X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=arch%2Fi386%2Fkernel%2Ftraps.c;h=0b9d741f80d2ce492a166e342994c18b4ad3605c;hb=34a75f0025b9cf803b6a88db032e6ad6950c9313;hp=f031de0588089595d4cdee4c11e0df35c407b455;hpb=9213980e6a70d8473e0ffd4b39ab5b6caaba9ff5;p=linux-2.6.git

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index f031de058..0b9d741f8 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -25,7 +25,9 @@
 #include <linux/highmem.h>
 #include <linux/kallsyms.h>
 #include <linux/ptrace.h>
-#include <linux/version.h>
+#include <linux/utsname.h>
+#include <linux/kprobes.h>
+#include <linux/kexec.h>
 
 #ifdef CONFIG_EISA
 #include <linux/ioport.h>
@@ -47,17 +49,15 @@
 #include <asm/nmi.h>
 
 #include <asm/smp.h>
-#include <asm/pgalloc.h>
 #include <asm/arch_hooks.h>
+#include <asm/kdebug.h>
 
-#include <linux/irq.h>
 #include <linux/module.h>
+#include <linux/vserver/debug.h>
 
 #include "mach_traps.h"
 
 asmlinkage int system_call(void);
-asmlinkage void lcall7(void);
-asmlinkage void lcall27(void);
 
 struct desc_struct default_ldt[] = { { 0, 0 }, { 0, 0 }, { 0, 0 },
 		{ 0, 0 }, { 0, 0 } };
@@ -93,58 +93,88 @@ asmlinkage void spurious_interrupt_bug(void);
 asmlinkage void machine_check(void);
 
 static int kstack_depth_to_print = 24;
+ATOMIC_NOTIFIER_HEAD(i386die_chain);
 
-static int valid_stack_ptr(struct task_struct *task, void *p)
+extern char last_sysfs_file[];
+
+int register_die_notifier(struct notifier_block *nb)
 {
-	if (p <= (void *)task->thread_info)
-		return 0;
-	if (kstack_end(p))
-		return 0;
-	return 1;
+	vmalloc_sync_all();
+	return atomic_notifier_chain_register(&i386die_chain, nb);
 }
+EXPORT_SYMBOL(register_die_notifier);
 
-#ifdef CONFIG_FRAME_POINTER
-void print_context_stack(struct task_struct *task, unsigned long *stack,
-			 unsigned long ebp)
+int unregister_die_notifier(struct notifier_block *nb)
 {
-	unsigned long addr;
+	return atomic_notifier_chain_unregister(&i386die_chain, nb);
+}
+EXPORT_SYMBOL(unregister_die_notifier);
 
-	while (valid_stack_ptr(task, (void *)ebp)) {
-		addr = *(unsigned long *)(ebp + 4);
-		printk(" [<%08lx>] ", addr);
-		print_symbol("%s", addr);
-		printk("\n");
-		ebp = *(unsigned long *)ebp;
-	}
+static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
+{
+	return	p > (void *)tinfo &&
+		p < (void *)tinfo + THREAD_SIZE - 3;
 }
+
+/*
+ * Print CONFIG_STACK_BACKTRACE_COLS address/symbol entries per line.
+ */
+static inline int print_addr_and_symbol(unsigned long addr, char *log_lvl,
+					int printed)
+{
+	if (!printed)
+		printk(log_lvl);
+
+#if CONFIG_STACK_BACKTRACE_COLS == 1
+	printk(" [<%08lx>] ", addr);
 #else
-void print_context_stack(struct task_struct *task, unsigned long *stack,
-			 unsigned long ebp)
+	printk(" <%08lx> ", addr);
+#endif
+	print_symbol("%s", addr);
+
+	printed = (printed + 1) % CONFIG_STACK_BACKTRACE_COLS;
+	if (printed)
+		printk(" ");
+	else
+		printk("\n");
+
+	return printed;
+}
+
+static inline unsigned long print_context_stack(struct thread_info *tinfo,
+				unsigned long *stack, unsigned long ebp,
+				char *log_lvl)
 {
 	unsigned long addr;
+	int printed = 0; /* nr of entries already printed on current line */
 
-	while (!kstack_end(stack)) {
+#ifdef	CONFIG_FRAME_POINTER
+	while (valid_stack_ptr(tinfo, (void *)ebp)) {
+		addr = *(unsigned long *)(ebp + 4);
+		printed = print_addr_and_symbol(addr, log_lvl, printed);
+		ebp = *(unsigned long *)ebp;
+	}
+#else
+	while (valid_stack_ptr(tinfo, stack)) {
 		addr = *stack++;
-		if (kernel_text_address(addr)) {
-			printk(" [<%08lx>] ", addr);
-			print_symbol("%s\n", addr);
-		}
+		if (__kernel_text_address(addr))
+			printed = print_addr_and_symbol(addr, log_lvl, printed);
 	}
-}
 #endif
+	if (printed)
+		printk("\n");
 
-void show_trace(struct task_struct *task, unsigned long * stack)
+	return ebp;
+}
+
+static void show_trace_log_lvl(struct task_struct *task,
+			       unsigned long *stack, char *log_lvl)
 {
 	unsigned long ebp;
 
 	if (!task)
 		task = current;
 
-	if (!valid_stack_ptr(task, stack)) {
-		printk("Stack pointer is garbage, not printing trace\n");
-		return;
-	}
-
 	if (task == current) {
 		/* Grab ebp right from our regs */
 		asm ("movl %%ebp, %0" : "=r" (ebp) : );
@@ -157,16 +187,21 @@ void show_trace(struct task_struct *task, unsigned long * stack)
 		struct thread_info *context;
 		context = (struct thread_info *)
 			((unsigned long)stack & (~(THREAD_SIZE - 1)));
-		print_context_stack(task, stack, ebp);
+		ebp = print_context_stack(context, stack, ebp, log_lvl);
 		stack = (unsigned long*)context->previous_esp;
 		if (!stack)
 			break;
-		printk(" =======================\n");
+		printk("%s =======================\n", log_lvl);
 	}
-	printk("\n");
 }
 
