[linux-2.6-block.git] / arch / x86 / kernel / process_32.c

/*
 *  Copyright (C) 1995  Linus Torvalds
 *
 *  Pentium III FXSR, SSE support
 *	Gareth Hughes <gareth@valinux.com>, May 2000
 */

/*
 * This file handles the architecture-dependent parts of process handling..
 */

#include <linux/stackprotector.h>
#include <linux/cpu.h>
#include <linux/errno.h>
#include <linux/sched.h>
#include <linux/fs.h>
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/elfcore.h>
#include <linux/smp.h>
#include <linux/stddef.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/user.h>
#include <linux/interrupt.h>
#include <linux/delay.h>
#include <linux/reboot.h>
#include <linux/init.h>
#include <linux/mc146818rtc.h>
#include <linux/module.h>
#include <linux/kallsyms.h>
#include <linux/ptrace.h>
#include <linux/personality.h>
#include <linux/tick.h>
#include <linux/percpu.h>
#include <linux/prctl.h>
#include <linux/ftrace.h>
#include <linux/uaccess.h>
#include <linux/io.h>
#include <linux/kdebug.h>

#include <asm/pgtable.h>
#include <asm/system.h>
#include <asm/ldt.h>
#include <asm/processor.h>
#include <asm/i387.h>
#include <asm/desc.h>
#ifdef CONFIG_MATH_EMULATION
#include <asm/math_emu.h>
#endif

#include <linux/err.h>

#include <asm/tlbflush.h>
#include <asm/cpu.h>
#include <asm/idle.h>
#include <asm/syscalls.h>
#include <asm/ds.h>
#include <asm/debugreg.h>

asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");

/*
 * Return saved PC of a blocked thread.
 */
unsigned long thread_saved_pc(struct task_struct *tsk)
{
	return ((unsigned long *)tsk->thread.sp)[3];
}

#ifndef CONFIG_SMP
static inline void play_dead(void)
{
	BUG();
}
#endif

/*
 * The idle thread. There's no useful work to be
 * done, so just try to conserve power and have a
 * low exit latency (ie sit in a loop waiting for
 * somebody to say that they'd like to reschedule)
 */
void cpu_idle(void)
{
	int cpu = smp_processor_id();

	/*
	 * If we're the non-boot CPU, nothing set the stack canary up
	 * for us.  CPU0 already has it initialized but no harm in
	 * doing it again.  This is a good place for updating it, as
	 * we wont ever return from this function (so the invalid
	 * canaries already on the stack wont ever trigger).
	 */
	boot_init_stack_canary();

	current_thread_info()->status |= TS_POLLING;

	/* endless idle loop with no priority at all */
	while (1) {
		tick_nohz_stop_sched_tick(1);
		while (!need_resched()) {

			check_pgt_cache();
			rmb();

			if (cpu_is_offline(cpu))
				play_dead();

			local_irq_disable();
			/* Don't trace irqs off for idle */
			stop_critical_timings();
			pm_idle();
			start_critical_timings();
		}
		tick_nohz_restart_sched_tick();
		preempt_enable_no_resched();
		schedule();
		preempt_disable();
	}
}

void __show_regs(struct pt_regs *regs, int all)
{
	unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
	unsigned long d0, d1, d2, d3, d6, d7;
	unsigned long sp;
	unsigned short ss, gs;

	if (user_mode_vm(regs)) {
		sp = regs->sp;
		ss = regs->ss & 0xffff;
		gs = get_user_gs(regs);
	} else {
		sp = kernel_stack_pointer(regs);
		savesegment(ss, ss);
		savesegment(gs, gs);
	}

	show_regs_common();

	printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n",
			(u16)regs->cs, regs->ip, regs->flags,
			smp_processor_id());
	print_symbol("EIP is at %s\n", regs->ip);

	printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n",
		regs->ax, regs->bx, regs->cx, regs->dx);
	printk("ESI: %08lx EDI: %08lx EBP: %08lx ESP: %08lx\n",
		regs->si, regs->di, regs->bp, sp);
	printk(" DS: %04x ES: %04x FS: %04x GS: %04x SS: %04x\n",
	       (u16)regs->ds, (u16)regs->es, (u16)regs->fs, gs, ss);

	if (!all)
		return;

	cr0 = read_cr0();
	cr2 = read_cr2();
	cr3 = read_cr3();
	cr4 = read_cr4_safe();
	printk("CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n",
			cr0, cr2, cr3, cr4);

	get_debugreg(d0, 0);
	get_debugreg(d1, 1);
	get_debugreg(d2, 2);
	get_debugreg(d3, 3);
	printk("DR0: %08lx DR1: %08lx DR2: %08lx DR3: %08lx\n",
			d0, d1, d2, d3);

	get_debugreg(d6, 6);
	get_debugreg(d7, 7);
	printk("DR6: %08lx DR7: %08lx\n",
			d6, d7);
}

void show_regs(struct pt_regs *regs)
{
	show_registers(regs);
	show_trace(NULL, regs, &regs->sp, regs->bp);
}

