 @@ extern int io_apic_get_redir_entries (int ioapic);
 extern int io_apic_set_pci_routing (int ioapic, int pin, int irq);
 #endif
 
+#ifdef CONFIG_ACPI_BOOT
+extern int io_apic_get_unique_id (int ioapic, int apic_id);
+extern int io_apic_get_version (int ioapic);
+extern int io_apic_get_redir_entries (int ioapic);
+extern int io_apic_set_pci_routing (int ioapic, int pin, int irq);
+#endif
+
 
 #else  /* !CONFIG_X86_IO_APIC */
 #define io_apic_assign_pci_irqs 0
 #endif
 
+void enable_NMI_through_LVT0 (void * dummy);
+
 #endif

include/asm-x86_64/ioctl32.h

+#include <linux/ioctl32.h>

include/asm-x86_64/ipc.h

-#ifndef __x86_64_IPC_H__
-#define __x86_64_IPC_H__
+#ifndef __x8664_IPC_H__
+#define __x8664_IPC_H__
 
 /* dummy */
 

include/asm-x86_64/ipcbuf.h

-#ifndef __i386_IPCBUF_H__
-#define __i386_IPCBUF_H__
+#ifndef __x86_64_IPCBUF_H__
+#define __x86_64_IPCBUF_H__
 
 /*
- * The ipc64_perm structure for i386 architecture.
+ * The ipc64_perm structure for x86_64 architecture.
  * Note extra padding because this structure is passed back and forth
  * between kernel and user space.
  *
@@ -26,4 +26,4 @@ struct ipc64_perm
 	unsigned long		__unused2;
 };
 
-#endif /* __i386_IPCBUF_H__ */
+#endif /* __x86_64_IPCBUF_H__ */

include/asm-x86_64/namei.h

-/* $Id: namei.h,v 1.2 2001/07/04 09:08:13 ak Exp $
- * linux/include/asm-i386/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
 #ifndef __X8664_NAMEI_H
 #define __X8664_NAMEI_H
 

include/asm-x86_64/nmi.h

+/*
+ *  linux/include/asm-i386/nmi.h
+ */
+#ifndef ASM_NMI_H
+#define ASM_NMI_H
+
+#include <linux/pm.h>
+ 
+struct pt_regs;
+ 
+typedef int (*nmi_callback_t)(struct pt_regs * regs, int cpu);
+ 
+/** 
+ * set_nmi_callback
+ *
+ * Set a handler for an NMI. Only one handler may be
+ * set. Return 1 if the NMI was handled.
+ */
+void set_nmi_callback(nmi_callback_t callback);
+ 
+/** 
+ * unset_nmi_callback
+ *
+ * Remove the handler previously set.
+ */
+void unset_nmi_callback(void);
+ 
+#ifdef CONFIG_PM
+ 
+/** Replace the PM callback routine for NMI. */
+struct pm_dev * set_nmi_pm_callback(pm_callback callback);
+
+/** Unset the PM callback routine back to the default. */
+void unset_nmi_pm_callback(struct pm_dev * dev);
+
+#else
+
+static inline struct pm_dev * set_nmi_pm_callback(pm_callback callback)
+{
+	return 0;
+} 
+ 
+static inline void unset_nmi_pm_callback(struct pm_dev * dev)
+{
+}
+
+#endif /* CONFIG_PM */
+ 
+#endif /* ASM_NMI_H */

include/asm-x86_64/param.h

@@ -2,7 +2,7 @@
 #define _ASMx86_64_PARAM_H
 
 #ifdef __KERNEL__
-# define HZ            100            /* Internal kernel timer frequency */
+# define HZ            1000            /* Internal kernel timer frequency */
 # define USER_HZ       100          /* .. some user interfaces are in "ticks */
 #define CLOCKS_PER_SEC        (USER_HZ)       /* like times() */
 #endif

include/asm-x86_64/proto.h

@@ -42,6 +42,7 @@ extern void exception_table_check(void);
 
 extern void acpi_boot_init(char *);
 
+int iommu_setup(char *opt);
 
 #define round_up(x,y) (((x) + (y) - 1) & ~((y)-1))
 #define round_down(x,y) ((x) & ~((y)-1))

include/asm-x86_64/softirq.h

@@ -9,6 +9,8 @@
 #define __local_bh_enable() \
 		do { barrier(); preempt_count() -= SOFTIRQ_OFFSET; } while (0)
 
+void do_softirq(void);
+
 #define local_bh_enable()						\
 do {									\
 	__local_bh_enable();						\

include/asm-x86_64/unistd.h

@@ -482,8 +482,10 @@ __SYSCALL(__NR_io_submit, sys_io_submit)
 __SYSCALL(__NR_io_cancel, sys_io_cancel)
 #define __NR_get_thread_area	211
 __SYSCALL(__NR_get_thread_area, sys_get_thread_area)
+#define __NR_lookup_dcookie	212
+__SYSCALL(__NR_lookup_dcookie, sys_lookup_dcookie)
 
-#define __NR_syscall_max __NR_get_thread_area
+#define __NR_syscall_max __NR_lookup_dcookie
 #ifndef __NO_STUBS
 
 /* user-visible error numbers are in the range -1 - -4095 */
@@ -503,7 +505,6 @@ do { \
 
 #define __syscall "syscall"
 
-/* XXX - _foo needs to be __foo, while __NR_bar could be _NR_bar. */
 #define _syscall0(type,name) \
 type name(void) \
 { \