initramfs.c 13.0 KB
Newer Older
1 2 3 4 5 6 7 8 9 10
/*
 * Many of the syscalls used in this file expect some of the arguments
 * to be __user pointers not __kernel pointers.  To limit the sparse
 * noise, turn off sparse checking for this file.
 */
#ifdef __CHECKER__
#undef __CHECKER__
#warning "Sparse checking disabled for this file"
#endif

L
Linus Torvalds 已提交
11 12 13 14 15 16 17
#include <linux/init.h>
#include <linux/fs.h>
#include <linux/slab.h>
#include <linux/types.h>
#include <linux/fcntl.h>
#include <linux/delay.h>
#include <linux/string.h>
18
#include <linux/dirent.h>
L
Linus Torvalds 已提交
19
#include <linux/syscalls.h>
20
#include <linux/utime.h>
L
Linus Torvalds 已提交
21 22 23 24 25 26 27 28 29 30

static __initdata char *message;
static void __init error(char *x)
{
	if (!message)
		message = x;
}

/* link hash */

31 32
#define N_ALIGN(len) ((((len) + 1) & ~3) + 2)

L
Linus Torvalds 已提交
33 34
static __initdata struct hash {
	int ino, minor, major;
A
Al Viro 已提交
35
	umode_t mode;
L
Linus Torvalds 已提交
36
	struct hash *next;
37
	char name[N_ALIGN(PATH_MAX)];
L
Linus Torvalds 已提交
38 39 40 41 42 43 44 45 46
} *head[32];

static inline int hash(int major, int minor, int ino)
{
	unsigned long tmp = ino + minor + (major << 3);
	tmp += tmp >> 5;
	return tmp & 31;
}

H
H. Peter Anvin 已提交
47
static char __init *find_link(int major, int minor, int ino,
A
Al Viro 已提交
48
			      umode_t mode, char *name)
L
Linus Torvalds 已提交
49 50 51 52 53 54 55 56 57
{
	struct hash **p, *q;
	for (p = head + hash(major, minor, ino); *p; p = &(*p)->next) {
		if ((*p)->ino != ino)
			continue;
		if ((*p)->minor != minor)
			continue;
		if ((*p)->major != major)
			continue;
H
H. Peter Anvin 已提交
58 59
		if (((*p)->mode ^ mode) & S_IFMT)
			continue;
L
Linus Torvalds 已提交
60 61
		return (*p)->name;
	}
62
	q = kmalloc(sizeof(struct hash), GFP_KERNEL);
L
Linus Torvalds 已提交
63 64 65
	if (!q)
		panic("can't allocate link hash entry");
	q->major = major;
H
H. Peter Anvin 已提交
66 67 68
	q->minor = minor;
	q->ino = ino;
	q->mode = mode;
69
	strcpy(q->name, name);
L
Linus Torvalds 已提交
70 71 72 73 74 75 76 77 78 79 80 81
	q->next = NULL;
	*p = q;
	return NULL;
}

static void __init free_hash(void)
{
	struct hash **p, *q;
	for (p = head; p < head + 32; p++) {
		while (*p) {
			q = *p;
			*p = q->next;
82
			kfree(q);
L
Linus Torvalds 已提交
83 84 85 86
		}
	}
}

87
static long __init do_utime(char *filename, time_t mtime)
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
{
	struct timespec t[2];

	t[0].tv_sec = mtime;
	t[0].tv_nsec = 0;
	t[1].tv_sec = mtime;
	t[1].tv_nsec = 0;

	return do_utimes(AT_FDCWD, filename, t, AT_SYMLINK_NOFOLLOW);
}

static __initdata LIST_HEAD(dir_list);
struct dir_entry {
	struct list_head list;
	char *name;
	time_t mtime;
};

static void __init dir_add(const char *name, time_t mtime)
{
	struct dir_entry *de = kmalloc(sizeof(struct dir_entry), GFP_KERNEL);
	if (!de)
		panic("can't allocate dir_entry buffer");
	INIT_LIST_HEAD(&de->list);
	de->name = kstrdup(name, GFP_KERNEL);
	de->mtime = mtime;
	list_add(&de->list, &dir_list);
}

static void __init dir_utime(void)
{
	struct dir_entry *de, *tmp;
	list_for_each_entry_safe(de, tmp, &dir_list, list) {
		list_del(&de->list);
		do_utime(de->name, de->mtime);
		kfree(de->name);
		kfree(de);
	}
}

static __initdata time_t mtime;

