read-cache.c 23 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1 2 3 4 5
/*
 * GIT - The information manager from hell
 *
 * Copyright (C) Linus Torvalds, 2005
 */
6
#include "cache.h"
7 8 9 10 11 12 13 14 15 16 17 18 19
#include "cache-tree.h"

/* Index extensions.
 *
 * The first letter should be 'A'..'Z' for extensions that are not
 * necessary for a correct operation (i.e. optimization data).
 * When new extensions are added that _needs_ to be understood in
 * order to correctly interpret the index file, pick character that
 * is outside the range, to cause the reader to abort.
 */

#define CACHE_EXT(s) ( (s[0]<<24)|(s[1]<<16)|(s[2]<<8)|(s[3]) )
#define CACHE_EXT_TREE 0x54524545	/* "TREE" */
20 21

struct cache_entry **active_cache = NULL;
Junio C Hamano's avatar
Junio C Hamano committed
22
static time_t index_file_timestamp;
23
unsigned int active_nr = 0, active_alloc = 0, active_cache_changed = 0;
24

25 26
struct cache_tree *active_cache_tree = NULL;

27 28 29 30 31 32 33 34 35
/*
 * This only updates the "non-critical" parts of the directory
 * cache, ie the parts that aren't tracked by GIT, and only used
 * to validate the cache.
 */
void fill_stat_cache_info(struct cache_entry *ce, struct stat *st)
{
	ce->ce_ctime.sec = htonl(st->st_ctime);
	ce->ce_mtime.sec = htonl(st->st_mtime);
36
#ifdef USE_NSEC
37 38 39 40 41 42 43 44
	ce->ce_ctime.nsec = htonl(st->st_ctim.tv_nsec);
	ce->ce_mtime.nsec = htonl(st->st_mtim.tv_nsec);
#endif
	ce->ce_dev = htonl(st->st_dev);
	ce->ce_ino = htonl(st->st_ino);
	ce->ce_uid = htonl(st->st_uid);
	ce->ce_gid = htonl(st->st_gid);
	ce->ce_size = htonl(st->st_size);
Junio C Hamano's avatar
Junio C Hamano committed
45 46 47

	if (assume_unchanged)
		ce->ce_flags |= htons(CE_VALID);
48 49
}

Junio C Hamano's avatar
Junio C Hamano committed
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
static int ce_compare_data(struct cache_entry *ce, struct stat *st)
{
	int match = -1;
	int fd = open(ce->name, O_RDONLY);

	if (fd >= 0) {
		unsigned char sha1[20];
		if (!index_fd(sha1, fd, st, 0, NULL))
			match = memcmp(sha1, ce->sha1, 20);
		close(fd);
	}
	return match;
}

static int ce_compare_link(struct cache_entry *ce, unsigned long expected_size)
{
	int match = -1;
	char *target;
	void *buffer;
	unsigned long size;
	char type[10];
	int len;

	target = xmalloc(expected_size);
	len = readlink(ce->name, target, expected_size);
	if (len != expected_size) {
		free(target);
		return -1;
	}
	buffer = read_sha1_file(ce->sha1, type, &size);
	if (!buffer) {
		free(target);
		return -1;
	}
	if (size == expected_size)
		match = memcmp(buffer, target, size);
	free(buffer);
	free(target);
	return match;
}

static int ce_modified_check_fs(struct cache_entry *ce, struct stat *st)
{
	switch (st->st_mode & S_IFMT) {
	case S_IFREG:
		if (ce_compare_data(ce, st))
			return DATA_CHANGED;
		break;
	case S_IFLNK:
		if (ce_compare_link(ce, st->st_size))
			return DATA_CHANGED;
		break;
	default:
		return TYPE_CHANGED;
	}
	return 0;
}

Junio C Hamano's avatar
Junio C Hamano committed
108
static int ce_match_stat_basic(struct cache_entry *ce, struct stat *st)
109 110 111
{
	unsigned int changed = 0;

