aboutsummaryrefslogtreecommitdiffstats
path: root/arch/um/os-Linux/skas/process.c
blob: 045ae0037456a3a723e5d58fdb9858af64349a61 (plain) (blame)
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
/*
 * Copyright (C) 2002- 2004 Jeff Dike (jdike@addtoit.com)
 * Licensed under the GPL
 */

#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <errno.h>
#include <signal.h>
#include <setjmp.h>
#include <sched.h>
#include "ptrace_user.h"
#include <sys/wait.h>
#include <sys/mman.h>
#include <sys/user.h>
#include <sys/time.h>
#include <asm/unistd.h>
#include <asm/types.h>
#include "user.h"
#include "sysdep/ptrace.h"
#include "user_util.h"
#include "kern_util.h"
#include "skas.h"
#include "stub-data.h"
#include "mm_id.h"
#include "sysdep/sigcontext.h"
#include "sysdep/stub.h"
#include "os.h"
#include "proc_mm.h"
#include "skas_ptrace.h"
#include "chan_user.h"
#include "registers.h"
#include "mem.h"
#include "uml-config.h"
#include "process.h"
#include "longjmp.h"

int is_skas_winch(int pid, int fd, void *data)
{
	if(pid != os_getpgrp())
		return(0);

	register_winch_irq(-1, fd, -1, data);
	return(1);
}

void wait_stub_done(int pid, int sig, char * fname)
{
	int n, status, err;

	do {
		if ( sig != -1 ) {
			err = ptrace(PTRACE_CONT, pid, 0, sig);
			if(err)
				panic("%s : continue failed, errno = %d\n",
				      fname, errno);
		}
		sig = 0;

		CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED));
	} while((n >= 0) && WIFSTOPPED(status) &&
	        ((WSTOPSIG(status) == SIGVTALRM) ||
		 /* running UML inside a detached screen can cause
		  * SIGWINCHes
		  */
		 (WSTOPSIG(status) == SIGWINCH)));

	if((n < 0) || !WIFSTOPPED(status) ||
	   (WSTOPSIG(status) != SIGUSR1 && WSTOPSIG(status) != SIGTRAP)){
		unsigned long regs[HOST_FRAME_SIZE];

		if(ptrace(PTRACE_GETREGS, pid, 0, regs) < 0)
			printk("Failed to get registers from stub, "
			       "errno = %d\n", errno);
		else {
			int i;

			printk("Stub registers -\n");
			for(i = 0; i < HOST_FRAME_SIZE; i++)
				printk("\t%d - %lx\n", i, regs[i]);
		}
		panic("%s : failed to wait for SIGUSR1/SIGTRAP, "
		      "pid = %d, n = %d, errno = %d, status = 0x%x\n",
		      fname, pid, n, errno, status);
	}
}

extern unsigned long current_stub_stack(void);

void get_skas_faultinfo(int pid, struct faultinfo * fi)
{
	int err;

	if(ptrace_faultinfo){
		err = ptrace(PTRACE_FAULTINFO, pid, 0, fi);
		if(err)
			panic("get_skas_faultinfo - PTRACE_FAULTINFO failed, "
			      "errno = %d\n", errno);

		/* Special handling for i386, which has different structs */
		if (sizeof(struct ptrace_faultinfo) < sizeof(struct faultinfo))
			memset((char *)fi + sizeof(struct ptrace_faultinfo), 0,
			       sizeof(struct faultinfo) -
			       sizeof(struct ptrace_faultinfo));
	}
	else {
		wait_stub_done(pid, SIGSEGV, "get_skas_faultinfo");

		/* faultinfo is prepared by the stub-segv-handler at start of
		 * the stub stack page. We just have to copy it.
		 */
		memcpy(fi, (void *)current_stub_stack(), sizeof(*fi));
	}
}

static void handle_segv(int pid, union uml_pt_regs * regs)
{
	get_skas_faultinfo(pid, &regs->skas.faultinfo);
	segv(regs->skas.faultinfo, 0, 1, NULL);
}

