summaryrefslogtreecommitdiffstats
path: root/sys/kern/sys_process.c
blob: 6faf442768ee45c31acd6a59d27cdf5e7b5aa12e (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
/*
 * Copyright (c) 1994, Sean Eric Fagan
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by Sean Eric Fagan.
 * 4. The name of the author may not be used to endorse or promote products
 *    derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 * $FreeBSD$
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/sysproto.h>
#include <sys/proc.h>
#include <sys/vnode.h>
#include <sys/ptrace.h>
#include <sys/sx.h>
#include <sys/user.h>

#include <machine/reg.h>

#include <vm/vm.h>
#include <vm/pmap.h>
#include <vm/vm_extern.h>
#include <vm/vm_map.h>
#include <vm/vm_kern.h>
#include <vm/vm_object.h>
#include <vm/vm_page.h>

/*
 * Functions implemented using PROC_ACTION():
 *
 * proc_read_regs(proc, regs)
 *	Get the current user-visible register set from the process
 *	and copy it into the regs structure (<machine/reg.h>).
 *	The process is stopped at the time read_regs is called.
 *
 * proc_write_regs(proc, regs)
 *	Update the current register set from the passed in regs
 *	structure.  Take care to avoid clobbering special CPU
 *	registers or privileged bits in the PSL.
 *	Depending on the architecture this may have fix-up work to do,
 *	especially if the IAR or PCW are modified.
 *	The process is stopped at the time write_regs is called.
 *
 * proc_read_fpregs, proc_write_fpregs
 *	deal with the floating point register set, otherwise as above.
 *
 * proc_read_dbregs, proc_write_dbregs
 *	deal with the processor debug register set, otherwise as above.
 *
 * proc_sstep(proc)
 *	Arrange for the process to trap after executing a single instruction.
 */

#define	PROC_ACTION(action) do {					\
	int error;							\
									\
	mtx_lock_spin(&sched_lock);					\
	if ((td->td_proc->p_sflag & PS_INMEM) == 0)			\
		error = EIO;						\
	else								\
		error = (action);					\
	mtx_unlock_spin(&sched_lock);					\
	return (error);							\
} while(0)
	
int
proc_read_regs(struct thread *td, struct reg *regs)
{

	PROC_ACTION(fill_regs(td, regs));
}

int
proc_write_regs(struct thread *td, struct reg *regs)
{

	PROC_ACTION(set_regs(td, regs));
}

int
proc_read_dbregs(struct thread *td, struct dbreg *dbregs)
{

	PROC_ACTION(fill_dbregs(td, dbregs));
}

int
proc_write_dbregs(struct thread *td, struct dbreg *dbregs)
{

	PROC_ACTION(set_dbregs(td, dbregs));
}

/*
 * Ptrace doesn't support fpregs at all, and there are no security holes
 * or translations for fpregs, so we can just copy them.
 */
int
proc_read_fpregs(struct thread *td, struct fpreg *fpregs)
{

	PROC_ACTION(fill_fpregs(td, fpregs));
}

int
proc_write_fpregs(struct thread *td, struct fpreg *fpregs)
{

	PROC_ACTION(set_fpregs(td, fpregs));
}

int
proc_sstep(struct thread *td)
{

	PROC_ACTION(ptrace_single_step(td));
}

int
proc_rwmem(struct proc *p, struct uio *uio)
{
	struct vmspace *vm;
	vm_map_t map;
	vm_object_t object = NULL;
	vm_offset_t pageno = 0;		/* page number */
	vm_prot_t reqprot;
	vm_offset_t kva;
	int error, writing;

	GIANT_REQUIRED;

