[linux-2.6-block.git] / kernel / ptrace.c

/*
 * linux/kernel/ptrace.c
 *
 * (C) Copyright 1999 Linus Torvalds
 *
 * Common interfaces for "ptrace()" which we do not want
 * to continually duplicate across every architecture.
 */

#include <linux/capability.h>
#include <linux/module.h>
#include <linux/sched.h>
#include <linux/errno.h>
#include <linux/mm.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
#include <linux/smp_lock.h>
#include <linux/ptrace.h>
#include <linux/security.h>
#include <linux/signal.h>

#include <asm/pgtable.h>
#include <asm/uaccess.h>

/*
 * ptrace a task: make the debugger its new parent and
 * move it to the ptrace list.
 *
 * Must be called with the tasklist lock write-held.
 */
void __ptrace_link(task_t *child, task_t *new_parent)
{
	BUG_ON(!list_empty(&child->ptrace_list));
	if (child->parent == new_parent)
		return;
	list_add(&child->ptrace_list, &child->parent->ptrace_children);
	remove_parent(child);
	child->parent = new_parent;
	add_parent(child);
}
 
/*
 * Turn a tracing stop into a normal stop now, since with no tracer there
 * would be no way to wake it up with SIGCONT or SIGKILL.  If there was a
 * signal sent that would resume the child, but didn't because it was in
 * TASK_TRACED, resume it now.
 * Requires that irqs be disabled.
 */
void ptrace_untrace(task_t *child)
{
	spin_lock(&child->sighand->siglock);
	if (child->state == TASK_TRACED) {
		if (child->signal->flags & SIGNAL_STOP_STOPPED) {
			child->state = TASK_STOPPED;
		} else {
			signal_wake_up(child, 1);
		}
	}
	spin_unlock(&child->sighand->siglock);
}

/*
 * unptrace a task: move it back to its original parent and
 * remove it from the ptrace list.
 *
 * Must be called with the tasklist lock write-held.
 */
void __ptrace_unlink(task_t *child)
{
	BUG_ON(!child->ptrace);

	child->ptrace = 0;
	if (!list_empty(&child->ptrace_list)) {
		list_del_init(&child->ptrace_list);
		remove_parent(child);
		child->parent = child->real_parent;
		add_parent(child);
	}

	if (child->state == TASK_TRACED)
		ptrace_untrace(child);
}

/*
 * Check that we have indeed attached to the thing..
 */
int ptrace_check_attach(struct task_struct *child, int kill)
{
	int ret = -ESRCH;

	/*
	 * We take the read lock around doing both checks to close a
	 * possible race where someone else was tracing our child and
	 * detached between these two checks.  After this locked check,
	 * we are sure that this is our traced child and that can only
	 * be changed by us so it's not changing right after this.
	 */
	read_lock(&tasklist_lock);
	if ((child->ptrace & PT_PTRACED) && child->parent == current &&
	    (!(child->ptrace & PT_ATTACHED) || child->real_parent != current)
	    && child->signal != NULL) {
		ret = 0;
		spin_lock_irq(&child->sighand->siglock);
		if (child->state == TASK_STOPPED) {
			child->state = TASK_TRACED;
		} else if (child->state != TASK_TRACED && !kill) {
			ret = -ESRCH;
		}
		spin_unlock_irq(&child->sighand->siglock);
	}
	read_unlock(&tasklist_lock);

	if (!ret && !kill) {
		wait_task_inactive(child);
	}

	/* All systems go.. */
	return ret;
}

static int may_attach(struct task_struct *task)
{
	/* May we inspect the given task?
	 * This check is used both for attaching with ptrace
	 * and for allowing access to sensitive information in /proc.
	 *
	 * ptrace_attach denies several cases that /proc allows
	 * because setting up the necessary parent/child relationship
	 * or halting the specified task is impossible.
	 */
	int dumpable = 0;
	/* Don't let security modules deny introspection */
	if (task == current)
		return 0;
	if (((current->uid != task->euid) ||
	     (current->uid != task->suid) ||
	     (current->uid != task->uid) ||
	     (current->gid != task->egid) ||
	     (current->gid != task->sgid) ||
	     (current->gid != task->gid)) && !capable(CAP_SYS_PTRACE))
		return -EPERM;
	smp_rmb();
	if (task->mm)
		dumpable = task->mm->dumpable;
	if (!dumpable && !capable(CAP_SYS_PTRACE))
		return -EPERM;