-void show_stack(struct task_struct *task, unsigned long *esp)
+void show_trace(struct task_struct *task, unsigned long * stack)
+{
+	show_trace_log_lvl(task, stack, "");
+}
+
+static void show_stack_log_lvl(struct task_struct *task, unsigned long *esp,
+			       char *log_lvl)
 {
 	unsigned long *stack;
 	int i;
@@ -183,11 +218,17 @@ void show_stack(struct task_struct *task, unsigned long *esp)
 		if (kstack_end(stack))
 			break;
 		if (i && ((i % 8) == 0))
-			printk("\n       ");
+			printk("\n%s       ", log_lvl);
 		printk("%08lx ", *stack++);
 	}
-	printk("\nCall Trace:\n");
-	show_trace(task, esp);
+	printk("\n%sCall Trace:\n", log_lvl);
+	show_trace_log_lvl(task, esp, log_lvl);
+}
+
+void show_stack(struct task_struct *task, unsigned long *esp)
+{
+	printk("       ");
+	show_stack_log_lvl(task, esp, "");
 }
 
 /*
@@ -210,48 +251,53 @@ void show_registers(struct pt_regs *regs)
 	unsigned short ss;
 
 	esp = (unsigned long) (&regs->esp);
-	ss = __KERNEL_DS;
-	if (regs->xcs & 3) {
+	savesegment(ss, ss);
+	if (user_mode_vm(regs)) {
 		in_kernel = 0;
 		esp = regs->esp;
 		ss = regs->xss & 0xffff;
 	}
 	print_modules();
-	printk("CPU:    %d\nEIP:    %04x:[<%08lx>]    %s\nEFLAGS: %08lx"
-			"   (%s) \n",
+	printk(KERN_EMERG "CPU:    %d\nEIP:    %04x:[<%08lx>]    %s VLI\n"
+			"EFLAGS: %08lx   (%s %.*s) \n",
 		smp_processor_id(), 0xffff & regs->xcs, regs->eip,
-		print_tainted(), regs->eflags, UTS_RELEASE);
-	print_symbol("EIP is at %s\n", regs->eip);
-	printk("eax: %08lx   ebx: %08lx   ecx: %08lx   edx: %08lx\n",
+		print_tainted(), regs->eflags, system_utsname.release,
+		(int)strcspn(system_utsname.version, " "),
+		system_utsname.version);
+	print_symbol(KERN_EMERG "EIP is at %s\n", regs->eip);
+	printk(KERN_EMERG "eax: %08lx   ebx: %08lx   ecx: %08lx   edx: %08lx\n",
 		regs->eax, regs->ebx, regs->ecx, regs->edx);
-	printk("esi: %08lx   edi: %08lx   ebp: %08lx   esp: %08lx\n",
+	printk(KERN_EMERG "esi: %08lx   edi: %08lx   ebp: %08lx   esp: %08lx\n",
 		regs->esi, regs->edi, regs->ebp, esp);
-	printk("ds: %04x   es: %04x   ss: %04x\n",
+	printk(KERN_EMERG "ds: %04x   es: %04x   ss: %04x\n",
 		regs->xds & 0xffff, regs->xes & 0xffff, ss);
-	printk("Process %s (pid: %d, threadinfo=%p task=%p)",
-		current->comm, current->pid, current_thread_info(), current);
+	printk(KERN_EMERG "Process %s (pid: %d[#%u], threadinfo=%p task=%p)",
+		current->comm, current->pid, current->xid,
+		current_thread_info(), current);
 	/*
 	 * When in-kernel, we also print out the stack and code at the
 	 * time of the fault..
 	 */
 	if (in_kernel) {
+		u8 __user *eip;
 
-		printk("\nStack: ");
-		show_stack(NULL, (unsigned long*)esp);
+		printk("\n" KERN_EMERG "Stack: ");
+		show_stack_log_lvl(NULL, (unsigned long *)esp, KERN_EMERG);
 
-		printk("Code: ");
-		if(regs->eip < PAGE_OFFSET)
-			goto bad;
+		printk(KERN_EMERG "Code: ");
 
-		for(i=0;i<20;i++)
-		{
+		eip = (u8 __user *)regs->eip - 43;
+		for (i = 0; i < 64; i++, eip++) {
 			unsigned char c;
-			if(__get_user(c, &((unsigned char*)regs->eip)[i])) {
-bad:
+
+			if (eip < (u8 __user *)PAGE_OFFSET || __get_user(c, eip)) {
 				printk(" Bad EIP value.");
 				break;
 			}
-			printk("%02x ", c);
+			if (eip == (u8 __user *)regs->eip)
+				printk("<%02x> ", c);
+			else
+				printk("%02x ", c);
 		}
 	}
 	printk("\n");
@@ -265,106 +311,156 @@ static void handle_BUG(struct pt_regs *regs)
 	char c;
 	unsigned long eip;
 
-	if (regs->xcs & 3)
-		goto no_bug;		/* Not in kernel */
-
 	eip = regs->eip;
 
 	if (eip < PAGE_OFFSET)
 		goto no_bug;
-	if (__get_user(ud2, (unsigned short *)eip))
+	if (__get_user(ud2, (unsigned short __user *)eip))
 		goto no_bug;
 	if (ud2 != 0x0b0f)
 		goto no_bug;
-	if (__get_user(line, (unsigned short *)(eip + 2)))
+	if (__get_user(line, (unsigned short __user *)(eip + 2)))
 		goto bug;
-	if (__get_user(file, (char **)(eip + 4)) ||
+	if (__get_user(file, (char * __user *)(eip + 4)) ||
 		(unsigned long)file < PAGE_OFFSET || __get_user(c, file))
 		file = "<bad filename>";
 