/*To use the same value of using_sysemu as the caller, ask it that value (in local_using_sysemu)*/
static void handle_trap(int pid, union uml_pt_regs *regs, int local_using_sysemu)
{
	int err, status;

	/* Mark this as a syscall */
	UPT_SYSCALL_NR(regs) = PT_SYSCALL_NR(regs->skas.regs);

	if (!local_using_sysemu)
	{
		err = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET,
			     __NR_getpid);
		if(err < 0)
			panic("handle_trap - nullifying syscall failed errno = %d\n",
			      errno);

		err = ptrace(PTRACE_SYSCALL, pid, 0, 0);
		if(err < 0)
			panic("handle_trap - continuing to end of syscall failed, "
			      "errno = %d\n", errno);

		CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED));
		if((err < 0) || !WIFSTOPPED(status) ||
		   (WSTOPSIG(status) != SIGTRAP + 0x80))
			panic("handle_trap - failed to wait at end of syscall, "
			      "errno = %d, status = %d\n", errno, status);
	}

	handle_syscall(regs);
}

extern int __syscall_stub_start;

static int userspace_tramp(void *stack)
{
	void *addr;

	ptrace(PTRACE_TRACEME, 0, 0, 0);

	init_new_thread_signals(1);
	enable_timer();

	if(!proc_mm){
		/* This has a pte, but it can't be mapped in with the usual
		 * tlb_flush mechanism because this is part of that mechanism
		 */
		int fd;
		__u64 offset;
		fd = phys_mapping(to_phys(&__syscall_stub_start), &offset);
		addr = mmap64((void *) UML_CONFIG_STUB_CODE, page_size(),
			      PROT_EXEC, MAP_FIXED | MAP_PRIVATE, fd, offset);
		if(addr == MAP_FAILED){
			printk("mapping mmap stub failed, errno = %d\n",
			       errno);
			exit(1);
		}

		if(stack != NULL){
			fd = phys_mapping(to_phys(stack), &offset);
			addr = mmap((void *) UML_CONFIG_STUB_DATA, page_size(),
				    PROT_READ | PROT_WRITE,
				    MAP_FIXED | MAP_SHARED, fd, offset);
			if(addr == MAP_FAILED){
				printk("mapping segfault stack failed, "
				       "errno = %d\n", errno);
				exit(1);
			}
		}
	}
	if(!ptrace_faultinfo && (stack != NULL)){
		unsigned long v = UML_CONFIG_STUB_CODE +
				  (unsigned long) stub_segv_handler -
				  (unsigned long) &__syscall_stub_start;

		set_sigstack((void *) UML_CONFIG_STUB_DATA, page_size());
		set_handler(SIGSEGV, (void *) v, SA_ONSTACK,
			    SIGIO, SIGWINCH, SIGALRM, SIGVTALRM,
			    SIGUSR1, -1);
	}

	os_stop_process(os_getpid());
	return(0);
}

/* Each element set once, and only accessed by a single processor anyway */
#undef NR_CPUS
#define NR_CPUS 1
int userspace_pid[NR_CPUS];

int start_userspace(unsigned long stub_stack)
{
	void *stack;
	unsigned long sp;
	int pid, status, n, flags;

	stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC,
		     MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
	if(stack == MAP_FAILED)
		panic("start_userspace : mmap failed, errno = %d", errno);
	sp = (unsigned long) stack + PAGE_SIZE - sizeof(void *);

	flags = CLONE_FILES | SIGCHLD;
	if(proc_mm) flags |= CLONE_VM;
	pid = clone(userspace_tramp, (void *) sp, flags, (void *) stub_stack);
	if(pid < 0)
		panic("start_userspace : clone failed, errno = %d", errno);

	do {
		CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED));
		if(n < 0)
			panic("start_userspace : wait failed, errno = %d",
			      errno);
	} while(WIFSTOPPED(status) && (WSTOPSIG(status) == SIGVTALRM));