	return security_ptrace(current, task);
}

int ptrace_may_attach(struct task_struct *task)
{
	int err;
	task_lock(task);
	err = may_attach(task);
	task_unlock(task);
	return !err;
}

int ptrace_attach(struct task_struct *task)
{
	int retval;

	retval = -EPERM;
	if (task->pid <= 1)
		goto out;
	if (task->tgid == current->tgid)
		goto out;

repeat:
	/*
	 * Nasty, nasty.
	 *
	 * We want to hold both the task-lock and the
	 * tasklist_lock for writing at the same time.
	 * But that's against the rules (tasklist_lock
	 * is taken for reading by interrupts on other
	 * cpu's that may have task_lock).
	 */
	task_lock(task);
	local_irq_disable();
	if (!write_trylock(&tasklist_lock)) {
		local_irq_enable();
		task_unlock(task);
		do {
			cpu_relax();
		} while (!write_can_lock(&tasklist_lock));
		goto repeat;
	}

	if (!task->mm)
		goto bad;
	/* the same process cannot be attached many times */
	if (task->ptrace & PT_PTRACED)
		goto bad;
	retval = may_attach(task);
	if (retval)
		goto bad;

	/* Go */
	task->ptrace |= PT_PTRACED | ((task->real_parent != current)
				      ? PT_ATTACHED : 0);
	if (capable(CAP_SYS_PTRACE))
		task->ptrace |= PT_PTRACE_CAP;

	__ptrace_link(task, current);

	force_sig_specific(SIGSTOP, task);

bad:
	write_unlock_irq(&tasklist_lock);
	task_unlock(task);
out:
	return retval;
}

void __ptrace_detach(struct task_struct *child, unsigned int data)
{
	child->exit_code = data;
	/* .. re-parent .. */
	__ptrace_unlink(child);
	/* .. and wake it up. */
	if (child->exit_state != EXIT_ZOMBIE)
		wake_up_process(child);
}

int ptrace_detach(struct task_struct *child, unsigned int data)
{
	if (!valid_signal(data))
		return -EIO;

	/* Architecture-specific hardware disable .. */
	ptrace_disable(child);

	write_lock_irq(&tasklist_lock);
	if (child->ptrace)
		__ptrace_detach(child, data);
	write_unlock_irq(&tasklist_lock);

	return 0;
}

/*
 * Access another process' address space.
 * Source/target buffer must be kernel space, 
 * Do not walk the page table directly, use get_user_pages
 */

int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write)
{
	struct mm_struct *mm;
	struct vm_area_struct *vma;
	struct page *page;
	void *old_buf = buf;

	mm = get_task_mm(tsk);
	if (!mm)
		return 0;

	down_read(&mm->mmap_sem);
	/* ignore errors, just check how much was sucessfully transfered */
	while (len) {
		int bytes, ret, offset;
		void *maddr;

		ret = get_user_pages(tsk, mm, addr, 1,
				write, 1, &page, &vma);
		if (ret <= 0)
			break;

		bytes = len;
		offset = addr & (PAGE_SIZE-1);
		if (bytes > PAGE_SIZE-offset)
			bytes = PAGE_SIZE-offset;

		maddr = kmap(page);
		if (write) {
			copy_to_user_page(vma, page, addr,
					  maddr + offset, buf, bytes);
			set_page_dirty_lock(page);
		} else {
			copy_from_user_page(vma, page, addr,
					    buf, maddr + offset, bytes);
		}
		kunmap(page);
		page_cache_release(page);
		len -= bytes;
		buf += bytes;
		addr += bytes;
	}
	up_read(&mm->mmap_sem);
	mmput(mm);
	
	return buf - old_buf;
}

int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst, int len)
{
	int copied = 0;

	while (len > 0) {
		char buf[128];
		int this_len, retval;

		this_len = (len > sizeof(buf)) ? sizeof(buf) : len;
		retval = access_process_vm(tsk, src, buf, this_len, 0);
		if (!retval) {
			if (copied)
				break;
			return -EIO;
		}
		if (copy_to_user(dst, buf, retval))
			return -EFAULT;
		copied += retval;
		src += retval;
		dst += retval;
		len -= retval;			
	}
	return copied;
}

int ptrace_writedata(struct task_struct *tsk, char __user *src, unsigned long dst, int len)
{
	int copied = 0;