-	printk("------------[ cut here ]------------\n");
-	printk(KERN_ALERT "kernel BUG at %s:%d!\n", file, line);
+	printk(KERN_EMERG "------------[ cut here ]------------\n");
+	printk(KERN_EMERG "kernel BUG at %s:%d!\n", file, line);
 
 no_bug:
 	return;
 
 	/* Here we know it was a BUG but file-n-line is unavailable */
 bug:
-	printk("Kernel BUG\n");
+	printk(KERN_EMERG "Kernel BUG\n");
 }
 
-spinlock_t die_lock = SPIN_LOCK_UNLOCKED;
-
+/* This is gone through when something in the kernel
+ * has done something bad and is about to be terminated.
+*/
 void die(const char * str, struct pt_regs * regs, long err)
 {
+	static struct {
+		spinlock_t lock;
+		u32 lock_owner;
+		int lock_owner_depth;
+	} die = {
+		.lock =			SPIN_LOCK_UNLOCKED,
+		.lock_owner =		-1,
+		.lock_owner_depth =	0
+	};
 	static int die_counter;
-	int nl = 0;
+	unsigned long flags;
 
-	console_verbose();
-	spin_lock_irq(&die_lock);
-	bust_spinlocks(1);
-	handle_BUG(regs);
-	printk(KERN_ALERT "%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter);
+	oops_enter();
+
+	vxh_throw_oops();
+
+	if (die.lock_owner != raw_smp_processor_id()) {
+		console_verbose();
+		spin_lock_irqsave(&die.lock, flags);
+		die.lock_owner = smp_processor_id();
+		die.lock_owner_depth = 0;
+		bust_spinlocks(1);
+	}
+	else
+		local_save_flags(flags);
+
+	if (++die.lock_owner_depth < 3) {
+		int nl = 0;
+		unsigned long esp;
+		unsigned short ss;
+
+		handle_BUG(regs);
+		printk(KERN_EMERG "%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter);
 #ifdef CONFIG_PREEMPT
-	printk("PREEMPT ");
-	nl = 1;
+		printk(KERN_EMERG "PREEMPT ");
+		nl = 1;
 #endif
 #ifdef CONFIG_SMP
-	printk("SMP ");
-	nl = 1;
+		if (!nl)
+			printk(KERN_EMERG);
+		printk("SMP ");
+		nl = 1;
 #endif
 #ifdef CONFIG_DEBUG_PAGEALLOC
-	printk("DEBUG_PAGEALLOC");
-	nl = 1;
+		if (!nl)
+			printk(KERN_EMERG);
+		printk("DEBUG_PAGEALLOC");
+		nl = 1;
 #endif
-	if (nl)
-		printk("\n");
-	show_registers(regs);
+		if (nl)
+			printk("\n");
+#ifdef CONFIG_SYSFS
+		printk(KERN_ALERT "last sysfs file: %s\n", last_sysfs_file);
+#endif
+		if (notify_die(DIE_OOPS, str, regs, err,
+			current->thread.trap_no, SIGSEGV) != NOTIFY_STOP) {
+			show_registers(regs);
+			vxh_dump_history();
+			/* Executive summary in case the oops scrolled away */
+			esp = (unsigned long) (&regs->esp);
+			savesegment(ss, ss);
+			if (user_mode(regs)) {
+				esp = regs->esp;
+				ss = regs->xss & 0xffff;
+			}
+			printk(KERN_EMERG "EIP: [<%08lx>] ", regs->eip);
+			print_symbol("%s", regs->eip);
+			printk(" SS:ESP %04x:%08lx\n", ss, esp);
+		}
+		else
+			regs = NULL;
+  	} else
+		printk(KERN_EMERG "Recursive die() failure, output suppressed\n");
+
 	bust_spinlocks(0);
-	spin_unlock_irq(&die_lock);
+	die.lock_owner = -1;
+	spin_unlock_irqrestore(&die.lock, flags);
+
+	if (!regs)
+		return;
+
+	if (kexec_should_crash(current))
+		crash_kexec(regs);
+
 	if (in_interrupt())
 		panic("Fatal exception in interrupt");
 
 	if (panic_on_oops) {
 		printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n");
-		set_current_state(TASK_UNINTERRUPTIBLE);
-		schedule_timeout(5 * HZ);
+		ssleep(5);
 		panic("Fatal exception");
 	}
+	oops_exit();
 	do_exit(SIGSEGV);
 }
 
 static inline void die_if_kernel(const char * str, struct pt_regs * regs, long err)
 {
-	if (!(regs->eflags & VM_MASK) && !(3 & regs->xcs))
+	if (!user_mode_vm(regs))
 		die(str, regs, err);
 }
 
-static inline unsigned long get_cr2(void)
+static void __kprobes do_trap(int trapnr, int signr, char *str, int vm86,
+			      struct pt_regs * regs, long error_code,
+			      siginfo_t *info)
 {
-	unsigned long address;
-
-	/* get the address */
-	__asm__("movl %%cr2,%0":"=r" (address));
-	return address;
-}
+	struct task_struct *tsk = current;
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = trapnr;
 
-static inline void do_trap(int trapnr, int signr, char *str, int vm86,
-			   struct pt_regs * regs, long error_code, siginfo_t *info)
-{
 	if (regs->eflags & VM_MASK) {
 		if (vm86)
 			goto vm86_trap;
 		goto trap_signal;
 	}
 