/*
 * This gets run with %bx containing the
 * function to call, and %dx containing
 * the "args".
 */
extern void kernel_thread_helper(void);

/*
 * Create a kernel thread
 */
int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
{
	struct pt_regs regs;

	memset(&regs, 0, sizeof(regs));

	regs.bx = (unsigned long) fn;
	regs.dx = (unsigned long) arg;

	regs.ds = __USER_DS;
	regs.es = __USER_DS;
	regs.fs = __KERNEL_PERCPU;
	regs.gs = __KERNEL_STACK_CANARY;
	regs.orig_ax = -1;
	regs.ip = (unsigned long) kernel_thread_helper;
	regs.cs = __KERNEL_CS | get_kernel_rpl();
	regs.flags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;

	/* Ok, create the new process.. */
	return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
}
EXPORT_SYMBOL(kernel_thread);

void release_thread(struct task_struct *dead_task)
{
	BUG_ON(dead_task->mm);
	release_vm86_irqs(dead_task);
}

/*
 * This gets called before we allocate a new thread and copy
 * the current task into it.
 */
void prepare_to_copy(struct task_struct *tsk)
{
	unlazy_fpu(tsk);
}

int copy_thread(unsigned long clone_flags, unsigned long sp,
	unsigned long unused,
	struct task_struct *p, struct pt_regs *regs)
{
	struct pt_regs *childregs;
	struct task_struct *tsk;
	int err;

	childregs = task_pt_regs(p);
	*childregs = *regs;
	childregs->ax = 0;
	childregs->sp = sp;

	p->thread.sp = (unsigned long) childregs;
	p->thread.sp0 = (unsigned long) (childregs+1);

	p->thread.ip = (unsigned long) ret_from_fork;

	task_user_gs(p) = get_user_gs(regs);

	p->thread.io_bitmap_ptr = NULL;
	tsk = current;
	err = -ENOMEM;

	memset(p->thread.ptrace_bps, 0, sizeof(p->thread.ptrace_bps));

	if (unlikely(test_tsk_thread_flag(tsk, TIF_IO_BITMAP))) {
		p->thread.io_bitmap_ptr = kmemdup(tsk->thread.io_bitmap_ptr,
						IO_BITMAP_BYTES, GFP_KERNEL);
		if (!p->thread.io_bitmap_ptr) {
			p->thread.io_bitmap_max = 0;
			return -ENOMEM;
		}
		set_tsk_thread_flag(p, TIF_IO_BITMAP);
	}

	err = 0;

	/*
	 * Set a new TLS for the child thread?
	 */
	if (clone_flags & CLONE_SETTLS)
		err = do_set_thread_area(p, -1,
			(struct user_desc __user *)childregs->si, 0);

	if (err && p->thread.io_bitmap_ptr) {
		kfree(p->thread.io_bitmap_ptr);
		p->thread.io_bitmap_max = 0;
	}

	clear_tsk_thread_flag(p, TIF_DS_AREA_MSR);
	p->thread.ds_ctx = NULL;

	clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR);
	p->thread.debugctlmsr = 0;

	return err;
}

void
start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
{
	set_user_gs(regs, 0);
	regs->fs		= 0;
	set_fs(USER_DS);
	regs->ds		= __USER_DS;
	regs->es		= __USER_DS;
	regs->ss		= __USER_DS;
	regs->cs		= __USER_CS;
	regs->ip		= new_ip;
	regs->sp		= new_sp;
	/*
	 * Free the old FP and other extended state
	 */
	free_thread_xstate(current);
}
EXPORT_SYMBOL_GPL(start_thread);