	while (len > 0) {
		char buf[128];
		int this_len, retval;

		this_len = (len > sizeof(buf)) ? sizeof(buf) : len;
		if (copy_from_user(buf, src, this_len))
			return -EFAULT;
		retval = access_process_vm(tsk, dst, buf, this_len, 1);
		if (!retval) {
			if (copied)
				break;
			return -EIO;
		}
		copied += retval;
		src += retval;
		dst += retval;
		len -= retval;			
	}
	return copied;
}

static int ptrace_setoptions(struct task_struct *child, long data)
{
	child->ptrace &= ~PT_TRACE_MASK;

	if (data & PTRACE_O_TRACESYSGOOD)
		child->ptrace |= PT_TRACESYSGOOD;

	if (data & PTRACE_O_TRACEFORK)
		child->ptrace |= PT_TRACE_FORK;

	if (data & PTRACE_O_TRACEVFORK)
		child->ptrace |= PT_TRACE_VFORK;

	if (data & PTRACE_O_TRACECLONE)
		child->ptrace |= PT_TRACE_CLONE;

	if (data & PTRACE_O_TRACEEXEC)
		child->ptrace |= PT_TRACE_EXEC;

	if (data & PTRACE_O_TRACEVFORKDONE)
		child->ptrace |= PT_TRACE_VFORK_DONE;

	if (data & PTRACE_O_TRACEEXIT)
		child->ptrace |= PT_TRACE_EXIT;

	return (data & ~PTRACE_O_MASK) ? -EINVAL : 0;
}

static int ptrace_getsiginfo(struct task_struct *child, siginfo_t __user * data)
{
	siginfo_t lastinfo;
	int error = -ESRCH;

	read_lock(&tasklist_lock);
	if (likely(child->sighand != NULL)) {
		error = -EINVAL;
		spin_lock_irq(&child->sighand->siglock);
		if (likely(child->last_siginfo != NULL)) {
			lastinfo = *child->last_siginfo;
			error = 0;
		}
		spin_unlock_irq(&child->sighand->siglock);
	}
	read_unlock(&tasklist_lock);
	if (!error)
		return copy_siginfo_to_user(data, &lastinfo);
	return error;
}

static int ptrace_setsiginfo(struct task_struct *child, siginfo_t __user * data)
{
	siginfo_t newinfo;
	int error = -ESRCH;

	if (copy_from_user(&newinfo, data, sizeof (siginfo_t)))
		return -EFAULT;

	read_lock(&tasklist_lock);
	if (likely(child->sighand != NULL)) {
		error = -EINVAL;
		spin_lock_irq(&child->sighand->siglock);
		if (likely(child->last_siginfo != NULL)) {
			*child->last_siginfo = newinfo;
			error = 0;
		}
		spin_unlock_irq(&child->sighand->siglock);
	}
	read_unlock(&tasklist_lock);
	return error;
}

int ptrace_request(struct task_struct *child, long request,
		   long addr, long data)
{
	int ret = -EIO;

	switch (request) {
#ifdef PTRACE_OLDSETOPTIONS
	case PTRACE_OLDSETOPTIONS:
#endif
	case PTRACE_SETOPTIONS:
		ret = ptrace_setoptions(child, data);
		break;
	case PTRACE_GETEVENTMSG:
		ret = put_user(child->ptrace_message, (unsigned long __user *) data);
		break;
	case PTRACE_GETSIGINFO:
		ret = ptrace_getsiginfo(child, (siginfo_t __user *) data);
		break;
	case PTRACE_SETSIGINFO:
		ret = ptrace_setsiginfo(child, (siginfo_t __user *) data);
		break;
	default:
		break;
	}

	return ret;
}

/**
 * ptrace_traceme  --  helper for PTRACE_TRACEME
 *
 * Performs checks and sets PT_PTRACED.
 * Should be used by all ptrace implementations for PTRACE_TRACEME.
 */
int ptrace_traceme(void)
{
	int ret = -EPERM;

	/*
	 * Are we already being traced?
	 */
	task_lock(current);
	if (!(current->ptrace & PT_PTRACED)) {
		ret = security_ptrace(current->parent, current);
		/*
		 * Set the ptrace bit in the process ptrace flags.
		 */
		if (!ret)
			current->ptrace |= PT_PTRACED;
	}
	task_unlock(current);
	return ret;
}