-	if (!(regs->xcs & 3))
+	if (!user_mode(regs))
 		goto kernel_trap;
 
 	trap_signal: {
-		struct task_struct *tsk = current;
-		tsk->thread.error_code = error_code;
-		tsk->thread.trap_no = trapnr;
 		if (info)
 			force_sig_info(signr, info, tsk);
 		else
@@ -386,80 +482,222 @@ static inline void do_trap(int trapnr, int signr, char *str, int vm86,
 }
 
 #define DO_ERROR(trapnr, signr, str, name) \
-asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
+fastcall void do_##name(struct pt_regs * regs, long error_code) \
 { \
+	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+						== NOTIFY_STOP) \
+		return; \
 	do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \
 }
 
 #define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
-asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
+fastcall void do_##name(struct pt_regs * regs, long error_code) \
 { \
 	siginfo_t info; \
 	info.si_signo = signr; \
 	info.si_errno = 0; \
 	info.si_code = sicode; \
-	info.si_addr = (void *)siaddr; \
+	info.si_addr = (void __user *)siaddr; \
+	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+						== NOTIFY_STOP) \
+		return; \
 	do_trap(trapnr, signr, str, 0, regs, error_code, &info); \
 }
 
 #define DO_VM86_ERROR(trapnr, signr, str, name) \
-asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
+fastcall void do_##name(struct pt_regs * regs, long error_code) \
 { \
+	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+						== NOTIFY_STOP) \
+		return; \
 	do_trap(trapnr, signr, str, 1, regs, error_code, NULL); \
 }
 
 #define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
-asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
+fastcall void do_##name(struct pt_regs * regs, long error_code) \
 { \
 	siginfo_t info; \
 	info.si_signo = signr; \
 	info.si_errno = 0; \
 	info.si_code = sicode; \
-	info.si_addr = (void *)siaddr; \
+	info.si_addr = (void __user *)siaddr; \
+	if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
+						== NOTIFY_STOP) \
+		return; \
 	do_trap(trapnr, signr, str, 1, regs, error_code, &info); \
 }
 
 DO_VM86_ERROR_INFO( 0, SIGFPE,  "divide error", divide_error, FPE_INTDIV, regs->eip)
+#ifndef CONFIG_KPROBES
 DO_VM86_ERROR( 3, SIGTRAP, "int3", int3)
+#endif
 DO_VM86_ERROR( 4, SIGSEGV, "overflow", overflow)
 DO_VM86_ERROR( 5, SIGSEGV, "bounds", bounds)
-DO_ERROR_INFO( 6, SIGILL,  "invalid operand", invalid_op, ILL_ILLOPN, regs->eip)
+DO_ERROR_INFO( 6, SIGILL,  "invalid opcode", invalid_op, ILL_ILLOPN, regs->eip)
 DO_ERROR( 9, SIGFPE,  "coprocessor segment overrun", coprocessor_segment_overrun)
 DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
 DO_ERROR(11, SIGBUS,  "segment not present", segment_not_present)
 DO_ERROR(12, SIGBUS,  "stack segment", stack_segment)
-DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, get_cr2())
+DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
+
 
-asmlinkage void do_general_protection(struct pt_regs * regs, long error_code)
+/*
+ * lazy-check for CS validity on exec-shield binaries:
+ *
+ * the original non-exec stack patch was written by
+ * Solar Designer <solar at openwall.com>. Thanks!
+ */
+static int
+check_lazy_exec_limit(int cpu, struct pt_regs *regs, long error_code)
 {
-	if (regs->eflags & X86_EFLAGS_IF)
-		local_irq_enable();
- 
+	struct desc_struct *desc1, *desc2;
+	struct vm_area_struct *vma;
+	unsigned long limit;
+
+	if (current->mm == NULL)
+		return 0;
+
+	limit = -1UL;
+	if (current->mm->context.exec_limit != -1UL) {
+		limit = PAGE_SIZE;
+		spin_lock(&current->mm->page_table_lock);
+		for (vma = current->mm->mmap; vma; vma = vma->vm_next)
+			if ((vma->vm_flags & VM_EXEC) && (vma->vm_end > limit))
+				limit = vma->vm_end;
+		spin_unlock(&current->mm->page_table_lock);
+		if (limit >= TASK_SIZE)
+			limit = -1UL;
+		current->mm->context.exec_limit = limit;
+	}
+	set_user_cs(&current->mm->context.user_cs, limit);
+
+	desc1 = &current->mm->context.user_cs;
+	desc2 = get_cpu_gdt_table(cpu) + GDT_ENTRY_DEFAULT_USER_CS;
+
+	if (desc1->a != desc2->a || desc1->b != desc2->b) {
+		/*
+		 * The CS was not in sync - reload it and retry the
+		 * instruction. If the instruction still faults then
+		 * we won't hit this branch next time around.
+		 */
+		if (print_fatal_signals >= 2) {
+			printk("#GPF fixup (%ld[seg:%lx]) at %08lx, CPU#%d.\n", error_code, error_code/8, regs->eip, smp_processor_id());
+			printk(" exec_limit: %08lx, user_cs: %08lx/%08lx, CPU_cs: %08lx/%08lx.\n", current->mm->context.exec_limit, desc1->a, desc1->b, desc2->a, desc2->b);
+		}
+		load_user_cs_desc(cpu, current->mm);
+		return 1;
+	}
+
+	return 0;
+}
+
+/*
+ * The fixup code for errors in iret jumps to here (iret_exc).  It loses
+ * the original trap number and error code.  The bogus trap 32 and error
+ * code 0 are what the vanilla kernel delivers via:
+ * DO_ERROR_INFO(32, SIGSEGV, "iret exception", iret_error, ILL_BADSTK, 0)
+ *
+ * In case of a general protection fault in the iret instruction, we
+ * need to check for a lazy CS update for exec-shield.
+ */
+fastcall void do_iret_error(struct pt_regs *regs, long error_code)
+{
+	int ok = check_lazy_exec_limit(get_cpu(), regs, error_code);
+	put_cpu();
+	if (!ok && notify_die(DIE_TRAP, "iret exception", regs,
+			      error_code, 32, SIGSEGV) != NOTIFY_STOP) {
+		siginfo_t info;
+		info.si_signo = SIGSEGV;
+		info.si_errno = 0;
+		info.si_code = ILL_BADSTK;
+		info.si_addr = 0;
+		do_trap(32, SIGSEGV, "iret exception", 0, regs, error_code,
+			&info);
+	}
+}
+
+fastcall void __kprobes do_general_protection(struct pt_regs * regs,
+					      long error_code)
+{
+	int cpu = get_cpu();
+	struct tss_struct *tss = &per_cpu(init_tss, cpu);
+	struct thread_struct *thread = &current->thread;
+	int ok;
+
+	/*
+	 * Perform the lazy TSS's I/O bitmap copy. If the TSS has an
+	 * invalid offset set (the LAZY one) and the faulting thread has
+	 * a valid I/O bitmap pointer, we copy the I/O bitmap in the TSS
+	 * and we set the offset field correctly. Then we let the CPU to
+	 * restart the faulting instruction.
+	 */
+	if (tss->io_bitmap_base == INVALID_IO_BITMAP_OFFSET_LAZY &&
+	    thread->io_bitmap_ptr) {
+		memcpy(tss->io_bitmap, thread->io_bitmap_ptr,
+		       thread->io_bitmap_max);
+		/*
+		 * If the previously set map was extending to higher ports
+		 * than the current one, pad extra space with 0xff (no access).
+		 */
+		if (thread->io_bitmap_max < tss->io_bitmap_max)
+			memset((char *) tss->io_bitmap +
+				thread->io_bitmap_max, 0xff,
+				tss->io_bitmap_max - thread->io_bitmap_max);
+		tss->io_bitmap_max = thread->io_bitmap_max;
+		tss->io_bitmap_base = IO_BITMAP_OFFSET;
+		tss->io_bitmap_owner = thread;
+		put_cpu();
+		return;
+	}
+
+	current->thread.error_code = error_code;
+	current->thread.trap_no = 13;
+
 	if (regs->eflags & VM_MASK)
 		goto gp_in_vm86;
 