/*
 *	switch_to(x,yn) should switch tasks from x to y.
 *
 * We fsave/fwait so that an exception goes off at the right time
 * (as a call from the fsave or fwait in effect) rather than to
 * the wrong process. Lazy FP saving no longer makes any sense
 * with modern CPU's, and this simplifies a lot of things (SMP
 * and UP become the same).
 *
 * NOTE! We used to use the x86 hardware context switching. The
 * reason for not using it any more becomes apparent when you
 * try to recover gracefully from saved state that is no longer
 * valid (stale segment register values in particular). With the
 * hardware task-switch, there is no way to fix up bad state in
 * a reasonable manner.
 *
 * The fact that Intel documents the hardware task-switching to
 * be slow is a fairly red herring - this code is not noticeably
 * faster. However, there _is_ some room for improvement here,
 * so the performance issues may eventually be a valid point.
 * More important, however, is the fact that this allows us much
 * more flexibility.
 *
 * The return value (in %ax) will be the "prev" task after
 * the task-switch, and shows up in ret_from_fork in entry.S,
 * for example.
 */
__notrace_funcgraph struct task_struct *
__switch_to(struct task_struct *prev_p, struct task_struct *next_p)
{
	struct thread_struct *prev = &prev_p->thread,
				 *next = &next_p->thread;
	int cpu = smp_processor_id();
	struct tss_struct *tss = &per_cpu(init_tss, cpu);
	bool preload_fpu;

	/* never put a printk in __switch_to... printk() calls wake_up*() indirectly */

	/*
	 * If the task has used fpu the last 5 timeslices, just do a full
	 * restore of the math state immediately to avoid the trap; the
	 * chances of needing FPU soon are obviously high now
	 */
	preload_fpu = tsk_used_math(next_p) && next_p->fpu_counter > 5;

	__unlazy_fpu(prev_p);

	/* we're going to use this soon, after a few expensive things */
	if (preload_fpu)
		prefetch(next->xstate);

	/*
	 * Reload esp0.
	 */
	load_sp0(tss, next);

	/*
	 * Save away %gs. No need to save %fs, as it was saved on the
	 * stack on entry.  No need to save %es and %ds, as those are
	 * always kernel segments while inside the kernel.  Doing this
	 * before setting the new TLS descriptors avoids the situation
	 * where we temporarily have non-reloadable segments in %fs
	 * and %gs.  This could be an issue if the NMI handler ever
	 * used %fs or %gs (it does not today), or if the kernel is
	 * running inside of a hypervisor layer.
	 */
	lazy_save_gs(prev->gs);

	/*
	 * Load the per-thread Thread-Local Storage descriptor.
	 */
	load_TLS(next, cpu);

	/*
	 * Restore IOPL if needed.  In normal use, the flags restore
	 * in the switch assembly will handle this.  But if the kernel
	 * is running virtualized at a non-zero CPL, the popf will
	 * not restore flags, so it must be done in a separate step.
	 */
	if (get_kernel_rpl() && unlikely(prev->iopl != next->iopl))
		set_iopl_mask(next->iopl);

	/*
	 * Now maybe handle debug registers and/or IO bitmaps
	 */
	if (unlikely(task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV ||
		     task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT))
		__switch_to_xtra(prev_p, next_p, tss);

	/* If we're going to preload the fpu context, make sure clts
	   is run while we're batching the cpu state updates. */
	if (preload_fpu)
		clts();

	/*
	 * Leave lazy mode, flushing any hypercalls made here.
	 * This must be done before restoring TLS segments so
	 * the GDT and LDT are properly updated, and must be
	 * done before math_state_restore, so the TS bit is up
	 * to date.
	 */
	arch_end_context_switch(next_p);

	if (preload_fpu)
		__math_state_restore();

	/*
	 * Restore %gs if needed (which is common)
	 */
	if (prev->gs | next->gs)
		lazy_load_gs(next->gs);

	percpu_write(current_task, next_p);

	return prev_p;
}

int sys_clone(struct pt_regs *regs)
{
	unsigned long clone_flags;
	unsigned long newsp;
	int __user *parent_tidptr, *child_tidptr;

	clone_flags = regs->bx;
	newsp = regs->cx;
	parent_tidptr = (int __user *)regs->dx;
	child_tidptr = (int __user *)regs->di;
	if (!newsp)
		newsp = regs->sp;
	return do_fork(clone_flags, newsp, regs, 0, parent_tidptr, child_tidptr);
}

/*
 * sys_execve() executes a new program.
 */
int sys_execve(struct pt_regs *regs)
{
	int error;
	char *filename;

	filename = getname((char __user *) regs->bx);
	error = PTR_ERR(filename);
	if (IS_ERR(filename))
		goto out;
	error = do_execve(filename,
			(char __user * __user *) regs->cx,
			(char __user * __user *) regs->dx,
			regs);
	if (error == 0) {
		/* Make sure we don't return using sysenter.. */
		set_thread_flag(TIF_IRET);
	}
	putname(filename);
out:
	return error;
}