/**
 * ptrace_get_task_struct  --  grab a task struct reference for ptrace
 * @pid:       process id to grab a task_struct reference of
 *
 * This function is a helper for ptrace implementations.  It checks
 * permissions and then grabs a task struct for use of the actual
 * ptrace implementation.
 *
 * Returns the task_struct for @pid or an ERR_PTR() on failure.
 */
struct task_struct *ptrace_get_task_struct(pid_t pid)
{
	struct task_struct *child;

	/*
	 * Tracing init is not allowed.
	 */
	if (pid == 1)
		return ERR_PTR(-EPERM);

	read_lock(&tasklist_lock);
	child = find_task_by_pid(pid);
	if (child)
		get_task_struct(child);
	read_unlock(&tasklist_lock);
	if (!child)
		return ERR_PTR(-ESRCH);
	return child;
}

#ifndef __ARCH_SYS_PTRACE
asmlinkage long sys_ptrace(long request, long pid, long addr, long data)
{
	struct task_struct *child;
	long ret;

	/*
	 * This lock_kernel fixes a subtle race with suid exec
	 */
	lock_kernel();
	if (request == PTRACE_TRACEME) {
		ret = ptrace_traceme();
		goto out;
	}

	child = ptrace_get_task_struct(pid);
	if (IS_ERR(child)) {
		ret = PTR_ERR(child);
		goto out;
	}

	if (request == PTRACE_ATTACH) {
		ret = ptrace_attach(child);
		goto out_put_task_struct;
	}

	ret = ptrace_check_attach(child, request == PTRACE_KILL);
	if (ret < 0)
		goto out_put_task_struct;

	ret = arch_ptrace(child, request, addr, data);
	if (ret < 0)
		goto out_put_task_struct;