	/*
	 * if the vmspace is in the midst of being deallocated or the
	 * process is exiting, don't try to grab anything.  The page table
	 * usage in that process can be messed up.
	 */
	vm = p->p_vmspace;
	if ((p->p_flag & P_WEXIT))
		return (EFAULT);
	if (vm->vm_refcnt < 1)
		return (EFAULT);
	++vm->vm_refcnt;
	/*
	 * The map we want...
	 */
	map = &vm->vm_map;

	writing = uio->uio_rw == UIO_WRITE;
	reqprot = writing ? (VM_PROT_WRITE | VM_PROT_OVERRIDE_WRITE) :
	    VM_PROT_READ;

	kva = kmem_alloc_pageable(kernel_map, PAGE_SIZE);

	/*
	 * Only map in one page at a time.  We don't have to, but it
	 * makes things easier.  This way is trivial - right?
	 */
	do {
		vm_map_t tmap;
		vm_offset_t uva;
		int page_offset;		/* offset into page */
		vm_map_entry_t out_entry;
		vm_prot_t out_prot;
		boolean_t wired;
		vm_pindex_t pindex;
		u_int len;
		vm_page_t m;

		object = NULL;

		uva = (vm_offset_t)uio->uio_offset;

		/*
		 * Get the page number of this segment.
		 */
		pageno = trunc_page(uva);
		page_offset = uva - pageno;

		/*
		 * How many bytes to copy
		 */
		len = min(PAGE_SIZE - page_offset, uio->uio_resid);

		/*
		 * Fault the page on behalf of the process
		 */
		error = vm_fault(map, pageno, reqprot, VM_FAULT_NORMAL);
		if (error) {
			error = EFAULT;
			break;
		}

		/*
		 * Now we need to get the page.  out_entry, out_prot, wired,
		 * and single_use aren't used.  One would think the vm code
		 * would be a *bit* nicer...  We use tmap because
		 * vm_map_lookup() can change the map argument.
		 */
		tmap = map;
		error = vm_map_lookup(&tmap, pageno, reqprot, &out_entry,
		    &object, &pindex, &out_prot, &wired);

		if (error) {
			error = EFAULT;

			/*
			 * Make sure that there is no residue in 'object' from
			 * an error return on vm_map_lookup.
			 */
			object = NULL;

			break;
		}

		m = vm_page_lookup(object, pindex);

		/* Allow fallback to backing objects if we are reading */

		while (m == NULL && !writing && object->backing_object) {

			pindex += OFF_TO_IDX(object->backing_object_offset);
			object = object->backing_object;
			
			m = vm_page_lookup(object, pindex);
		}

		if (m == NULL) {
			error = EFAULT;

			/*
			 * Make sure that there is no residue in 'object' from
			 * an error return on vm_map_lookup.
			 */
			object = NULL;

			vm_map_lookup_done(tmap, out_entry);

			break;
		}

		/*
		 * Wire the page into memory
		 */
		vm_page_wire(m);

		/*
		 * We're done with tmap now.
		 * But reference the object first, so that we won't loose
		 * it.
		 */
		vm_object_reference(object);
		vm_map_lookup_done(tmap, out_entry);

		pmap_qenter(kva, &m, 1);

		/*
		 * Now do the i/o move.
		 */
		error = uiomove((caddr_t)(kva + page_offset), len, uio);

		pmap_qremove(kva, 1);

		/*
		 * release the page and the object
		 */
		vm_page_unwire(m, 1);
		vm_object_deallocate(object);

		object = NULL;

	} while (error == 0 && uio->uio_resid > 0);

	if (object)
		vm_object_deallocate(object);

	kmem_free(kernel_map, kva, PAGE_SIZE);
	vmspace_free(vm);
	return (error);
}

/*
 * Process debugging system call.
 */
#ifndef _SYS_SYSPROTO_H_
struct ptrace_args {
	int	req;
	pid_t	pid;
	caddr_t	addr;
	int	data;
};
#endif

int
ptrace(struct thread *td, struct ptrace_args *uap)
{
	struct iovec iov;
	struct uio uio;
	/*
	 * XXX this obfuscation is to reduce stack usage, but the register
	 * structs may be too large to put on the stack anyway.
	 */
	union {
		struct ptrace_io_desc piod;
		struct dbreg dbreg;
		struct fpreg fpreg;
		struct reg reg;
	} r;
	struct proc *curp, *p, *pp;
	struct thread *td2;
	int error, write;
	int proctree_locked = 0;

	curp = td->td_proc;