L
Linus Torvalds 已提交
130 131 132
/* cpio header parsing */

static __initdata unsigned long ino, major, minor, nlink;
A
Al Viro 已提交
133
static __initdata umode_t mode;
L
Linus Torvalds 已提交
134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154
static __initdata unsigned long body_len, name_len;
static __initdata uid_t uid;
static __initdata gid_t gid;
static __initdata unsigned rdev;

static void __init parse_header(char *s)
{
	unsigned long parsed[12];
	char buf[9];
	int i;

	buf[8] = '\0';
	for (i = 0, s += 6; i < 12; i++, s += 8) {
		memcpy(buf, s, 8);
		parsed[i] = simple_strtoul(buf, NULL, 16);
	}
	ino = parsed[0];
	mode = parsed[1];
	uid = parsed[2];
	gid = parsed[3];
	nlink = parsed[4];
155
	mtime = parsed[5];
L
Linus Torvalds 已提交
156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179
	body_len = parsed[6];
	major = parsed[7];
	minor = parsed[8];
	rdev = new_encode_dev(MKDEV(parsed[9], parsed[10]));
	name_len = parsed[11];
}

/* FSM */

static __initdata enum state {
	Start,
	Collect,
	GotHeader,
	SkipIt,
	GotName,
	CopyFile,
	GotSymlink,
	Reset
} state, next_state;

static __initdata char *victim;
static __initdata unsigned count;
static __initdata loff_t this_header, next_header;

A
Al Viro 已提交
180
static inline void __init eat(unsigned n)
L
Linus Torvalds 已提交
181 182 183 184 185 186
{
	victim += n;
	this_header += n;
	count -= n;
}

187
static __initdata char *vcollected;
L
Linus Torvalds 已提交
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
static __initdata char *collected;
static __initdata int remains;
static __initdata char *collect;

static void __init read_into(char *buf, unsigned size, enum state next)
{
	if (count >= size) {
		collected = victim;
		eat(size);
		state = next;
	} else {
		collect = collected = buf;
		remains = size;
		next_state = next;
		state = Collect;
	}
}

static __initdata char *header_buf, *symlink_buf, *name_buf;

static int __init do_start(void)
{
	read_into(header_buf, 110, GotHeader);
	return 0;
}

static int __init do_collect(void)
{
	unsigned n = remains;
	if (count < n)
		n = count;
	memcpy(collect, victim, n);
	eat(n);
	collect += n;
	if ((remains -= n) != 0)
		return 1;
	state = next_state;
	return 0;
}

static int __init do_header(void)
{
230 231 232 233
	if (memcmp(collected, "070707", 6)==0) {
		error("incorrect cpio method used: use -H newc option");
		return 1;
	}
L
Linus Torvalds 已提交
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
	if (memcmp(collected, "070701", 6)) {
		error("no cpio magic");
		return 1;
	}
	parse_header(collected);
	next_header = this_header + N_ALIGN(name_len) + body_len;
	next_header = (next_header + 3) & ~3;
	state = SkipIt;
	if (name_len <= 0 || name_len > PATH_MAX)
		return 0;
	if (S_ISLNK(mode)) {
		if (body_len > PATH_MAX)
			return 0;
		collect = collected = symlink_buf;
		remains = N_ALIGN(name_len) + body_len;
		next_state = GotSymlink;
		state = Collect;
		return 0;
	}
	if (S_ISREG(mode) || !body_len)
		read_into(name_buf, N_ALIGN(name_len), GotName);
	return 0;
}

static int __init do_skip(void)
{
	if (this_header + count < next_header) {
		eat(count);
		return 1;
	} else {
		eat(next_header - this_header);
		state = next_state;
		return 0;
	}
}

static int __init do_reset(void)
{
	while(count && *victim == '\0')
		eat(1);
	if (count && (this_header & 3))
		error("broken padding");
	return 1;
}

static int __init maybe_link(void)
{
	if (nlink >= 2) {
H
H. Peter Anvin 已提交
282
		char *old = find_link(major, minor, ino, mode, collected);
L
Linus Torvalds 已提交
283 284 285 286 287 288
		if (old)
			return (sys_link(old, collected) < 0) ? -1 : 1;
	}
	return 0;
}