 out_put_task_struct:
	put_task_struct(child);
 out:
	unlock_kernel();
	return ret;
}
#endif /* __ARCH_SYS_PTRACE */
Commit	Line	Data
1da177e4 LT	1	/*
	2	* linux/kernel/ptrace.c
	3	*
	4	* (C) Copyright 1999 Linus Torvalds
	5	*
	6	* Common interfaces for "ptrace()" which we do not want
	7	* to continually duplicate across every architecture.
	8	*/
	9
c59ede7b	10	#include <linux/capability.h>
1da177e4 LT	11	#include <linux/module.h>
	12	#include <linux/sched.h>
	13	#include <linux/errno.h>
	14	#include <linux/mm.h>
	15	#include <linux/highmem.h>
	16	#include <linux/pagemap.h>
	17	#include <linux/smp_lock.h>
	18	#include <linux/ptrace.h>
	19	#include <linux/security.h>
7ed20e1a	20	#include <linux/signal.h>
1da177e4 LT	21
	22	#include <asm/pgtable.h>
	23	#include <asm/uaccess.h>
	24
	25	/*
	26	* ptrace a task: make the debugger its new parent and
	27	* move it to the ptrace list.
	28	*
	29	* Must be called with the tasklist lock write-held.
	30	*/
	31	void __ptrace_link(task_t child, task_t new_parent)
	32	{
524223ca	33	BUG_ON(!list_empty(&child->ptrace_list));
1da177e4 LT	34	if (child->parent == new_parent)
	35	return;
	36	list_add(&child->ptrace_list, &child->parent->ptrace_children);
9b678ece	37	remove_parent(child);
1da177e4	38	child->parent = new_parent;
9b678ece	39	add_parent(child);
1da177e4 LT	40	}
	41
	42	/*
	43	* Turn a tracing stop into a normal stop now, since with no tracer there
	44	* would be no way to wake it up with SIGCONT or SIGKILL. If there was a
	45	* signal sent that would resume the child, but didn't because it was in
	46	* TASK_TRACED, resume it now.
	47	* Requires that irqs be disabled.
	48	*/
	49	void ptrace_untrace(task_t *child)
	50	{
	51	spin_lock(&child->sighand->siglock);
	52	if (child->state == TASK_TRACED) {
	53	if (child->signal->flags & SIGNAL_STOP_STOPPED) {
	54	child->state = TASK_STOPPED;
	55	} else {
	56	signal_wake_up(child, 1);
	57	}
	58	}
	59	spin_unlock(&child->sighand->siglock);
	60	}
	61
	62	/*
	63	* unptrace a task: move it back to its original parent and
	64	* remove it from the ptrace list.
	65	*
	66	* Must be called with the tasklist lock write-held.
	67	*/
	68	void __ptrace_unlink(task_t *child)
	69	{
5ecfbae0 ON	70	BUG_ON(!child->ptrace);
5ecfbae0 ON	71
1da177e4 LT	72	child->ptrace = 0;
	73	if (!list_empty(&child->ptrace_list)) {
	74	list_del_init(&child->ptrace_list);
9b678ece	75	remove_parent(child);
1da177e4	76	child->parent = child->real_parent;
9b678ece	77	add_parent(child);
1da177e4 LT	78	}
1da177e4 LT	79
e57a5059 RM	80	if (child->state == TASK_TRACED)
e57a5059 RM	81	ptrace_untrace(child);
1da177e4 LT	82	}
	83
	84	/*
	85	* Check that we have indeed attached to the thing..
	86	*/
	87	int ptrace_check_attach(struct task_struct *child, int kill)
	88	{
	89	int ret = -ESRCH;
	90
	91	/*
	92	* We take the read lock around doing both checks to close a
	93	* possible race where someone else was tracing our child and
	94	* detached between these two checks. After this locked check,
	95	* we are sure that this is our traced child and that can only
	96	* be changed by us so it's not changing right after this.
	97	*/
	98	read_lock(&tasklist_lock);
	99	if ((child->ptrace & PT_PTRACED) && child->parent == current &&
	100	(!(child->ptrace & PT_ATTACHED) \|\| child->real_parent != current)
	101	&& child->signal != NULL) {
	102	ret = 0;
	103	spin_lock_irq(&child->sighand->siglock);
	104	if (child->state == TASK_STOPPED) {
	105	child->state = TASK_TRACED;
	106	} else if (child->state != TASK_TRACED && !kill) {
	107	ret = -ESRCH;
	108	}
	109	spin_unlock_irq(&child->sighand->siglock);
	110	}
	111	read_unlock(&tasklist_lock);
	112
	113	if (!ret && !kill) {
	114	wait_task_inactive(child);
	115	}
	116
	117	/* All systems go.. */
	118	return ret;
	119	}
	120
ab8d11be MS	121	static int may_attach(struct task_struct *task)