	/*
	 * Do copyin() early before getting locks and lock proctree before
	 * locking the process.
	 */
	switch (uap->req) {
	case PT_TRACE_ME:
	case PT_ATTACH:
	case PT_STEP:
	case PT_CONTINUE:
	case PT_DETACH:
		sx_xlock(&proctree_lock);
		proctree_locked = 1;
		break;
#ifdef PT_SETREGS
	case PT_SETREGS:
		error = copyin(uap->addr, &r.reg, sizeof r.reg);
		if (error)
			return (error);
		break;
#endif /* PT_SETREGS */
#ifdef PT_SETFPREGS
	case PT_SETFPREGS:
		error = copyin(uap->addr, &r.fpreg, sizeof r.fpreg);
		if (error)
			return (error);
		break;
#endif /* PT_SETFPREGS */
#ifdef PT_SETDBREGS
	case PT_SETDBREGS:
		error = copyin(uap->addr, &r.dbreg, sizeof r.dbreg);
		if (error)
			return (error);
		break;
#endif /* PT_SETDBREGS */
	default:
		break;
	}
		
	write = 0;
	if (uap->req == PT_TRACE_ME) {
		p = td->td_proc;
		PROC_LOCK(p);
	} else {
		if ((p = pfind(uap->pid)) == NULL) {
			if (proctree_locked)
				sx_xunlock(&proctree_lock);
			return (ESRCH);
		}
	}
	if (p_cansee(td->td_proc, p)) {
		error = ESRCH;
		goto fail;
	}

	if ((error = p_candebug(td->td_proc, p)) != 0)
		goto fail;

	/*
	 * System processes can't be debugged.
	 */
	if ((p->p_flag & P_SYSTEM) != 0) {
		error = EINVAL;
		goto fail;
	}
	
	/*
	 * Permissions check
	 */
	switch (uap->req) {
	case PT_TRACE_ME:
		/* Always legal. */
		break;

	case PT_ATTACH:
		/* Self */
		if (p->p_pid == td->td_proc->p_pid) {
			error = EINVAL;
			goto fail;
		}

		/* Already traced */
		if (p->p_flag & P_TRACED) {
			error = EBUSY;
			goto fail;
		}

		/* Can't trace an ancestor if you're being traced. */
		if (curp->p_flag & P_TRACED) {
			for (pp = curp->p_pptr; pp != NULL; pp = pp->p_pptr) {
				if (pp == p) {
					error = EINVAL;
					goto fail;
				}
			}
		}


		/* OK */
		break;

	case PT_READ_I:
	case PT_READ_D:
	case PT_WRITE_I:
	case PT_WRITE_D:
	case PT_IO:
	case PT_CONTINUE:
	case PT_KILL:
	case PT_STEP:
	case PT_DETACH:
	case PT_GETREGS:
	case PT_SETREGS:
	case PT_GETFPREGS:
	case PT_SETFPREGS:
	case PT_GETDBREGS:
	case PT_SETDBREGS:
		/* not being traced... */
		if ((p->p_flag & P_TRACED) == 0) {
			error = EPERM;
			goto fail;
		}

		/* not being traced by YOU */
		if (p->p_pptr != td->td_proc) {
			error = EBUSY;
			goto fail;
		}

		/* not currently stopped */
		if (p->p_stat != SSTOP || (p->p_flag & P_WAITED) == 0) {
			error = EBUSY;
			goto fail;
		}

		/* OK */
		break;

	default:
		error = EINVAL;
		goto fail;
	}

	td2 = FIRST_THREAD_IN_PROC(p);
#ifdef FIX_SSTEP
	/*
	 * Single step fixup ala procfs
	 */
	FIX_SSTEP(td2);			/* XXXKSE */
#endif