112 113 114
	switch (ntohl(ce->ce_mode) & S_IFMT) {
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
Junio C Hamano's avatar
Junio C Hamano committed
115 116 117 118 119
		/* We consider only the owner x bit to be relevant for
		 * "mode changes"
		 */
		if (trust_executable_bit &&
		    (0100 & (ntohl(ce->ce_mode) ^ st->st_mode)))
120
			changed |= MODE_CHANGED;
121 122 123 124 125 126 127
		break;
	case S_IFLNK:
		changed |= !S_ISLNK(st->st_mode) ? TYPE_CHANGED : 0;
		break;
	default:
		die("internal error: ce_mode is %o", ntohl(ce->ce_mode));
	}
128
	if (ce->ce_mtime.sec != htonl(st->st_mtime))
129
		changed |= MTIME_CHANGED;
130 131 132
	if (ce->ce_ctime.sec != htonl(st->st_ctime))
		changed |= CTIME_CHANGED;

133
#ifdef USE_NSEC
134 135 136 137 138
	/*
	 * nsec seems unreliable - not all filesystems support it, so
	 * as long as it is in the inode cache you get right nsec
	 * but after it gets flushed, you get zero nsec.
	 */
139
	if (ce->ce_mtime.nsec != htonl(st->st_mtim.tv_nsec))
140
		changed |= MTIME_CHANGED;
141
	if (ce->ce_ctime.nsec != htonl(st->st_ctim.tv_nsec))
142
		changed |= CTIME_CHANGED;
143 144 145 146
#endif	

	if (ce->ce_uid != htonl(st->st_uid) ||
	    ce->ce_gid != htonl(st->st_gid))
147
		changed |= OWNER_CHANGED;
148
	if (ce->ce_ino != htonl(st->st_ino))
149
		changed |= INODE_CHANGED;
150 151 152 153 154 155 156 157 158 159 160

#ifdef USE_STDEV
	/*
	 * st_dev breaks on network filesystems where different
	 * clients will have different views of what "device"
	 * the filesystem is on
	 */
	if (ce->ce_dev != htonl(st->st_dev))
		changed |= INODE_CHANGED;
#endif

161
	if (ce->ce_size != htonl(st->st_size))
162
		changed |= DATA_CHANGED;
163

Junio C Hamano's avatar
Junio C Hamano committed
164 165 166
	return changed;
}

Junio C Hamano's avatar
Junio C Hamano committed
167
int ce_match_stat(struct cache_entry *ce, struct stat *st, int ignore_valid)
Junio C Hamano's avatar
Junio C Hamano committed
168
{
Junio C Hamano's avatar
Junio C Hamano committed
169 170 171 172 173 174 175 176 177 178
	unsigned int changed;

	/*
	 * If it's marked as always valid in the index, it's
	 * valid whatever the checked-out copy says.
	 */
	if (!ignore_valid && (ce->ce_flags & htons(CE_VALID)))
		return 0;

	changed = ce_match_stat_basic(ce, st);
Junio C Hamano's avatar
Junio C Hamano committed
179

Junio C Hamano's avatar
Junio C Hamano committed
180 181 182 183 184 185 186 187 188 189 190 191 192 193
	/*
	 * Within 1 second of this sequence:
	 * 	echo xyzzy >file && git-update-index --add file
	 * running this command:
	 * 	echo frotz >file
	 * would give a falsely clean cache entry.  The mtime and
	 * length match the cache, and other stat fields do not change.
	 *
	 * We could detect this at update-index time (the cache entry
	 * being registered/updated records the same time as "now")
	 * and delay the return from git-update-index, but that would
	 * effectively mean we can make at most one commit per second,
	 * which is not acceptable.  Instead, we check cache entries
	 * whose mtime are the same as the index file timestamp more
Junio C Hamano's avatar
Junio C Hamano committed
194
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
195 196 197 198 199
	 */
	if (!changed &&
	    index_file_timestamp &&
	    index_file_timestamp <= ntohl(ce->ce_mtime.sec))
		changed |= ce_modified_check_fs(ce, st);
200

Junio C Hamano's avatar
Junio C Hamano committed
201
	return changed;
202 203
}

Junio C Hamano's avatar
Junio C Hamano committed
204
int ce_modified(struct cache_entry *ce, struct stat *st, int really)
205
{
Junio C Hamano's avatar
Junio C Hamano committed
206
	int changed, changed_fs;
Junio C Hamano's avatar
Junio C Hamano committed
207
	changed = ce_match_stat(ce, st, really);
208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
	if (!changed)
		return 0;
	/*
	 * If the mode or type has changed, there's no point in trying
	 * to refresh the entry - it's not going to match
	 */
	if (changed & (MODE_CHANGED | TYPE_CHANGED))
		return changed;