#define top_esp                (THREAD_SIZE - sizeof(unsigned long))
#define top_ebp                (THREAD_SIZE - 2*sizeof(unsigned long))

unsigned long get_wchan(struct task_struct *p)
{
	unsigned long bp, sp, ip;
	unsigned long stack_page;
	int count = 0;
	if (!p || p == current || p->state == TASK_RUNNING)
		return 0;
	stack_page = (unsigned long)task_stack_page(p);
	sp = p->thread.sp;
	if (!stack_page || sp < stack_page || sp > top_esp+stack_page)
		return 0;
	/* include/asm-i386/system.h:switch_to() pushes bp last. */
	bp = *(unsigned long *) sp;
	do {
		if (bp < stack_page || bp > top_ebp+stack_page)
			return 0;
		ip = *(unsigned long *) (bp+4);
		if (!in_sched_functions(ip))
			return ip;
		bp = *(unsigned long *) bp;
	} while (count++ < 16);
	return 0;
}
Commit	Line	Data
1da177e4	1	/*
1da177e4 LT	2	* Copyright (C) 1995 Linus Torvalds
	3	*
	4	* Pentium III FXSR, SSE support
	5	* Gareth Hughes <gareth@valinux.com>, May 2000
	6	*/
	7
	8	/*
	9	* This file handles the architecture-dependent parts of process handling..
	10	*/
	11
5c79d2a5	12	#include <linux/stackprotector.h>
f3705136	13	#include <linux/cpu.h>
1da177e4 LT	14	#include <linux/errno.h>
	15	#include <linux/sched.h>
	16	#include <linux/fs.h>
	17	#include <linux/kernel.h>
	18	#include <linux/mm.h>
	19	#include <linux/elfcore.h>
	20	#include <linux/smp.h>
1da177e4 LT	21	#include <linux/stddef.h>
	22	#include <linux/slab.h>
	23	#include <linux/vmalloc.h>
	24	#include <linux/user.h>
1da177e4	25	#include <linux/interrupt.h>
1da177e4 LT	26	#include <linux/delay.h>
	27	#include <linux/reboot.h>
	28	#include <linux/init.h>
	29	#include <linux/mc146818rtc.h>
	30	#include <linux/module.h>
	31	#include <linux/kallsyms.h>
	32	#include <linux/ptrace.h>
c16b63e0	33	#include <linux/personality.h>
74167347	34	#include <linux/tick.h>
7c3576d2	35	#include <linux/percpu.h>
529e25f6	36	#include <linux/prctl.h>
8b96f011	37	#include <linux/ftrace.h>
befa9e78 JSR	38	#include <linux/uaccess.h>
	39	#include <linux/io.h>
	40	#include <linux/kdebug.h>
1da177e4	41
1da177e4 LT	42	#include <asm/pgtable.h>
1da177e4 LT	43	#include <asm/system.h>
1da177e4 LT	44	#include <asm/ldt.h>
	45	#include <asm/processor.h>
	46	#include <asm/i387.h>
1da177e4 LT	47	#include <asm/desc.h>
	48	#ifdef CONFIG_MATH_EMULATION
	49	#include <asm/math_emu.h>
	50	#endif
	51
1da177e4 LT	52	#include <linux/err.h>
1da177e4 LT	53
f3705136 ZM	54	#include <asm/tlbflush.h>
f3705136 ZM	55	#include <asm/cpu.h>
1eda8149	56	#include <asm/idle.h>
bbc1f698	57	#include <asm/syscalls.h>
bf53de90	58	#include <asm/ds.h>
66cb5917	59	#include <asm/debugreg.h>
f3705136	60
1da177e4 LT	61	asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
1da177e4 LT	62
1da177e4 LT	63	/*
	64	* Return saved PC of a blocked thread.
	65	*/
	66	unsigned long thread_saved_pc(struct task_struct *tsk)
	67	{
faca6227	68	return ((unsigned long *)tsk->thread.sp)[3];
1da177e4 LT	69	}
1da177e4 LT	70
913da64b AN	71	#ifndef CONFIG_SMP
	72	static inline void play_dead(void)
	73	{
	74	BUG();
	75	}
	76	#endif
	77
1da177e4 LT	78	/*
	79	* The idle thread. There's no useful work to be
	80	* done, so just try to conserve power and have a
	81	* low exit latency (ie sit in a loop waiting for
	82	* somebody to say that they'd like to reschedule)
	83	*/
f3705136	84	void cpu_idle(void)
1da177e4	85	{
5bfb5d69	86	int cpu = smp_processor_id();
f3705136	87
5c79d2a5 TH	88	/*
	89	* If we're the non-boot CPU, nothing set the stack canary up
	90	* for us. CPU0 already has it initialized but no harm in