-	if (!(regs->xcs & 3))
+	if (!user_mode(regs))
 		goto gp_in_kernel;
 
+	ok = check_lazy_exec_limit(cpu, regs, error_code);
+
+	put_cpu();
+
+	if (ok)
+		return;
+
+	if (print_fatal_signals) {
+		printk("#GPF(%ld[seg:%lx]) at %08lx, CPU#%d.\n", error_code, error_code/8, regs->eip, smp_processor_id());
+		printk(" exec_limit: %08lx, user_cs: %08lx/%08lx.\n", current->mm->context.exec_limit, current->mm->context.user_cs.a, current->mm->context.user_cs.b);
+	}
+
 	current->thread.error_code = error_code;
 	current->thread.trap_no = 13;
 	force_sig(SIGSEGV, current);
 	return;
 
 gp_in_vm86:
+	put_cpu();
 	local_irq_enable();
 	handle_vm86_fault((struct kernel_vm86_regs *) regs, error_code);
 	return;
 
 gp_in_kernel:
-	if (!fixup_exception(regs))
+	put_cpu();
+	if (!fixup_exception(regs)) {
+		if (notify_die(DIE_GPF, "general protection fault", regs,
+				error_code, 13, SIGSEGV) == NOTIFY_STOP)
+			return;
 		die("general protection fault", regs, error_code);
+	}
 }
 
 static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
 {
-	printk("Uhhuh. NMI received. Dazed and confused, but trying to continue\n");
-	printk("You probably have a hardware problem with your RAM chips\n");
+	printk(KERN_EMERG "Uhhuh. NMI received. Dazed and confused, but trying "
+			"to continue\n");
+	printk(KERN_EMERG "You probably have a hardware problem with your RAM "
+			"chips\n");
 
 	/* Clear and disable the memory parity error line. */
 	clear_mem_error(reason);
@@ -469,7 +707,7 @@ static void io_check_error(unsigned char reason, struct pt_regs * regs)
 {
 	unsigned long i;
 
-	printk("NMI: IOCK error (debug interrupt?)\n");
+	printk(KERN_EMERG "NMI: IOCK error (debug interrupt?)\n");
 	show_registers(regs);
 
 	/* Re-enable the IOCK line, wait for a few seconds */
@@ -497,11 +735,52 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 	printk("Do you have a strange power saving mode enabled?\n");
 }
 
+static DEFINE_SPINLOCK(nmi_print_lock);
+
+void die_nmi (struct pt_regs *regs, const char *msg)
+{
+	if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) ==
+	    NOTIFY_STOP)
+		return;
+
+	spin_lock(&nmi_print_lock);
+	/*
+	* We are in trouble anyway, lets at least try
+	* to get a message out.
+	*/
+	bust_spinlocks(1);
+	printk(KERN_EMERG "%s", msg);
+	printk(" on CPU%d, eip %08lx, registers:\n",
+		smp_processor_id(), regs->eip);
+	show_registers(regs);
+	printk(KERN_EMERG "console shuts up ...\n");
+	console_silent();
+	spin_unlock(&nmi_print_lock);
+	bust_spinlocks(0);
+
+	/* If we are in kernel we are probably nested up pretty bad
+	 * and might aswell get out now while we still can.
+	*/
+	if (!user_mode_vm(regs)) {
+		current->thread.trap_no = 2;
+		crash_kexec(regs);
+	}
+
+	do_exit(SIGSEGV);
+}
+
 static void default_do_nmi(struct pt_regs * regs)
 {
-	unsigned char reason = get_nmi_reason();
+	unsigned char reason = 0;
+
+	/* Only the BSP gets external NMIs from the system.  */
+	if (!smp_processor_id())
+		reason = get_nmi_reason();
  