	if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP))
		panic("start_userspace : expected SIGSTOP, got status = %d",
		      status);

	if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL, (void *)PTRACE_O_TRACESYSGOOD) < 0)
		panic("start_userspace : PTRACE_OLDSETOPTIONS failed, errno=%d\n",
		      errno);

	if(munmap(stack, PAGE_SIZE) < 0)
		panic("start_userspace : munmap failed, errno = %d\n", errno);

	return(pid);
}

void userspace(union uml_pt_regs *regs)
{
	int err, status, op, pid = userspace_pid[0];
	int local_using_sysemu; /*To prevent races if using_sysemu changes under us.*/

	while(1){
		restore_registers(pid, regs);

		/* Now we set local_using_sysemu to be used for one loop */
		local_using_sysemu = get_using_sysemu();

		op = SELECT_PTRACE_OPERATION(local_using_sysemu, singlestepping(NULL));

		err = ptrace(op, pid, 0, 0);
		if(err)
			panic("userspace - could not resume userspace process, "
			      "pid=%d, ptrace operation = %d, errno = %d\n",
			      pid, op, errno);

		CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED));
		if(err < 0)
			panic("userspace - waitpid failed, errno = %d\n",
			      errno);

		regs->skas.is_user = 1;
		save_registers(pid, regs);
		UPT_SYSCALL_NR(regs) = -1; /* Assume: It's not a syscall */

		if(WIFSTOPPED(status)){
		  	switch(WSTOPSIG(status)){
			case SIGSEGV:
				if(PTRACE_FULL_FAULTINFO || !ptrace_faultinfo)
					user_signal(SIGSEGV, regs, pid);
				else handle_segv(pid, regs);
				break;
			case SIGTRAP + 0x80:
			        handle_trap(pid, regs, local_using_sysemu);
				break;
			case SIGTRAP:
				relay_signal(SIGTRAP, regs);
				break;
			case SIGIO:
			case SIGVTALRM:
			case SIGILL:
			case SIGBUS:
			case SIGFPE:
			case SIGWINCH:
				user_signal(WSTOPSIG(status), regs, pid);
				break;
			default:
			        printk("userspace - child stopped with signal "
				       "%d\n", WSTOPSIG(status));
			}
			pid = userspace_pid[0];
			interrupt_end();

			/* Avoid -ERESTARTSYS handling in host */
			if(PT_SYSCALL_NR_OFFSET != PT_SYSCALL_RET_OFFSET)
				PT_SYSCALL_NR(regs->skas.regs) = -1;
		}
	}
}

int copy_context_skas0(unsigned long new_stack, int pid)
{
	int err;
	unsigned long regs[HOST_FRAME_SIZE];
	unsigned long fp_regs[HOST_FP_SIZE];
	unsigned long current_stack = current_stub_stack();
	struct stub_data *data = (struct stub_data *) current_stack;
	struct stub_data *child_data = (struct stub_data *) new_stack;
	__u64 new_offset;
	int new_fd = phys_mapping(to_phys((void *)new_stack), &new_offset);

	/* prepare offset and fd of child's stack as argument for parent's
	 * and child's mmap2 calls
	 */
	*data = ((struct stub_data) { .offset	= MMAP_OFFSET(new_offset),
				      .fd	= new_fd,
				      .timer    = ((struct itimerval)
					            { { 0, 1000000 / hz() },
						      { 0, 1000000 / hz() }})});
	get_safe_registers(regs, fp_regs);