	91	* doing it again. This is a good place for updating it, as
	92	* we wont ever return from this function (so the invalid
	93	* canaries already on the stack wont ever trigger).
	94	*/
	95	boot_init_stack_canary();
	96
495ab9c0	97	current_thread_info()->status \|= TS_POLLING;
64c7c8f8	98
1da177e4 LT	99	/* endless idle loop with no priority at all */
1da177e4 LT	100	while (1) {
b8f8c3cf	101	tick_nohz_stop_sched_tick(1);
1da177e4	102	while (!need_resched()) {
1da177e4	103
f1d1a842	104	check_pgt_cache();
1da177e4	105	rmb();
1da177e4	106
f3705136 ZM	107	if (cpu_is_offline(cpu))
	108	play_dead();
	109
7f424a8b	110	local_irq_disable();
6cd8a4bb SR	111	/* Don't trace irqs off for idle */
6cd8a4bb SR	112	stop_critical_timings();
6ddd2a27	113	pm_idle();
6cd8a4bb	114	start_critical_timings();
1da177e4	115	}
74167347	116	tick_nohz_restart_sched_tick();
5bfb5d69	117	preempt_enable_no_resched();
1da177e4	118	schedule();
5bfb5d69	119	preempt_disable();
1da177e4 LT	120	}
	121	}
	122
e2ce07c8	123	void __show_regs(struct pt_regs *regs, int all)
1da177e4 LT	124	{
1da177e4 LT	125	unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
bb1995d5	126	unsigned long d0, d1, d2, d3, d6, d7;
65ea5b03	127	unsigned long sp;
9d975ebd PE	128	unsigned short ss, gs;
	129
	130	if (user_mode_vm(regs)) {
65ea5b03 PA	131	sp = regs->sp;
65ea5b03 PA	132	ss = regs->ss & 0xffff;
d9a89a26	133	gs = get_user_gs(regs);
9d975ebd	134	} else {
def3c5d0	135	sp = kernel_stack_pointer(regs);
9d975ebd PE	136	savesegment(ss, ss);
	137	savesegment(gs, gs);
	138	}
1da177e4	139
814e2c84	140	show_regs_common();
9d975ebd PE	141
9d975ebd PE	142	printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n",
92bc2056	143	(u16)regs->cs, regs->ip, regs->flags,
9d975ebd	144	smp_processor_id());
65ea5b03	145	print_symbol("EIP is at %s\n", regs->ip);
1da177e4	146
1da177e4	147	printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n",
65ea5b03	148	regs->ax, regs->bx, regs->cx, regs->dx);
9d975ebd	149	printk("ESI: %08lx EDI: %08lx EBP: %08lx ESP: %08lx\n",
65ea5b03	150	regs->si, regs->di, regs->bp, sp);
9d975ebd	151	printk(" DS: %04x ES: %04x FS: %04x GS: %04x SS: %04x\n",
92bc2056	152	(u16)regs->ds, (u16)regs->es, (u16)regs->fs, gs, ss);
9d975ebd PE	153
	154	if (!all)
	155	return;
1da177e4	156
4bb0d3ec ZA	157	cr0 = read_cr0();
	158	cr2 = read_cr2();
	159	cr3 = read_cr3();
ff6e8c0d	160	cr4 = read_cr4_safe();
9d975ebd PE	161	printk("CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n",
9d975ebd PE	162	cr0, cr2, cr3, cr4);
bb1995d5 AS	163
	164	get_debugreg(d0, 0);
	165	get_debugreg(d1, 1);
	166	get_debugreg(d2, 2);
	167	get_debugreg(d3, 3);
	168	printk("DR0: %08lx DR1: %08lx DR2: %08lx DR3: %08lx\n",
	169	d0, d1, d2, d3);
9d975ebd	170
bb1995d5 AS	171	get_debugreg(d6, 6);
bb1995d5 AS	172	get_debugreg(d7, 7);
9d975ebd PE	173	printk("DR6: %08lx DR7: %08lx\n",
	174	d6, d7);
	175	}
bb1995d5	176
9d975ebd PE	177	void show_regs(struct pt_regs *regs)
9d975ebd PE	178	{
a489ca35	179	show_registers(regs);
5bc27dc2	180	show_trace(NULL, regs, &regs->sp, regs->bp);
1da177e4 LT	181	}
	182
	183	/*
65ea5b03 PA	184	* This gets run with %bx containing the
65ea5b03 PA	185	* function to call, and %dx containing
1da177e4 LT	186	* the "args".
	187	*/
	188	extern void kernel_thread_helper(void);
1da177e4 LT	189
	190	/*
	191	* Create a kernel thread
	192	*/
befa9e78	193	int kernel_thread(int (fn)(void ), void *arg, unsigned long flags)
1da177e4 LT	194	{
	195	struct pt_regs regs;
	196