A
Al Viro 已提交
289
static void __init clean_path(char *path, umode_t mode)
H
H. Peter Anvin 已提交
290 291 292 293 294 295 296 297 298 299 300
{
	struct stat st;

	if (!sys_newlstat(path, &st) && (st.st_mode^mode) & S_IFMT) {
		if (S_ISDIR(st.st_mode))
			sys_rmdir(path);
		else
			sys_unlink(path);
	}
}

L
Linus Torvalds 已提交
301 302 303 304 305 306 307 308 309 310
static __initdata int wfd;

static int __init do_name(void)
{
	state = SkipIt;
	next_state = Reset;
	if (strcmp(collected, "TRAILER!!!") == 0) {
		free_hash();
		return 0;
	}
H
H. Peter Anvin 已提交
311
	clean_path(collected, mode);
L
Linus Torvalds 已提交
312
	if (S_ISREG(mode)) {
H
H. Peter Anvin 已提交
313 314 315 316 317 318 319
		int ml = maybe_link();
		if (ml >= 0) {
			int openflags = O_WRONLY|O_CREAT;
			if (ml != 1)
				openflags |= O_TRUNC;
			wfd = sys_open(collected, openflags, mode);

L
Linus Torvalds 已提交
320 321 322
			if (wfd >= 0) {
				sys_fchown(wfd, uid, gid);
				sys_fchmod(wfd, mode);
323 324
				if (body_len)
					sys_ftruncate(wfd, body_len);
325
				vcollected = kstrdup(collected, GFP_KERNEL);
L
Linus Torvalds 已提交
326 327 328 329 330 331 332
				state = CopyFile;
			}
		}
	} else if (S_ISDIR(mode)) {
		sys_mkdir(collected, mode);
		sys_chown(collected, uid, gid);
		sys_chmod(collected, mode);
333
		dir_add(collected, mtime);
L
Linus Torvalds 已提交
334 335 336 337 338 339
	} else if (S_ISBLK(mode) || S_ISCHR(mode) ||
		   S_ISFIFO(mode) || S_ISSOCK(mode)) {
		if (maybe_link() == 0) {
			sys_mknod(collected, mode, rdev);
			sys_chown(collected, uid, gid);
			sys_chmod(collected, mode);
340
			do_utime(collected, mtime);
L
Linus Torvalds 已提交
341 342 343 344 345 346 347 348 349 350
		}
	}
	return 0;
}

static int __init do_copy(void)
{
	if (count >= body_len) {
		sys_write(wfd, victim, body_len);
		sys_close(wfd);
351 352
		do_utime(vcollected, mtime);
		kfree(vcollected);
L
Linus Torvalds 已提交
353 354 355 356 357 358 359 360 361 362 363 364 365 366
		eat(body_len);
		state = SkipIt;
		return 0;
	} else {
		sys_write(wfd, victim, count);
		body_len -= count;
		eat(count);
		return 1;
	}
}

static int __init do_symlink(void)
{
	collected[N_ALIGN(name_len) + body_len] = '\0';
H
H. Peter Anvin 已提交
367
	clean_path(collected, 0);
L
Linus Torvalds 已提交
368 369
	sys_symlink(collected + N_ALIGN(name_len), collected);
	sys_lchown(collected, uid, gid);
370
	do_utime(collected, mtime);
L
Linus Torvalds 已提交
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
	state = SkipIt;
	next_state = Reset;
	return 0;
}

static __initdata int (*actions[])(void) = {
	[Start]		= do_start,
	[Collect]	= do_collect,
	[GotHeader]	= do_header,
	[SkipIt]	= do_skip,
	[GotName]	= do_name,
	[CopyFile]	= do_copy,
	[GotSymlink]	= do_symlink,
	[Reset]		= do_reset,
};

static int __init write_buffer(char *buf, unsigned len)
{
	count = len;
	victim = buf;

	while (!actions[state]())
		;
	return len - count;
}

397
static int __init flush_buffer(void *bufv, unsigned len)
L
Linus Torvalds 已提交
398
{
399
	char *buf = (char *) bufv;
L
Linus Torvalds 已提交
400
	int written;
401
	int origLen = len;
L
Linus Torvalds 已提交
402
	if (message)
403
		return -1;
L
Linus Torvalds 已提交
404 405 406 407 408 409 410 411 412 413 414 415 416
	while ((written = write_buffer(buf, len)) < len && !message) {
		char c = buf[written];
		if (c == '0') {
			buf += written;
			len -= written;
			state = Start;
		} else if (c == 0) {
			buf += written;
			len -= written;
			state = Reset;
		} else
			error("junk in compressed archive");
	}
417
	return origLen;
L
Linus Torvalds 已提交
418 419
}