	/* Set parent's instruction pointer to start of clone-stub */
	regs[REGS_IP_INDEX] = UML_CONFIG_STUB_CODE +
				(unsigned long) stub_clone_handler -
				(unsigned long) &__syscall_stub_start;
	regs[REGS_SP_INDEX] = UML_CONFIG_STUB_DATA + PAGE_SIZE -
		sizeof(void *);
#ifdef __SIGNAL_FRAMESIZE
	regs[REGS_SP_INDEX] -= __SIGNAL_FRAMESIZE;
#endif
	err = ptrace_setregs(pid, regs);
	if(err < 0)
		panic("copy_context_skas0 : PTRACE_SETREGS failed, "
		      "pid = %d, errno = %d\n", pid, errno);

	err = ptrace_setfpregs(pid, fp_regs);
	if(err < 0)
		panic("copy_context_skas0 : PTRACE_SETFPREGS failed, "
		      "pid = %d, errno = %d\n", pid, errno);

	/* set a well known return code for detection of child write failure */
	child_data->err = 12345678;

	/* Wait, until parent has finished its work: read child's pid from
	 * parent's stack, and check, if bad result.
	 */
	wait_stub_done(pid, 0, "copy_context_skas0");

	pid = data->err;
	if(pid < 0)
		panic("copy_context_skas0 - stub-parent reports error %d\n",
		      pid);

	/* Wait, until child has finished too: read child's result from
	 * child's stack and check it.
	 */
	wait_stub_done(pid, -1, "copy_context_skas0");
	if (child_data->err != UML_CONFIG_STUB_DATA)
		panic("copy_context_skas0 - stub-child reports error %ld\n",
		      child_data->err);

	if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL,
		   (void *)PTRACE_O_TRACESYSGOOD) < 0)
		panic("copy_context_skas0 : PTRACE_OLDSETOPTIONS failed, "
		      "errno = %d\n", errno);

	return pid;
}

/*
 * This is used only, if stub pages are needed, while proc_mm is
 * availabl. Opening /proc/mm creates a new mm_context, which lacks
 * the stub-pages. Thus, we map them using /proc/mm-fd
 */
void map_stub_pages(int fd, unsigned long code,
		    unsigned long data, unsigned long stack)
{
	struct proc_mm_op mmop;
	int n;
	__u64 code_offset;
	int code_fd = phys_mapping(to_phys((void *) &__syscall_stub_start),
				   &code_offset);

	mmop = ((struct proc_mm_op) { .op        = MM_MMAP,
				      .u         =
				      { .mmap    =
					{ .addr    = code,
					  .len     = PAGE_SIZE,
					  .prot    = PROT_EXEC,
					  .flags   = MAP_FIXED | MAP_PRIVATE,
					  .fd      = code_fd,
					  .offset  = code_offset
	} } });
	n = os_write_file(fd, &mmop, sizeof(mmop));
	if(n != sizeof(mmop))
		panic("map_stub_pages : /proc/mm map for code failed, "
		      "err = %d\n", -n);

	if ( stack ) {
		__u64 map_offset;
		int map_fd = phys_mapping(to_phys((void *)stack), &map_offset);
		mmop = ((struct proc_mm_op)
				{ .op        = MM_MMAP,
				  .u         =
				  { .mmap    =
				    { .addr    = data,
				      .len     = PAGE_SIZE,
				      .prot    = PROT_READ | PROT_WRITE,
				      .flags   = MAP_FIXED | MAP_SHARED,
				      .fd      = map_fd,
				      .offset  = map_offset
		} } });
		n = os_write_file(fd, &mmop, sizeof(mmop));
		if(n != sizeof(mmop))
			panic("map_stub_pages : /proc/mm map for data failed, "
			      "err = %d\n", -n);
	}
}

void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr,
		void (*handler)(int))
{
	unsigned long flags;
	sigjmp_buf switch_buf, fork_buf;
	int enable;

	*switch_buf_ptr = &switch_buf;
	*fork_buf_ptr = &fork_buf;