ab8d11be MS	122	{
df26c40e EB	123	/* May we inspect the given task?
	124	* This check is used both for attaching with ptrace
	125	* and for allowing access to sensitive information in /proc.
	126	*
	127	* ptrace_attach denies several cases that /proc allows
	128	* because setting up the necessary parent/child relationship
	129	* or halting the specified task is impossible.
	130	*/
	131	int dumpable = 0;
	132	/* Don't let security modules deny introspection */
	133	if (task == current)
	134	return 0;
ab8d11be MS	135	if (((current->uid != task->euid) \|\|
	136	(current->uid != task->suid) \|\|
	137	(current->uid != task->uid) \|\|
	138	(current->gid != task->egid) \|\|
	139	(current->gid != task->sgid) \|\|
	140	(current->gid != task->gid)) && !capable(CAP_SYS_PTRACE))
	141	return -EPERM;
	142	smp_rmb();
df26c40e EB	143	if (task->mm)
	144	dumpable = task->mm->dumpable;
	145	if (!dumpable && !capable(CAP_SYS_PTRACE))
ab8d11be MS	146	return -EPERM;
	147
	148	return security_ptrace(current, task);
	149	}
	150
	151	int ptrace_may_attach(struct task_struct *task)
	152	{
	153	int err;
	154	task_lock(task);
	155	err = may_attach(task);
	156	task_unlock(task);
	157	return !err;
	158	}
	159
1da177e4 LT	160	int ptrace_attach(struct task_struct *task)
	161	{
	162	int retval;
f5b40e36	163
1da177e4 LT	164	retval = -EPERM;
1da177e4 LT	165	if (task->pid <= 1)
f5b40e36	166	goto out;
28d838cc	167	if (task->tgid == current->tgid)
f5b40e36 LT	168	goto out;
f5b40e36 LT	169
f358166a LT	170	repeat:
	171	/*
	172	* Nasty, nasty.
	173	*
	174	* We want to hold both the task-lock and the
	175	* tasklist_lock for writing at the same time.
	176	* But that's against the rules (tasklist_lock
	177	* is taken for reading by interrupts on other
	178	* cpu's that may have task_lock).
	179	*/
f5b40e36	180	task_lock(task);
f358166a LT	181	local_irq_disable();
	182	if (!write_trylock(&tasklist_lock)) {
	183	local_irq_enable();
	184	task_unlock(task);
	185	do {
	186	cpu_relax();
	187	} while (!write_can_lock(&tasklist_lock));
	188	goto repeat;
	189	}
f5b40e36	190
df26c40e EB	191	if (!task->mm)
df26c40e EB	192	goto bad;
1da177e4 LT	193	/* the same process cannot be attached many times */
	194	if (task->ptrace & PT_PTRACED)
	195	goto bad;
ab8d11be	196	retval = may_attach(task);
1da177e4 LT	197	if (retval)
	198	goto bad;
	199
	200	/* Go */
	201	task->ptrace \|= PT_PTRACED \| ((task->real_parent != current)
	202	? PT_ATTACHED : 0);
	203	if (capable(CAP_SYS_PTRACE))
	204	task->ptrace \|= PT_PTRACE_CAP;
1da177e4	205
1da177e4	206	__ptrace_link(task, current);
1da177e4 LT	207
1da177e4 LT	208	force_sig_specific(SIGSTOP, task);
1da177e4 LT	209
1da177e4 LT	210	bad:
f5b40e36	211	write_unlock_irq(&tasklist_lock);
1da177e4	212	task_unlock(task);
f5b40e36	213	out:
1da177e4 LT	214	return retval;
	215	}
	216
5ecfbae0 ON	217	void __ptrace_detach(struct task_struct *child, unsigned int data)
	218	{
	219	child->exit_code = data;
	220	/* .. re-parent .. */
	221	__ptrace_unlink(child);
	222	/* .. and wake it up. */
	223	if (child->exit_state != EXIT_ZOMBIE)
	224	wake_up_process(child);
	225	}
	226
1da177e4 LT	227	int ptrace_detach(struct task_struct *child, unsigned int data)
1da177e4 LT	228	{
7ed20e1a	229	if (!valid_signal(data))
5ecfbae0	230	return -EIO;
1da177e4 LT	231
	232	/* Architecture-specific hardware disable .. */
	233	ptrace_disable(child);
	234
1da177e4	235	write_lock_irq(&tasklist_lock);
5ecfbae0 ON	236	if (child->ptrace)
5ecfbae0 ON	237	__ptrace_detach(child, data);
1da177e4 LT	238	write_unlock_irq(&tasklist_lock);
	239
	240	return 0;
	241	}
	242
	243	/*
	244	* Access another process' address space.
	245	* Source/target buffer must be kernel space,
	246	* Do not walk the page table directly, use get_user_pages
	247	*/
	248