	197	memset(&regs, 0, sizeof(regs));
	198
65ea5b03 PA	199	regs.bx = (unsigned long) fn;
65ea5b03 PA	200	regs.dx = (unsigned long) arg;
1da177e4	201
65ea5b03 PA	202	regs.ds = __USER_DS;
	203	regs.es = __USER_DS;
	204	regs.fs = __KERNEL_PERCPU;
60a5317f	205	regs.gs = __KERNEL_STACK_CANARY;
65ea5b03 PA	206	regs.orig_ax = -1;
	207	regs.ip = (unsigned long) kernel_thread_helper;
	208	regs.cs = __KERNEL_CS \| get_kernel_rpl();
	209	regs.flags = X86_EFLAGS_IF \| X86_EFLAGS_SF \| X86_EFLAGS_PF \| 0x2;
1da177e4 LT	210
1da177e4 LT	211	/* Ok, create the new process.. */
8cf2c519	212	return do_fork(flags \| CLONE_VM \| CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
1da177e4	213	}
129f6946	214	EXPORT_SYMBOL(kernel_thread);
1da177e4	215
1da177e4 LT	216	void release_thread(struct task_struct *dead_task)
1da177e4 LT	217	{
2684927c	218	BUG_ON(dead_task->mm);
1da177e4 LT	219	release_vm86_irqs(dead_task);
	220	}
	221
	222	/*
	223	* This gets called before we allocate a new thread and copy
	224	* the current task into it.
	225	*/
	226	void prepare_to_copy(struct task_struct *tsk)
	227	{
	228	unlazy_fpu(tsk);
	229	}
	230
6f2c55b8	231	int copy_thread(unsigned long clone_flags, unsigned long sp,
1da177e4	232	unsigned long unused,
befa9e78	233	struct task_struct p, struct pt_regs regs)
1da177e4	234	{
befa9e78	235	struct pt_regs *childregs;
1da177e4 LT	236	struct task_struct *tsk;
	237	int err;
	238
07b047fc	239	childregs = task_pt_regs(p);
f48d9663	240	childregs = regs;
65ea5b03 PA	241	childregs->ax = 0;
65ea5b03 PA	242	childregs->sp = sp;
f48d9663	243
faca6227 PA	244	p->thread.sp = (unsigned long) childregs;
faca6227 PA	245	p->thread.sp0 = (unsigned long) (childregs+1);
1da177e4	246
faca6227	247	p->thread.ip = (unsigned long) ret_from_fork;
1da177e4	248
d9a89a26	249	task_user_gs(p) = get_user_gs(regs);
1da177e4	250
66cb5917	251	p->thread.io_bitmap_ptr = NULL;
1da177e4	252	tsk = current;
66cb5917	253	err = -ENOMEM;
24f1e32c FW	254
24f1e32c FW	255	memset(p->thread.ptrace_bps, 0, sizeof(p->thread.ptrace_bps));
66cb5917	256
b3cf2576	257	if (unlikely(test_tsk_thread_flag(tsk, TIF_IO_BITMAP))) {
52978be6 AD	258	p->thread.io_bitmap_ptr = kmemdup(tsk->thread.io_bitmap_ptr,
52978be6 AD	259	IO_BITMAP_BYTES, GFP_KERNEL);
1da177e4 LT	260	if (!p->thread.io_bitmap_ptr) {
	261	p->thread.io_bitmap_max = 0;
	262	return -ENOMEM;
	263	}
b3cf2576	264	set_tsk_thread_flag(p, TIF_IO_BITMAP);
1da177e4 LT	265	}
1da177e4 LT	266
efd1ca52 RM	267	err = 0;
efd1ca52 RM	268
1da177e4 LT	269	/*
	270	* Set a new TLS for the child thread?
	271	*/
efd1ca52 RM	272	if (clone_flags & CLONE_SETTLS)
efd1ca52 RM	273	err = do_set_thread_area(p, -1,
65ea5b03	274	(struct user_desc __user *)childregs->si, 0);
1da177e4	275
1da177e4 LT	276	if (err && p->thread.io_bitmap_ptr) {
	277	kfree(p->thread.io_bitmap_ptr);
	278	p->thread.io_bitmap_max = 0;
	279	}
bf53de90	280
2311f0de MM	281	clear_tsk_thread_flag(p, TIF_DS_AREA_MSR);
2311f0de MM	282	p->thread.ds_ctx = NULL;
bf53de90 MM	283
	284	clear_tsk_thread_flag(p, TIF_DEBUGCTLMSR);
	285	p->thread.debugctlmsr = 0;
	286
1da177e4 LT	287	return err;
	288	}
	289
513ad84b IM	290	void
	291	start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
	292	{
d9a89a26	293	set_user_gs(regs, 0);
513ad84b IM	294	regs->fs = 0;
	295	set_fs(USER_DS);
	296	regs->ds = __USER_DS;
	297	regs->es = __USER_DS;
	298	regs->ss = __USER_DS;
	299	regs->cs = __USER_CS;
	300	regs->ip = new_ip;
	301	regs->sp = new_sp;
aa283f49 SS	302	/*
	303	* Free the old FP and other extended state
	304	*/