	/* Somewhat subtle - siglongjmp restores the signal mask before doing
	 * the longjmp.  This means that when jumping from one stack to another
	 * when the target stack has interrupts enabled, an interrupt may occur
	 * on the source stack.  This is bad when starting up a process because
	 * it's not supposed to get timer ticks until it has been scheduled.
	 * So, we disable interrupts around the sigsetjmp to ensure that
	 * they can't happen until we get back here where they are safe.
	 */
	flags = get_signals();
	block_signals();
	if(UML_SIGSETJMP(&fork_buf, enable) == 0)
		new_thread_proc(stack, handler);

	remove_sigstack();

	set_signals(flags);
}

#define INIT_JMP_NEW_THREAD 0
#define INIT_JMP_REMOVE_SIGSTACK 1
#define INIT_JMP_CALLBACK 2
#define INIT_JMP_HALT 3
#define INIT_JMP_REBOOT 4

void thread_wait(void *sw, void *fb)
{
	sigjmp_buf buf, **switch_buf = sw, *fork_buf;
	int enable;

	*switch_buf = &buf;
	fork_buf = fb;
	if(UML_SIGSETJMP(&buf, enable) == 0)
		siglongjmp(*fork_buf, INIT_JMP_REMOVE_SIGSTACK);
}

void switch_threads(void *me, void *next)
{
	sigjmp_buf my_buf, **me_ptr = me, *next_buf = next;
	int enable;

	*me_ptr = &my_buf;
	if(UML_SIGSETJMP(&my_buf, enable) == 0)
		UML_SIGLONGJMP(next_buf, 1);
}

static sigjmp_buf initial_jmpbuf;

/* XXX Make these percpu */
static void (*cb_proc)(void *arg);
static void *cb_arg;
static sigjmp_buf *cb_back;

int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr)
{
	sigjmp_buf **switch_buf = switch_buf_ptr;
	int n, enable;

	set_handler(SIGWINCH, (__sighandler_t) sig_handler,
		    SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGALRM,
		    SIGVTALRM, -1);

	*fork_buf_ptr = &initial_jmpbuf;
	n = UML_SIGSETJMP(&initial_jmpbuf, enable);
	switch(n){
	case INIT_JMP_NEW_THREAD:
		new_thread_proc((void *) stack, new_thread_handler);
		break;
	case INIT_JMP_REMOVE_SIGSTACK:
		remove_sigstack();
		break;
	case INIT_JMP_CALLBACK:
		(*cb_proc)(cb_arg);
		UML_SIGLONGJMP(cb_back, 1);
		break;
	case INIT_JMP_HALT:
		kmalloc_ok = 0;
		return(0);
	case INIT_JMP_REBOOT:
		kmalloc_ok = 0;
		return(1);
	default:
		panic("Bad sigsetjmp return in start_idle_thread - %d\n", n);
	}
	UML_SIGLONGJMP(*switch_buf, 1);
}

void initial_thread_cb_skas(void (*proc)(void *), void *arg)
{
	sigjmp_buf here;
	int enable;

	cb_proc = proc;
	cb_arg = arg;
	cb_back = &here;

	block_signals();
	if(UML_SIGSETJMP(&here, enable) == 0)
		UML_SIGLONGJMP(&initial_jmpbuf, INIT_JMP_CALLBACK);
	unblock_signals();

	cb_proc = NULL;
	cb_arg = NULL;
	cb_back = NULL;
}

void halt_skas(void)
{
	block_signals();
	UML_SIGLONGJMP(&initial_jmpbuf, INIT_JMP_HALT);
}

void reboot_skas(void)
{
	block_signals();
	UML_SIGLONGJMP(&initial_jmpbuf, INIT_JMP_REBOOT);
}

void switch_mm_skas(struct mm_id *mm_idp)
{
	int err;

#warning need cpu pid in switch_mm_skas
	if(proc_mm){
		err = ptrace(PTRACE_SWITCH_MM, userspace_pid[0], 0,
			     mm_idp->u.mm_fd);
		if(err)
			panic("switch_mm_skas - PTRACE_SWITCH_MM failed, "
			      "errno = %d\n", errno);
	}
	else userspace_pid[0] = mm_idp->u.pid;
}