420
static unsigned my_inptr;   /* index of next byte to be processed in inbuf */
L
Linus Torvalds 已提交
421

422
#include <linux/decompress/generic.h>
L
Linus Torvalds 已提交
423

424
static char * __init unpack_to_rootfs(char *buf, unsigned len)
L
Linus Torvalds 已提交
425
{
426
	int written, res;
427
	decompress_fn decompress;
428 429
	const char *compress_name;
	static __initdata char msg_buf[64];
430

431 432 433
	header_buf = kmalloc(110, GFP_KERNEL);
	symlink_buf = kmalloc(PATH_MAX + N_ALIGN(PATH_MAX) + 1, GFP_KERNEL);
	name_buf = kmalloc(N_ALIGN(PATH_MAX), GFP_KERNEL);
434 435

	if (!header_buf || !symlink_buf || !name_buf)
L
Linus Torvalds 已提交
436
		panic("can't allocate buffers");
437

L
Linus Torvalds 已提交
438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456
	state = Start;
	this_header = 0;
	message = NULL;
	while (!message && len) {
		loff_t saved_offset = this_header;
		if (*buf == '0' && !(this_header & 3)) {
			state = Start;
			written = write_buffer(buf, len);
			buf += written;
			len -= written;
			continue;
		}
		if (!*buf) {
			buf++;
			len--;
			this_header++;
			continue;
		}
		this_header = 0;
457
		decompress = decompress_method(buf, len, &compress_name);
458 459
		if (decompress) {
			res = decompress(buf, len, NULL, flush_buffer, NULL,
460
				   &my_inptr, error);
461 462 463
			if (res)
				error("decompressor failed");
		} else if (compress_name) {
464 465 466 467 468 469
			if (!message) {
				snprintf(msg_buf, sizeof msg_buf,
					 "compression method %s not configured",
					 compress_name);
				message = msg_buf;
			}
470 471
		} else
			error("junk in compressed archive");
L
Linus Torvalds 已提交
472
		if (state != Reset)
473 474 475 476
			error("junk in compressed archive");
		this_header = saved_offset + my_inptr;
		buf += my_inptr;
		len -= my_inptr;
L
Linus Torvalds 已提交
477
	}
478
	dir_utime();
479 480 481
	kfree(name_buf);
	kfree(symlink_buf);
	kfree(header_buf);
L
Linus Torvalds 已提交
482 483 484
	return message;
}

485 486 487 488 489 490 491 492 493 494 495
static int __initdata do_retain_initrd;

static int __init retain_initrd_param(char *str)
{
	if (*str)
		return 0;
	do_retain_initrd = 1;
	return 1;
}
__setup("retain_initrd", retain_initrd_param);

496 497
extern char __initramfs_start[];
extern unsigned long __initramfs_size;
L
Linus Torvalds 已提交
498
#include <linux/initrd.h>
499
#include <linux/kexec.h>
J
Jan Beulich 已提交
500 501 502

static void __init free_initrd(void)
{
503 504 505
#ifdef CONFIG_KEXEC
	unsigned long crashk_start = (unsigned long)__va(crashk_res.start);
	unsigned long crashk_end   = (unsigned long)__va(crashk_res.end);
506 507 508
#endif
	if (do_retain_initrd)
		goto skip;
509