	249	int access_process_vm(struct task_struct tsk, unsigned long addr, void buf, int len, int write)
	250	{
	251	struct mm_struct *mm;
	252	struct vm_area_struct *vma;
	253	struct page *page;
	254	void *old_buf = buf;
	255
	256	mm = get_task_mm(tsk);
	257	if (!mm)
	258	return 0;
	259
	260	down_read(&mm->mmap_sem);
	261	/* ignore errors, just check how much was sucessfully transfered */
	262	while (len) {
	263	int bytes, ret, offset;
	264	void *maddr;
	265
	266	ret = get_user_pages(tsk, mm, addr, 1,
	267	write, 1, &page, &vma);
	268	if (ret <= 0)
	269	break;
	270
	271	bytes = len;
	272	offset = addr & (PAGE_SIZE-1);
	273	if (bytes > PAGE_SIZE-offset)
	274	bytes = PAGE_SIZE-offset;
	275
	276	maddr = kmap(page);
	277	if (write) {
	278	copy_to_user_page(vma, page, addr,
	279	maddr + offset, buf, bytes);
16bf1348	280	set_page_dirty_lock(page);
1da177e4 LT	281	} else {
	282	copy_from_user_page(vma, page, addr,
	283	buf, maddr + offset, bytes);
	284	}
	285	kunmap(page);
	286	page_cache_release(page);
	287	len -= bytes;
	288	buf += bytes;
	289	addr += bytes;
	290	}
	291	up_read(&mm->mmap_sem);
	292	mmput(mm);
	293
	294	return buf - old_buf;
	295	}
	296
	297	int ptrace_readdata(struct task_struct tsk, unsigned long src, char __user dst, int len)
	298	{
	299	int copied = 0;
	300
	301	while (len > 0) {
	302	char buf[128];
	303	int this_len, retval;
	304
	305	this_len = (len > sizeof(buf)) ? sizeof(buf) : len;
	306	retval = access_process_vm(tsk, src, buf, this_len, 0);
	307	if (!retval) {
	308	if (copied)
	309	break;
	310	return -EIO;
	311	}
	312	if (copy_to_user(dst, buf, retval))
	313	return -EFAULT;
	314	copied += retval;
	315	src += retval;
	316	dst += retval;
	317	len -= retval;
	318	}
	319	return copied;
	320	}
	321
	322	int ptrace_writedata(struct task_struct tsk, char __user src, unsigned long dst, int len)
	323	{
	324	int copied = 0;
	325
	326	while (len > 0) {
	327	char buf[128];
	328	int this_len, retval;
	329
	330	this_len = (len > sizeof(buf)) ? sizeof(buf) : len;
	331	if (copy_from_user(buf, src, this_len))
	332	return -EFAULT;
	333	retval = access_process_vm(tsk, dst, buf, this_len, 1);
	334	if (!retval) {
	335	if (copied)
	336	break;
	337	return -EIO;
	338	}
	339	copied += retval;
	340	src += retval;
	341	dst += retval;
	342	len -= retval;
	343	}
	344	return copied;
345	}
346
347	static int ptrace_setoptions(struct task_struct *child, long data)
348	{
349	child->ptrace &= ~PT_TRACE_MASK;
350
351	if (data & PTRACE_O_TRACESYSGOOD)
352	child->ptrace \|= PT_TRACESYSGOOD;
353
354	if (data & PTRACE_O_TRACEFORK)
355	child->ptrace \|= PT_TRACE_FORK;
356
357	if (data & PTRACE_O_TRACEVFORK)
358	child->ptrace \|= PT_TRACE_VFORK;
359
360	if (data & PTRACE_O_TRACECLONE)
361	child->ptrace \|= PT_TRACE_CLONE;
362
363	if (data & PTRACE_O_TRACEEXEC)
364	child->ptrace \|= PT_TRACE_EXEC;
365
366	if (data & PTRACE_O_TRACEVFORKDONE)
367	child->ptrace \|= PT_TRACE_VFORK_DONE;
368
369	if (data & PTRACE_O_TRACEEXIT)
370	child->ptrace \|= PT_TRACE_EXIT;
371
372	return (data & ~PTRACE_O_MASK) ? -EINVAL : 0;
373	}
374
375	static int ptrace_getsiginfo(struct task_struct child, siginfo_t __user data)
376	{
377	siginfo_t lastinfo;
378	int error = -ESRCH;
379
380	read_lock(&tasklist_lock);
381	if (likely(child->sighand != NULL)) {
382	error = -EINVAL;
383	spin_lock_irq(&child->sighand->siglock);
384	if (likely(child->last_siginfo != NULL)) {
385	lastinfo = *child->last_siginfo;
386	error = 0;
387	}
388	spin_unlock_irq(&child->sighand->siglock);
389	}
390	read_unlock(&tasklist_lock);
391	if (!error)
392	return copy_siginfo_to_user(data, &lastinfo);
393	return error;
394	}
395
396	static int ptrace_setsiginfo(struct task_struct child, siginfo_t __user data)
397	{
398	siginfo_t newinfo;
399	int error = -ESRCH;
400
401	if (copy_from_user(&newinfo, data, sizeof (siginfo_t)))