 	if (!(reason & 0xc0)) {
+		if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
+							== NOTIFY_STOP)
+			return;
 #ifdef CONFIG_X86_LOCAL_APIC
 		/*
 		 * Ok, so this is none of the documented NMI sources,
@@ -515,6 +794,8 @@ static void default_do_nmi(struct pt_regs * regs)
 		unknown_nmi_error(reason, regs);
 		return;
 	}
+	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
+		return;
 	if (reason & 0x80)
 		mem_parity_error(reason, regs);
 	if (reason & 0x40)
@@ -533,16 +814,17 @@ static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
  
 static nmi_callback_t nmi_callback = dummy_nmi_callback;
  
-asmlinkage void do_nmi(struct pt_regs * regs, long error_code)
+fastcall void do_nmi(struct pt_regs * regs, long error_code)
 {
 	int cpu;
 
 	nmi_enter();
 
 	cpu = smp_processor_id();
+
 	++nmi_count(cpu);
 
-	if (!nmi_callback(regs, cpu))
+	if (!rcu_dereference(nmi_callback)(regs, cpu))
 		default_do_nmi(regs);
 
 	nmi_exit();
@@ -550,13 +832,29 @@ asmlinkage void do_nmi(struct pt_regs * regs, long error_code)
 
 void set_nmi_callback(nmi_callback_t callback)
 {
-	nmi_callback = callback;
+	vmalloc_sync_all();
+	rcu_assign_pointer(nmi_callback, callback);
 }
+EXPORT_SYMBOL_GPL(set_nmi_callback);
 
 void unset_nmi_callback(void)
 {
 	nmi_callback = dummy_nmi_callback;
 }
+EXPORT_SYMBOL_GPL(unset_nmi_callback);
+
+#ifdef CONFIG_KPROBES
+fastcall void __kprobes do_int3(struct pt_regs *regs, long error_code)
+{
+	if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP)
+			== NOTIFY_STOP)
+		return;
+	/* This is an interrupt gate, because kprobes wants interrupts
+	disabled.  Normal trap handlers don't. */
+	restore_interrupts(regs);
+	do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL);
+}
+#endif
 
 /*
  * Our handling of the processor debug registers is non-trivial.
@@ -580,14 +878,16 @@ void unset_nmi_callback(void)
  * find every occurrence of the TF bit that could be saved away even
  * by user code)
  */
-asmlinkage void do_debug(struct pt_regs * regs, long error_code)
+fastcall void __kprobes do_debug(struct pt_regs * regs, long error_code)
 {
 	unsigned int condition;
 	struct task_struct *tsk = current;
-	siginfo_t info;
 
-	__asm__ __volatile__("movl %%db6,%0" : "=r" (condition));
+	get_debugreg(condition, 6);
 
+	if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
+					SIGTRAP) == NOTIFY_STOP)
+		return;
 	/* It's safe to allow irq's after DR6 has been saved */
 	if (regs->eflags & X86_EFLAGS_IF)
 		local_irq_enable();
@@ -604,44 +904,28 @@ asmlinkage void do_debug(struct pt_regs * regs, long error_code)
 	/* Save debug status register where ptrace can see it */
 	tsk->thread.debugreg[6] = condition;
 
-	/* Mask out spurious TF errors due to lazy TF clearing */
+	/*
+	 * Single-stepping through TF: make sure we ignore any events in
+	 * kernel space (but re-enable TF when returning to user mode).
+	 */
 	if (condition & DR_STEP) {
 		/*
-		 * The TF error should be masked out only if the current
-		 * process is not traced and if the TRAP flag has been set
-		 * previously by a tracing process (condition detected by
-		 * the PT_DTRACE flag); remember that the i386 TRAP flag
-		 * can be modified by the process itself in user mode,
-		 * allowing programs to debug themselves without the ptrace()
-		 * interface.
+		 * We already checked v86 mode above, so we can
+		 * check for kernel mode by just checking the CPL
+		 * of CS.
 		 */
-		if ((regs->xcs & 3) == 0)
+		if (!user_mode(regs))
 			goto clear_TF_reenable;
-		if ((tsk->ptrace & (PT_DTRACE|PT_PTRACED)) == PT_DTRACE)
-			goto clear_TF;
 	}
 
 	/* Ok, finally something we can handle */
-	tsk->thread.trap_no = 1;
-	tsk->thread.error_code = error_code;
-	info.si_signo = SIGTRAP;
-	info.si_errno = 0;
-	info.si_code = TRAP_BRKPT;
-	
-	/* If this is a kernel mode trap, save the user PC on entry to 
-	 * the kernel, that's what the debugger can make sense of.
-	 */
-	info.si_addr = ((regs->xcs & 3) == 0) ? (void *)tsk->thread.eip : 
-	                                        (void *)regs->eip;
-	force_sig_info(SIGTRAP, &info, tsk);
+	send_sigtrap(tsk, regs, error_code);
 