	305	free_thread_xstate(current);
513ad84b IM	306	}
	307	EXPORT_SYMBOL_GPL(start_thread);
	308
1da177e4 LT	309
	310	/*
	311	* switch_to(x,yn) should switch tasks from x to y.
	312	*
	313	* We fsave/fwait so that an exception goes off at the right time
	314	* (as a call from the fsave or fwait in effect) rather than to
	315	* the wrong process. Lazy FP saving no longer makes any sense
	316	* with modern CPU's, and this simplifies a lot of things (SMP
	317	* and UP become the same).
	318	*
	319	* NOTE! We used to use the x86 hardware context switching. The
	320	* reason for not using it any more becomes apparent when you
	321	* try to recover gracefully from saved state that is no longer
	322	* valid (stale segment register values in particular). With the
	323	* hardware task-switch, there is no way to fix up bad state in
	324	* a reasonable manner.
	325	*
	326	* The fact that Intel documents the hardware task-switching to
	327	* be slow is a fairly red herring - this code is not noticeably
	328	* faster. However, there _is_ some room for improvement here,
	329	* so the performance issues may eventually be a valid point.
	330	* More important, however, is the fact that this allows us much
	331	* more flexibility.
	332	*
65ea5b03	333	* The return value (in %ax) will be the "prev" task after
1da177e4 LT	334	* the task-switch, and shows up in ret_from_fork in entry.S,
	335	* for example.
	336	*/
8b96f011 FW	337	__notrace_funcgraph struct task_struct *
8b96f011 FW	338	__switch_to(struct task_struct prev_p, struct task_struct next_p)
1da177e4 LT	339	{
	340	struct thread_struct *prev = &prev_p->thread,
	341	*next = &next_p->thread;
	342	int cpu = smp_processor_id();
	343	struct tss_struct *tss = &per_cpu(init_tss, cpu);
2fcddce1	344	bool preload_fpu;
1da177e4 LT	345
	346	/* never put a printk in __switch_to... printk() calls wake_up() indirectly /
	347
2fcddce1 JF	348	/*
	349	* If the task has used fpu the last 5 timeslices, just do a full
	350	* restore of the math state immediately to avoid the trap; the
	351	* chances of needing FPU soon are obviously high now
	352	*/
	353	preload_fpu = tsk_used_math(next_p) && next_p->fpu_counter > 5;
1da177e4	354
2fcddce1	355	__unlazy_fpu(prev_p);
acc20761 CE	356
acc20761 CE	357	/* we're going to use this soon, after a few expensive things */
2fcddce1	358	if (preload_fpu)
61c4628b	359	prefetch(next->xstate);
acc20761	360
1da177e4	361	/*
e7a2ff59	362	* Reload esp0.
1da177e4	363	*/
faca6227	364	load_sp0(tss, next);
1da177e4 LT	365
1da177e4 LT	366	/*
464d1a78	367	* Save away %gs. No need to save %fs, as it was saved on the
f95d47ca JF	368	* stack on entry. No need to save %es and %ds, as those are
	369	* always kernel segments while inside the kernel. Doing this
	370	* before setting the new TLS descriptors avoids the situation
	371	* where we temporarily have non-reloadable segments in %fs
	372	* and %gs. This could be an issue if the NMI handler ever
	373	* used %fs or %gs (it does not today), or if the kernel is
	374	* running inside of a hypervisor layer.
1da177e4	375	*/
ccbeed3a	376	lazy_save_gs(prev->gs);
1da177e4 LT	377
1da177e4 LT	378	/*
e7a2ff59	379	* Load the per-thread Thread-Local Storage descriptor.
1da177e4	380	*/
e7a2ff59	381	load_TLS(next, cpu);
1da177e4	382
8b151144 ZA	383	/*
	384	* Restore IOPL if needed. In normal use, the flags restore
	385	* in the switch assembly will handle this. But if the kernel
	386	* is running virtualized at a non-zero CPL, the popf will