	/* Immediately after read-tree or update-index --cacheinfo,
	 * the length field is zero.  For other cases the ce_size
	 * should match the SHA1 recorded in the index entry.
	 */
	if ((changed & DATA_CHANGED) && ce->ce_size != htonl(0))
		return changed;

Junio C Hamano's avatar
Junio C Hamano committed
224 225 226
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
227 228 229
	return 0;
}

230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
int base_name_compare(const char *name1, int len1, int mode1,
		      const char *name2, int len2, int mode2)
{
	unsigned char c1, c2;
	int len = len1 < len2 ? len1 : len2;
	int cmp;

	cmp = memcmp(name1, name2, len);
	if (cmp)
		return cmp;
	c1 = name1[len];
	c2 = name2[len];
	if (!c1 && S_ISDIR(mode1))
		c1 = '/';
	if (!c2 && S_ISDIR(mode2))
		c2 = '/';
	return (c1 < c2) ? -1 : (c1 > c2) ? 1 : 0;
}

249
int cache_name_compare(const char *name1, int flags1, const char *name2, int flags2)
250
{
251 252
	int len1 = flags1 & CE_NAMEMASK;
	int len2 = flags2 & CE_NAMEMASK;
253 254 255 256 257 258 259 260 261 262
	int len = len1 < len2 ? len1 : len2;
	int cmp;

	cmp = memcmp(name1, name2, len);
	if (cmp)
		return cmp;
	if (len1 < len2)
		return -1;
	if (len1 > len2)
		return 1;
Junio C Hamano's avatar
Junio C Hamano committed
263

264 265 266
	/* Compare stages  */
	flags1 &= CE_STAGEMASK;
	flags2 &= CE_STAGEMASK;
Junio C Hamano's avatar
Junio C Hamano committed
267

268 269 270 271
	if (flags1 < flags2)
		return -1;
	if (flags1 > flags2)
		return 1;
272 273 274 275 276 277 278 279 280 281 282 283
	return 0;
}

int cache_name_pos(const char *name, int namelen)
{
	int first, last;

	first = 0;
	last = active_nr;
	while (last > first) {
		int next = (last + first) >> 1;
		struct cache_entry *ce = active_cache[next];
284
		int cmp = cache_name_compare(name, namelen, ce->name, ntohs(ce->ce_flags));
285
		if (!cmp)
286
			return next;
287 288 289 290 291 292
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
293
	return -first-1;
294 295
}

296
/* Remove entry, return true if there are more entries to go.. */
297
int remove_cache_entry_at(int pos)
298
{
299
	active_cache_changed = 1;
300 301 302 303 304 305 306
	active_nr--;
	if (pos >= active_nr)
		return 0;
	memmove(active_cache + pos, active_cache + pos + 1, (active_nr - pos) * sizeof(struct cache_entry *));
	return 1;
}

Junio C Hamano's avatar
Junio C Hamano committed
307
int remove_file_from_cache(const char *path)
308 309
{
	int pos = cache_name_pos(path, strlen(path));
310 311 312
	if (pos < 0)
		pos = -pos-1;
	while (pos < active_nr && !strcmp(active_cache[pos]->name, path))
313
		remove_cache_entry_at(pos);
314 315 316
	return 0;
}

317
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
318 319 320 321 322
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342
int ce_path_match(const struct cache_entry *ce, const char **pathspec)
{
	const char *match, *name;
	int len;

	if (!pathspec)
		return 1;

	len = ce_namelen(ce);
	name = ce->name;
	while ((match = *pathspec++) != NULL) {
		int matchlen = strlen(match);
		if (matchlen > len)
			continue;
		if (memcmp(name, match, matchlen))
			continue;
		if (matchlen && name[matchlen-1] == '/')
			return 1;
		if (name[matchlen] == '/' || !name[matchlen])
			return 1;
343 344
		if (!matchlen)
			return 1;
345 346 347 348
	}
	return 0;
}

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
/*
 * We fundamentally don't like some paths: we don't want
 * dot or dot-dot anywhere, and for obvious reasons don't
 * want to recurse into ".git" either.
 *
 * Also, we don't want double slashes or slashes at the
 * end that can make pathnames ambiguous.
 */
static int verify_dotfile(const char *rest)
{
	/*
	 * The first character was '.', but that
	 * has already been discarded, we now test
	 * the rest.
	 */
	switch (*rest) {
	/* "." is not allowed */
	case '\0': case '/':
		return 0;