402	return -EFAULT;
403
404	read_lock(&tasklist_lock);
405	if (likely(child->sighand != NULL)) {
406	error = -EINVAL;
407	spin_lock_irq(&child->sighand->siglock);
408	if (likely(child->last_siginfo != NULL)) {
409	*child->last_siginfo = newinfo;
410	error = 0;
411	}
412	spin_unlock_irq(&child->sighand->siglock);
413	}
414	read_unlock(&tasklist_lock);
415	return error;
416	}
417
418	int ptrace_request(struct task_struct *child, long request,
419	long addr, long data)
420	{
421	int ret = -EIO;
422
423	switch (request) {
424	#ifdef PTRACE_OLDSETOPTIONS
425	case PTRACE_OLDSETOPTIONS:
426	#endif
427	case PTRACE_SETOPTIONS:
428	ret = ptrace_setoptions(child, data);
429	break;
430	case PTRACE_GETEVENTMSG:
431	ret = put_user(child->ptrace_message, (unsigned long __user *) data);
432	break;
433	case PTRACE_GETSIGINFO:
434	ret = ptrace_getsiginfo(child, (siginfo_t __user *) data);
435	break;
436	case PTRACE_SETSIGINFO:
437	ret = ptrace_setsiginfo(child, (siginfo_t __user *) data);
438	break;
439	default:
440	break;
441	}
442
443	return ret;
444	}
481bed45	445
6b9c7ed8 CH	446	/**
	447	* ptrace_traceme -- helper for PTRACE_TRACEME
	448	*
	449	* Performs checks and sets PT_PTRACED.
	450	* Should be used by all ptrace implementations for PTRACE_TRACEME.
	451	*/
	452	int ptrace_traceme(void)
481bed45	453	{
f5b40e36	454	int ret = -EPERM;
481bed45 CH	455
481bed45 CH	456	/*
6b9c7ed8 CH	457	* Are we already being traced?
6b9c7ed8 CH	458	*/
f5b40e36 LT	459	task_lock(current);
	460	if (!(current->ptrace & PT_PTRACED)) {
	461	ret = security_ptrace(current->parent, current);
	462	/*
	463	* Set the ptrace bit in the process ptrace flags.
	464	*/
	465	if (!ret)
	466	current->ptrace \|= PT_PTRACED;
	467	}
	468	task_unlock(current);
	469	return ret;
6b9c7ed8	470	}
481bed45	471
6b9c7ed8 CH	472	/**
	473	* ptrace_get_task_struct -- grab a task struct reference for ptrace
	474	* @pid: process id to grab a task_struct reference of
	475	*
	476	* This function is a helper for ptrace implementations. It checks
	477	* permissions and then grabs a task struct for use of the actual
	478	* ptrace implementation.
	479	*
	480	* Returns the task_struct for @pid or an ERR_PTR() on failure.
	481	*/
	482	struct task_struct *ptrace_get_task_struct(pid_t pid)
	483	{
	484	struct task_struct *child;
481bed45 CH	485
481bed45 CH	486	/*
6b9c7ed8	487	* Tracing init is not allowed.
481bed45 CH	488	*/
481bed45 CH	489	if (pid == 1)
6b9c7ed8	490	return ERR_PTR(-EPERM);
481bed45	491
481bed45 CH	492	read_lock(&tasklist_lock);
	493	child = find_task_by_pid(pid);
	494	if (child)
	495	get_task_struct(child);
	496	read_unlock(&tasklist_lock);
	497	if (!child)
6b9c7ed8 CH	498	return ERR_PTR(-ESRCH);
6b9c7ed8 CH	499	return child;
481bed45 CH	500	}
481bed45 CH	501
6b9c7ed8	502	#ifndef __ARCH_SYS_PTRACE
481bed45 CH	503	asmlinkage long sys_ptrace(long request, long pid, long addr, long data)
	504	{
	505	struct task_struct *child;
	506	long ret;
	507
	508	/*
	509	* This lock_kernel fixes a subtle race with suid exec
	510	*/
	511	lock_kernel();
6b9c7ed8 CH	512	if (request == PTRACE_TRACEME) {
6b9c7ed8 CH	513	ret = ptrace_traceme();
481bed45	514	goto out;
6b9c7ed8 CH	515	}
	516
	517	child = ptrace_get_task_struct(pid);
	518	if (IS_ERR(child)) {
	519	ret = PTR_ERR(child);
	520	goto out;
	521	}
481bed45 CH	522
	523	if (request == PTRACE_ATTACH) {
	524	ret = ptrace_attach(child);
005f18df	525	goto out_put_task_struct;
481bed45 CH	526	}
	527
	528	ret = ptrace_check_attach(child, request == PTRACE_KILL);
	529	if (ret < 0)
	530	goto out_put_task_struct;
	531
	532	ret = arch_ptrace(child, request, addr, data);
	533	if (ret < 0)
	534	goto out_put_task_struct;
	535
	536	out_put_task_struct:
	537	put_task_struct(child);
	538	out:
	539	unlock_kernel();
	540	return ret;
	541	}
	542	#endif /* __ARCH_SYS_PTRACE */