	/*
	 * Actually do the requests
	 */

	td->td_retval[0] = 0;

	switch (uap->req) {
	case PT_TRACE_ME:
		/* set my trace flag and "owner" so it can read/write me */
		p->p_flag |= P_TRACED;
		p->p_oppid = p->p_pptr->p_pid;
		PROC_UNLOCK(p);
		sx_xunlock(&proctree_lock);
		return (0);

	case PT_ATTACH:
		/* security check done above */
		p->p_flag |= P_TRACED;
		p->p_oppid = p->p_pptr->p_pid;
		if (p->p_pptr != td->td_proc)
			proc_reparent(p, td->td_proc);
		uap->data = SIGSTOP;
		goto sendsig;	/* in PT_CONTINUE below */

	case PT_STEP:
	case PT_CONTINUE:
	case PT_DETACH:
		/* XXX uap->data is used even in the PT_STEP case. */
		if ((uap->req != PT_STEP) && ((unsigned)uap->data >= NSIG)) {
			error = EINVAL;
			goto fail;
		}

		_PHOLD(p);

		if (uap->req == PT_STEP) {
			error = ptrace_single_step(td2);
			if (error) {
				_PRELE(p);
				goto fail;
			}
		}

		if (uap->addr != (caddr_t)1) {
			fill_kinfo_proc(p, &p->p_uarea->u_kproc);
			error = ptrace_set_pc(td2,
			    (u_long)(uintfptr_t)uap->addr);
			if (error) {
				_PRELE(p);
				goto fail;
			}
		}
		_PRELE(p);

		if (uap->req == PT_DETACH) {
			/* reset process parent */
			if (p->p_oppid != p->p_pptr->p_pid) {
				struct proc *pp;

				PROC_UNLOCK(p);
				pp = pfind(p->p_oppid);
				if (pp == NULL)
					pp = initproc;
				else
					PROC_UNLOCK(pp);
				PROC_LOCK(p);
				proc_reparent(p, pp);
			}
			p->p_flag &= ~(P_TRACED | P_WAITED);
			p->p_oppid = 0;

			/* should we send SIGCHLD? */
		}

	sendsig:
		if (proctree_locked)
			sx_xunlock(&proctree_lock);
		/* deliver or queue signal */
		if (p->p_stat == SSTOP) {
			p->p_xstat = uap->data;
			mtx_lock_spin(&sched_lock);
			setrunnable(td2);	/* XXXKSE */
			mtx_unlock_spin(&sched_lock);
		} else if (uap->data)		      
			psignal(p, uap->data);
		PROC_UNLOCK(p);
		
		return (0);

	case PT_WRITE_I:
	case PT_WRITE_D:
		write = 1;
		/* fallthrough */
	case PT_READ_I:
	case PT_READ_D:
		PROC_UNLOCK(p);
		/* write = 0 set above */
		iov.iov_base = write ? (caddr_t)&uap->data :
		    (caddr_t)td->td_retval;
		iov.iov_len = sizeof(int);
		uio.uio_iov = &iov;
		uio.uio_iovcnt = 1;
		uio.uio_offset = (off_t)(uintptr_t)uap->addr;
		uio.uio_resid = sizeof(int);
		uio.uio_segflg = UIO_SYSSPACE;	/* i.e.: the uap */
		uio.uio_rw = write ? UIO_WRITE : UIO_READ;
		uio.uio_td = td;
		error = proc_rwmem(p, &uio);
		if (uio.uio_resid != 0) {
			/*
			 * XXX proc_rwmem() doesn't currently return ENOSPC,
			 * so I think write() can bogusly return 0.
			 * XXX what happens for short writes?  We don't want
			 * to write partial data.
			 * XXX proc_rwmem() returns EPERM for other invalid
			 * addresses.  Convert this to EINVAL.  Does this
			 * clobber returns of EPERM for other reasons?
			 */
			if (error == 0 || error == ENOSPC || error == EPERM)
				error = EINVAL;	/* EOF */
		}
		return (error);