	/*
	 * ".git" followed by  NUL or slash is bad. This
	 * shares the path end test with the ".." case.
	 */
	case 'g':
		if (rest[1] != 'i')
			break;
		if (rest[2] != 't')
			break;
		rest += 2;
	/* fallthrough */
	case '.':
		if (rest[1] == '\0' || rest[1] == '/')
			return 0;
	}
	return 1;
}

int verify_path(const char *path)
{
	char c;

	goto inside;
	for (;;) {
		if (!c)
			return 1;
		if (c == '/') {
inside:
			c = *path++;
			switch (c) {
			default:
				continue;
			case '/': case '\0':
				break;
			case '.':
				if (verify_dotfile(path))
					continue;
			}
			return 0;
		}
		c = *path++;
	}
}

413 414 415
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
416
 */
417
static int has_file_name(const struct cache_entry *ce, int pos, int ok_to_replace)
418
{
419 420
	int retval = 0;
	int len = ce_namelen(ce);
421
	int stage = ce_stage(ce);
422
	const char *name = ce->name;
423

424 425
	while (pos < active_nr) {
		struct cache_entry *p = active_cache[pos++];
426

427
		if (len >= ce_namelen(p))
428
			break;
429 430
		if (memcmp(name, p->name, len))
			break;
431 432
		if (ce_stage(p) != stage)
			continue;
433 434 435 436 437 438
		if (p->name[len] != '/')
			continue;
		retval = -1;
		if (!ok_to_replace)
			break;
		remove_cache_entry_at(--pos);
439
	}
440 441
	return retval;
}
442

443 444 445 446 447 448 449
/*
 * Do we have another file with a pathname that is a proper
 * subset of the name we're trying to add?
 */
static int has_dir_name(const struct cache_entry *ce, int pos, int ok_to_replace)
{
	int retval = 0;
450
	int stage = ce_stage(ce);
451 452
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
453

454 455
	for (;;) {
		int len;
456

457 458 459 460 461 462 463
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
464

465
		pos = cache_name_pos(name, ntohs(create_ce_flags(len, stage)));
466 467 468 469
		if (pos >= 0) {
			retval = -1;
			if (ok_to_replace)
				break;
470
			remove_cache_entry_at(pos);
471 472 473 474 475 476
			continue;
		}

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
477
		 * we're ok, and we can exit.
478 479
		 */
		pos = -pos-1;
480
		while (pos < active_nr) {
481
			struct cache_entry *p = active_cache[pos];
482 483 484 485 486 487 488 489 490 491 492 493
			if ((ce_namelen(p) <= len) ||
			    (p->name[len] != '/') ||
			    memcmp(p->name, name, len))
				break; /* not our subdirectory */
			if (ce_stage(p) == stage)
				/* p is at the same stage as our entry, and
				 * is a subdirectory of what we are looking
				 * at, so we cannot have conflicts at our
				 * level or anything shorter.
				 */
				return retval;
			pos++;
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
	return retval;
}

/* We may be in a situation where we already have path/file and path
 * is being added, or we already have path and path/file is being
 * added.  Either one would result in a nonsense tree that has path
 * twice when git-write-tree tries to write it out.  Prevent it.
 * 
 * If ok-to-replace is specified, we remove the conflicting entries
 * from the cache so the caller should recompute the insert position.
 * When this happens, we return non-zero.
 */
static int check_file_directory_conflict(const struct cache_entry *ce, int pos, int ok_to_replace)
{
	/*
	 * We check if the path is a sub-path of a subsequent pathname
	 * first, since removing those will not change the position
	 * in the array
	 */
	int retval = has_file_name(ce, pos, ok_to_replace);
	/*
	 * Then check if the path might have a clashing sub-directory
	 * before it.
	 */
	return retval + has_dir_name(ce, pos, ok_to_replace);
521 522
}

523
int add_cache_entry(struct cache_entry *ce, int option)
524 525
{
	int pos;
526 527
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
528
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
Junio C Hamano's avatar
Junio C Hamano committed
529

530
	pos = cache_name_pos(ce->name, ntohs(ce->ce_flags));
531

Junio C Hamano's avatar
Junio C Hamano committed
532
	/* existing match? Just replace it. */
533
	if (pos >= 0) {
534
		active_cache_changed = 1;
535
		active_cache[pos] = ce;
536 537
		return 0;
	}
538
	pos = -pos-1;
539