510
#ifdef CONFIG_KEXEC
511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527
	/*
	 * If the initrd region is overlapped with crashkernel reserved region,
	 * free only memory that is not part of crashkernel region.
	 */
	if (initrd_start < crashk_end && initrd_end > crashk_start) {
		/*
		 * Initialize initrd memory region since the kexec boot does
		 * not do.
		 */
		memset((void *)initrd_start, 0, initrd_end - initrd_start);
		if (initrd_start < crashk_start)
			free_initrd_mem(initrd_start, crashk_start);
		if (initrd_end > crashk_end)
			free_initrd_mem(crashk_end, initrd_end);
	} else
#endif
		free_initrd_mem(initrd_start, initrd_end);
528
skip:
J
Jan Beulich 已提交
529 530 531 532
	initrd_start = 0;
	initrd_end = 0;
}

533
#ifdef CONFIG_BLK_DEV_RAM
534 535 536 537 538 539
#define BUF_SIZE 1024
static void __init clean_rootfs(void)
{
	int fd;
	void *buf;
	struct linux_dirent64 *dirp;
540
	int num;
541

542
	fd = sys_open("/", O_RDONLY, 0);
543 544 545 546 547 548 549 550 551 552 553
	WARN_ON(fd < 0);
	if (fd < 0)
		return;
	buf = kzalloc(BUF_SIZE, GFP_KERNEL);
	WARN_ON(!buf);
	if (!buf) {
		sys_close(fd);
		return;
	}

	dirp = buf;
554 555 556
	num = sys_getdents64(fd, dirp, BUF_SIZE);
	while (num > 0) {
		while (num > 0) {
557 558 559 560 561 562 563 564 565 566 567 568
			struct stat st;
			int ret;

			ret = sys_newlstat(dirp->d_name, &st);
			WARN_ON_ONCE(ret);
			if (!ret) {
				if (S_ISDIR(st.st_mode))
					sys_rmdir(dirp->d_name);
				else
					sys_unlink(dirp->d_name);
			}

569
			num -= dirp->d_reclen;
570 571 572 573
			dirp = (void *)dirp + dirp->d_reclen;
		}
		dirp = buf;
		memset(buf, 0, BUF_SIZE);
574
		num = sys_getdents64(fd, dirp, BUF_SIZE);
575 576 577 578 579
	}

	sys_close(fd);
	kfree(buf);
}
580
#endif
581

582
static int __init populate_rootfs(void)
L
Linus Torvalds 已提交
583
{
584
	char *err = unpack_to_rootfs(__initramfs_start, __initramfs_size);
L
Linus Torvalds 已提交
585
	if (err)
586
		panic(err);	/* Failed to decompress INTERNAL initramfs */
L
Linus Torvalds 已提交
587
	if (initrd_start) {
588
#ifdef CONFIG_BLK_DEV_RAM
L
Linus Torvalds 已提交
589
		int fd;
590
		printk(KERN_INFO "Trying to unpack rootfs image as initramfs...\n");
L
Linus Torvalds 已提交
591
		err = unpack_to_rootfs((char *)initrd_start,
592
			initrd_end - initrd_start);
L
Linus Torvalds 已提交
593
		if (!err) {
J
Jan Beulich 已提交
594
			free_initrd();
595
			goto done;
596 597
		} else {
			clean_rootfs();
598
			unpack_to_rootfs(__initramfs_start, __initramfs_size);
L
Linus Torvalds 已提交
599
		}
600 601
		printk(KERN_INFO "rootfs image is not initramfs (%s)"
				"; looks like an initrd\n", err);
602
		fd = sys_open("/initrd.image",
603
			      O_WRONLY|O_CREAT, 0700);
L
Linus Torvalds 已提交
604 605 606 607
		if (fd >= 0) {
			sys_write(fd, (char *)initrd_start,
					initrd_end - initrd_start);
			sys_close(fd);
J
Jan Beulich 已提交
608
			free_initrd();
L
Linus Torvalds 已提交
609
		}
610
	done:
611
#else
612
		printk(KERN_INFO "Unpacking initramfs...\n");
613
		err = unpack_to_rootfs((char *)initrd_start,
614
			initrd_end - initrd_start);
615 616
		if (err)
			printk(KERN_EMERG "Initramfs unpacking failed: %s\n", err);
617 618
		free_initrd();
#endif
619 620 621 622 623
		/*
		 * Try loading default modules from initramfs.  This gives
		 * us a chance to load before device_initcalls.
		 */
		load_default_modules();
L
Linus Torvalds 已提交
624
	}
625
	return 0;
L
Linus Torvalds 已提交
626
}
627
rootfs_initcall(populate_rootfs);