 	/* Disable additional traps. They'll be re-enabled when
 	 * the signal is delivered.
 	 */
 clear_dr7:
-	__asm__("movl %0,%%db7"
-		: /* no output */
-		: "r" (0));
+	set_debugreg(0, 7);
 	return;
 
 debug_vm86:
@@ -650,7 +934,6 @@ debug_vm86:
 
 clear_TF_reenable:
 	set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
-clear_TF:
 	regs->eflags &= ~TF_MASK;
 	return;
 }
@@ -660,7 +943,7 @@ clear_TF:
  * the correct behaviour even in the presence of the asynchronous
  * IRQ13 behaviour
  */
-void math_error(void *eip)
+void math_error(void __user *eip)
 {
 	struct task_struct * task;
 	siginfo_t info;
@@ -689,15 +972,18 @@ void math_error(void *eip)
 	 */
 	cwd = get_fpu_cwd(task);
 	swd = get_fpu_swd(task);
-	switch (((~cwd) & swd & 0x3f) | (swd & 0x240)) {
-		case 0x000:
-		default:
+	switch (swd & ~cwd & 0x3f) {
+		case 0x000: /* No unmasked exception */
+			return;
+		default:    /* Multiple exceptions */
 			break;
 		case 0x001: /* Invalid Op */
-		case 0x041: /* Stack Fault */
-		case 0x241: /* Stack Fault | Direction */
+			/*
+			 * swd & 0x240 == 0x040: Stack Underflow
+			 * swd & 0x240 == 0x240: Stack Overflow
+			 * User must clear the SF bit (0x40) if set
+			 */
 			info.si_code = FPE_FLTINV;
-			/* Should we clear the SF or let user space do it ???? */
 			break;
 		case 0x002: /* Denormalize */
 		case 0x010: /* Underflow */
@@ -716,13 +1002,13 @@ void math_error(void *eip)
 	force_sig_info(SIGFPE, &info, task);
 }
 
-asmlinkage void do_coprocessor_error(struct pt_regs * regs, long error_code)
+fastcall void do_coprocessor_error(struct pt_regs * regs, long error_code)
 {
 	ignore_fpu_irq = 1;
-	math_error((void *)regs->eip);
+	math_error((void __user *)regs->eip);
 }
 
-void simd_math_error(void *eip)
+static void simd_math_error(void __user *eip)
 {
 	struct task_struct * task;
 	siginfo_t info;
@@ -770,13 +1056,13 @@ void simd_math_error(void *eip)
 	force_sig_info(SIGFPE, &info, task);
 }
 
-asmlinkage void do_simd_coprocessor_error(struct pt_regs * regs,
+fastcall void do_simd_coprocessor_error(struct pt_regs * regs,
 					  long error_code)
 {
 	if (cpu_has_xmm) {
 		/* Handle SIMD FPU exceptions on PIII+ processors. */
 		ignore_fpu_irq = 1;
-		simd_math_error((void *)regs->eip);
+		simd_math_error((void __user *)regs->eip);
 	} else {
 		/*
 		 * Handle strange cache flush from user space exception
@@ -787,14 +1073,14 @@ asmlinkage void do_simd_coprocessor_error(struct pt_regs * regs,
 					  error_code);
 			return;
 		}
-		die_if_kernel("cache flush denied", regs, error_code);
 		current->thread.trap_no = 19;
 		current->thread.error_code = error_code;
+		die_if_kernel("cache flush denied", regs, error_code);
 		force_sig(SIGSEGV, current);
 	}
 }
 
-asmlinkage void do_spurious_interrupt_bug(struct pt_regs * regs,
+fastcall void do_spurious_interrupt_bug(struct pt_regs * regs,
 					  long error_code)
 {
 #if 0
@@ -803,6 +1089,51 @@ asmlinkage void do_spurious_interrupt_bug(struct pt_regs * regs,
 #endif
 }
 
+fastcall void setup_x86_bogus_stack(unsigned char * stk)
+{
+	unsigned long *switch16_ptr, *switch32_ptr;
+	struct pt_regs *regs;
+	unsigned long stack_top, stack_bot;
+	unsigned short iret_frame16_off;
+	int cpu = smp_processor_id();
+	/* reserve the space on 32bit stack for the magic switch16 pointer */
+	memmove(stk, stk + 8, sizeof(struct pt_regs));
+	switch16_ptr = (unsigned long *)(stk + sizeof(struct pt_regs));
+	regs = (struct pt_regs *)stk;
+	/* now the switch32 on 16bit stack */
+	stack_bot = (unsigned long)&per_cpu(cpu_16bit_stack, cpu);
+	stack_top = stack_bot +	CPU_16BIT_STACK_SIZE;
+	switch32_ptr = (unsigned long *)(stack_top - 8);
+	iret_frame16_off = CPU_16BIT_STACK_SIZE - 8 - 20;
+	/* copy iret frame on 16bit stack */
+	memcpy((void *)(stack_bot + iret_frame16_off), &regs->eip, 20);
+	/* fill in the switch pointers */
+	switch16_ptr[0] = (regs->esp & 0xffff0000) | iret_frame16_off;
+	switch16_ptr[1] = __ESPFIX_SS;
+	switch32_ptr[0] = (unsigned long)stk + sizeof(struct pt_regs) +
+		8 - CPU_16BIT_STACK_SIZE;
+	switch32_ptr[1] = __KERNEL_DS;
+}
+
+fastcall unsigned char * fixup_x86_bogus_stack(unsigned short sp)
+{
+	unsigned long *switch32_ptr;
+	unsigned char *stack16, *stack32;
+	unsigned long stack_top, stack_bot;
+	int len;
+	int cpu = smp_processor_id();
+	stack_bot = (unsigned long)&per_cpu(cpu_16bit_stack, cpu);
+	stack_top = stack_bot +	CPU_16BIT_STACK_SIZE;
+	switch32_ptr = (unsigned long *)(stack_top - 8);
+	/* copy the data from 16bit stack to 32bit stack */
+	len = CPU_16BIT_STACK_SIZE - 8 - sp;
+	stack16 = (unsigned char *)(stack_bot + sp);
+	stack32 = (unsigned char *)
+		(switch32_ptr[0] + CPU_16BIT_STACK_SIZE - 8 - len);
+	memcpy(stack32, stack16, len);
+	return stack32;
+}
+
 /*
  *  'math_state_restore()' saves the current math information in the
  * old math state array, and gets the new ones from the current task
@@ -819,7 +1150,7 @@ asmlinkage void math_state_restore(struct pt_regs regs)
 	struct task_struct *tsk = thread->task;
 