	case PT_IO:
		error = copyin(uap->addr, &r.piod, sizeof r.piod);
		if (error)
			return (error);
		iov.iov_base = r.piod.piod_addr;
		iov.iov_len = r.piod.piod_len;
		uio.uio_iov = &iov;
		uio.uio_iovcnt = 1;
		uio.uio_offset = (off_t)(uintptr_t)r.piod.piod_offs;
		uio.uio_resid = r.piod.piod_len;
		uio.uio_segflg = UIO_USERSPACE;
		uio.uio_td = td;
		switch (r.piod.piod_op) {
		case PIOD_READ_D:
		case PIOD_READ_I:
			uio.uio_rw = UIO_READ;
			break;
		case PIOD_WRITE_D:
		case PIOD_WRITE_I:
			uio.uio_rw = UIO_WRITE;
			break;
		default:
			return (EINVAL);
		}
		error = proc_rwmem(p, &uio);
		r.piod.piod_len -= uio.uio_resid;
		(void)copyout(&r.piod, uap->addr, sizeof r.piod);
		return (error);

	case PT_KILL:
		uap->data = SIGKILL;
		goto sendsig;	/* in PT_CONTINUE above */

	case PT_SETREGS:
		_PHOLD(p);
		error = proc_write_regs(td2, &r.reg);
		_PRELE(p);
		PROC_UNLOCK(p);
		return (error);

	case PT_GETREGS:
		_PHOLD(p);
		error = proc_read_regs(td2, &r.reg);
		_PRELE(p);
		PROC_UNLOCK(p);
		if (error == 0)
			error = copyout(&r.reg, uap->addr, sizeof r.reg);
		return (error);

	case PT_SETFPREGS:
		_PHOLD(p);
		error = proc_write_fpregs(td2, &r.fpreg);
		_PRELE(p);
		PROC_UNLOCK(p);
		return (error);

	case PT_GETFPREGS:
		_PHOLD(p);
		error = proc_read_fpregs(td2, &r.fpreg);
		_PRELE(p);
		PROC_UNLOCK(p);
		if (error == 0)
			error = copyout(&r.fpreg, uap->addr, sizeof r.fpreg);
		return (error);

	case PT_SETDBREGS:
		_PHOLD(p);
		error = proc_write_dbregs(td2, &r.dbreg);
		_PRELE(p);
		PROC_UNLOCK(p);
		return (error);

	case PT_GETDBREGS:
		_PHOLD(p);
		error = proc_read_dbregs(td2, &r.dbreg);
		_PRELE(p);
		PROC_UNLOCK(p);
		if (error == 0)
			error = copyout(&r.dbreg, uap->addr, sizeof r.dbreg);
		return (error);

	default:
		KASSERT(0, ("unreachable code\n"));
		break;
	}

	KASSERT(0, ("unreachable code\n"));
	return (0);

fail:
	PROC_UNLOCK(p);
	if (proctree_locked)
		sx_xunlock(&proctree_lock);
	return (error);
}

/*
 * Stop a process because of a debugging event;
 * stay stopped until p->p_step is cleared
 * (cleared by PIOCCONT in procfs).
 */
void
stopevent(struct proc *p, unsigned int event, unsigned int val)
{

	PROC_LOCK_ASSERT(p, MA_OWNED | MA_NOTRECURSED);
	p->p_step = 1;

	do {
		p->p_xstat = val;
		p->p_stype = event;	/* Which event caused the stop? */
		wakeup(&p->p_stype);	/* Wake up any PIOCWAIT'ing procs */
		msleep(&p->p_step, &p->p_mtx, PWAIT, "stopevent", 0);
	} while (p->p_step);
}
OpenPOWER on IntegriCloud