	387	* not restore flags, so it must be done in a separate step.
	388	*/
	389	if (get_kernel_rpl() && unlikely(prev->iopl != next->iopl))
	390	set_iopl_mask(next->iopl);
	391
1da177e4	392	/*
b3cf2576	393	* Now maybe handle debug registers and/or IO bitmaps
1da177e4	394	*/
cf99abac AA	395	if (unlikely(task_thread_info(prev_p)->flags & _TIF_WORK_CTXSW_PREV \|\|
	396	task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT))
	397	__switch_to_xtra(prev_p, next_p, tss);
ffaa8bd6	398
2fcddce1 JF	399	/* If we're going to preload the fpu context, make sure clts
	400	is run while we're batching the cpu state updates. */
	401	if (preload_fpu)
	402	clts();
	403
9226d125 ZA	404	/*
	405	* Leave lazy mode, flushing any hypercalls made here.
	406	* This must be done before restoring TLS segments so
	407	* the GDT and LDT are properly updated, and must be
	408	* done before math_state_restore, so the TS bit is up
	409	* to date.
	410	*/
224101ed	411	arch_end_context_switch(next_p);
9226d125	412
2fcddce1 JF	413	if (preload_fpu)
2fcddce1 JF	414	__math_state_restore();
acc20761	415
9226d125 ZA	416	/*
	417	* Restore %gs if needed (which is common)
	418	*/
	419	if (prev->gs \| next->gs)
ccbeed3a	420	lazy_load_gs(next->gs);
9226d125	421
6dbde353	422	percpu_write(current_task, next_p);
9226d125	423
1da177e4 LT	424	return prev_p;
	425	}
	426
b12bdaf1	427	int sys_clone(struct pt_regs *regs)
1da177e4	428	{
b12bdaf1 BG	429	unsigned long clone_flags;
	430	unsigned long newsp;
	431	int __user parent_tidptr, child_tidptr;
	432
	433	clone_flags = regs->bx;
	434	newsp = regs->cx;
	435	parent_tidptr = (int __user *)regs->dx;
	436	child_tidptr = (int __user *)regs->di;
1da177e4	437	if (!newsp)
253f29a4 BG	438	newsp = regs->sp;
253f29a4 BG	439	return do_fork(clone_flags, newsp, regs, 0, parent_tidptr, child_tidptr);
1da177e4 LT	440	}
1da177e4 LT	441
1da177e4 LT	442	/*
	443	* sys_execve() executes a new program.
	444	*/
b12bdaf1	445	int sys_execve(struct pt_regs *regs)
1da177e4 LT	446	{
1da177e4 LT	447	int error;
befa9e78	448	char *filename;
1da177e4	449
b12bdaf1	450	filename = getname((char __user *) regs->bx);
1da177e4 LT	451	error = PTR_ERR(filename);
	452	if (IS_ERR(filename))
	453	goto out;
b12bdaf1 BG	454	error = do_execve(filename,
	455	(char __user * __user *) regs->cx,
	456	(char __user * __user *) regs->dx,
	457	regs);
1da177e4	458	if (error == 0) {
1da177e4 LT	459	/* Make sure we don't return using sysenter.. */
	460	set_thread_flag(TIF_IRET);
	461	}
	462	putname(filename);
	463	out:
	464	return error;
	465	}
	466
	467	#define top_esp (THREAD_SIZE - sizeof(unsigned long))
	468	#define top_ebp (THREAD_SIZE - 2*sizeof(unsigned long))
	469
	470	unsigned long get_wchan(struct task_struct *p)
	471	{
65ea5b03	472	unsigned long bp, sp, ip;
1da177e4 LT	473	unsigned long stack_page;
	474	int count = 0;
	475	if (!p \|\| p == current \|\| p->state == TASK_RUNNING)
	476	return 0;
65e0fdff	477	stack_page = (unsigned long)task_stack_page(p);
faca6227	478	sp = p->thread.sp;
65ea5b03	479	if (!stack_page \|\| sp < stack_page \|\| sp > top_esp+stack_page)
1da177e4	480	return 0;
65ea5b03 PA	481	/* include/asm-i386/system.h:switch_to() pushes bp last. */
65ea5b03 PA	482	bp = (unsigned long ) sp;
1da177e4	483	do {
65ea5b03	484	if (bp < stack_page \|\| bp > top_ebp+stack_page)
1da177e4	485	return 0;
65ea5b03 PA	486	ip = (unsigned long ) (bp+4);
	487	if (!in_sched_functions(ip))
	488	return ip;
	489	bp = (unsigned long ) bp;
1da177e4 LT	490	} while (count++ < 16);
	491	return 0;
	492	}
	493