 	clts();		/* Allow maths ops (or we recurse) */
-	if (!tsk->used_math)
+	if (!tsk_used_math(tsk))
 		init_fpu(tsk);
 	restore_fpu(tsk);
 	thread->status |= TS_USEDFPU;	/* So we fnsave on switch_to() */
@@ -829,8 +1160,8 @@ asmlinkage void math_state_restore(struct pt_regs regs)
 
 asmlinkage void math_emulate(long arg)
 {
-	printk("math-emulation not enabled and no coprocessor found.\n");
-	printk("killing %s.\n",current->comm);
+	printk(KERN_EMERG "math-emulation not enabled and no coprocessor found.\n");
+	printk(KERN_EMERG "killing %s.\n",current->comm);
 	force_sig(SIGFPE,current);
 	schedule();
 }
@@ -847,7 +1178,7 @@ void __init trap_init_f00f_bug(void)
 	 * it uses the read-only mapped virtual address.
 	 */
 	idt_descr.address = fix_to_virt(FIX_F00F_IDT);
-	__asm__ __volatile__("lidt %0" : : "m" (idt_descr));
+	load_idt(&idt_descr);
 }
 #endif
 
@@ -876,6 +1207,14 @@ void set_intr_gate(unsigned int n, void *addr)
 	_set_gate(idt_table+n,14,0,addr,__KERNEL_CS);
 }
 
+/*
+ * This routine sets up an interrupt gate at directory privilege level 3.
+ */
+static inline void set_system_intr_gate(unsigned int n, void *addr)
+{
+	_set_gate(idt_table+n, 14, 3, addr, __KERNEL_CS);
+}
+
 static void __init set_trap_gate(unsigned int n, void *addr)
 {
 	_set_gate(idt_table+n,15,0,addr,__KERNEL_CS);
@@ -886,11 +1225,6 @@ static void __init set_system_gate(unsigned int n, void *addr)
 	_set_gate(idt_table+n,15,3,addr,__KERNEL_CS);
 }
 
-static void __init set_call_gate(void *a, void *addr)
-{
-	_set_gate(a,12,3,addr,__KERNEL_CS);
-}
-
 static void __init set_task_gate(unsigned int n, unsigned int gdt_entry)
 {
 	_set_gate(idt_table+n,5,0,0,(gdt_entry<<3));
@@ -900,9 +1234,11 @@ static void __init set_task_gate(unsigned int n, unsigned int gdt_entry)
 void __init trap_init(void)
 {
 #ifdef CONFIG_EISA
-	if (isa_readl(0x0FFFD9) == 'E'+('I'<<8)+('S'<<16)+('A'<<24)) {
+	void __iomem *p = ioremap(0x0FFFD9, 4);
+	if (readl(p) == 'E'+('I'<<8)+('S'<<16)+('A'<<24)) {
 		EISA_bus = 1;
 	}
+	iounmap(p);
 #endif
 
 #ifdef CONFIG_X86_LOCAL_APIC
@@ -912,9 +1248,9 @@ void __init trap_init(void)
 	set_trap_gate(0,&divide_error);
 	set_intr_gate(1,&debug);
 	set_intr_gate(2,&nmi);
-	set_system_gate(3,&int3);	/* int3-5 can be called from all */
+	set_system_intr_gate(3, &int3); /* int3/4 can be called from all */
 	set_system_gate(4,&overflow);
-	set_system_gate(5,&bounds);
+	set_trap_gate(5,&bounds);
 	set_trap_gate(6,&invalid_op);
 	set_trap_gate(7,&device_not_available);
 	set_task_gate(8,GDT_ENTRY_DOUBLEFAULT_TSS);
@@ -932,14 +1268,29 @@ void __init trap_init(void)
 #endif
 	set_trap_gate(19,&simd_coprocessor_error);
 
-	set_system_gate(SYSCALL_VECTOR,&system_call);
+	if (cpu_has_fxsr) {
+		/*
+		 * Verify that the FXSAVE/FXRSTOR data will be 16-byte aligned.
+		 * Generates a compile-time "error: zero width for bit-field" if
+		 * the alignment is wrong.
+		 */
+		struct fxsrAlignAssert {
+			int _:!(offsetof(struct task_struct,
+					thread.i387.fxsave) & 15);
+		};
+
+		printk(KERN_INFO "Enabling fast FPU save and restore... ");
+		set_in_cr4(X86_CR4_OSFXSR);
+		printk("done.\n");
+	}
+	if (cpu_has_xmm) {
+		printk(KERN_INFO "Enabling unmasked SIMD FPU exception "
+				"support... ");
+		set_in_cr4(X86_CR4_OSXMMEXCPT);
+		printk("done.\n");
+	}
 
-	/*
-	 * default LDT is a single-entry callgate to lcall7 for iBCS
-	 * and a callgate to lcall27 for Solaris/x86 binaries
-	 */
-	set_call_gate(&default_ldt[0],lcall7);
-	set_call_gate(&default_ldt[4],lcall27);
+	set_system_gate(SYSCALL_VECTOR,&system_call);
 
 	/*
 	 * Should be a barrier for any external CPU state.
@@ -948,3 +1299,10 @@ void __init trap_init(void)
 
 	trap_init_hook();
 }
+
+static int __init kstack_setup(char *s)
+{
+	kstack_depth_to_print = simple_strtoul(s, NULL, 0);
+	return 1;
+}
+__setup("kstack=", kstack_setup);