540 541 542 543 544
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
	if (pos < active_nr && ce_stage(ce) == 0) {
545
		while (ce_same_name(active_cache[pos], ce)) {
546
			ok_to_add = 1;
547
			if (!remove_cache_entry_at(pos))
548 549 550 551
				break;
		}
	}

552 553
	if (!ok_to_add)
		return -1;
554 555
	if (!verify_path(ce->name))
		return -1;
556

Junio C Hamano's avatar
Junio C Hamano committed
557 558
	if (!skip_df_check &&
	    check_file_directory_conflict(ce, pos, ok_to_replace)) {
559 560
		if (!ok_to_replace)
			return -1;
561
		pos = cache_name_pos(ce->name, ntohs(ce->ce_flags));
562 563
		pos = -pos-1;
	}
564

565 566 567
	/* Make sure the array is big enough .. */
	if (active_nr == active_alloc) {
		active_alloc = alloc_nr(active_alloc);
568
		active_cache = xrealloc(active_cache, active_alloc * sizeof(struct cache_entry *));
569 570 571 572 573 574 575
	}

	/* Add it in.. */
	active_nr++;
	if (active_nr > pos)
		memmove(active_cache + pos + 1, active_cache + pos, (active_nr - pos - 1) * sizeof(ce));
	active_cache[pos] = ce;
576
	active_cache_changed = 1;
577 578 579
	return 0;
}

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
/* Three functions to allow overloaded pointer return; see linux/err.h */
static inline void *ERR_PTR(long error)
{
	return (void *) error;
}

static inline long PTR_ERR(const void *ptr)
{
	return (long) ptr;
}

static inline long IS_ERR(const void *ptr)
{
	return (unsigned long)ptr > (unsigned long)-1000L;
}

/*
 * "refresh" does not calculate a new sha1 file or bring the
 * cache up-to-date for mode/content changes. But what it
 * _does_ do is to "re-match" the stat information of a file
 * with the cache, so that you can refresh the cache for a
 * file that hasn't been changed but where the stat entry is
 * out of date.
 *
 * For example, you'd want to do this after doing a "git-read-tree",
 * to link up the stat cache details with the proper files.
 */
static struct cache_entry *refresh_entry(struct cache_entry *ce, int really)
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;

	if (lstat(ce->name, &st) < 0)
		return ERR_PTR(-errno);

	changed = ce_match_stat(ce, &st, really);
	if (!changed) {
		if (really && assume_unchanged &&
		    !(ce->ce_flags & htons(CE_VALID)))
			; /* mark this one VALID again */
		else
			return NULL;
	}

	if (ce_modified(ce, &st, really))
		return ERR_PTR(-EINVAL);

	size = ce_size(ce);
	updated = xmalloc(size);
	memcpy(updated, ce, size);
	fill_stat_cache_info(updated, &st);

	/* In this case, if really is not set, we should leave
	 * CE_VALID bit alone.  Otherwise, paths marked with
	 * --no-assume-unchanged (i.e. things to be edited) will
	 * reacquire CE_VALID bit automatically, which is not
	 * really what we want.
	 */
	if (!really && assume_unchanged && !(ce->ce_flags & htons(CE_VALID)))
		updated->ce_flags &= ~htons(CE_VALID);

	return updated;
}

int refresh_cache(unsigned int flags)
{
	int i;
	int has_errors = 0;
	int really = (flags & REFRESH_REALLY) != 0;
	int allow_unmerged = (flags & REFRESH_UNMERGED) != 0;
	int quiet = (flags & REFRESH_QUIET) != 0;
	int not_new = (flags & REFRESH_IGNORE_MISSING) != 0;

	for (i = 0; i < active_nr; i++) {
		struct cache_entry *ce, *new;
		ce = active_cache[i];
		if (ce_stage(ce)) {
			while ((i < active_nr) &&
			       ! strcmp(active_cache[i]->name, ce->name))
				i++;
			i--;
			if (allow_unmerged)
				continue;
			printf("%s: needs merge\n", ce->name);
			has_errors = 1;
			continue;
		}

		new = refresh_entry(ce, really);
		if (!new)
			continue;
		if (IS_ERR(new)) {
			if (not_new && PTR_ERR(new) == -ENOENT)
				continue;
			if (really && PTR_ERR(new) == -EINVAL) {
				/* If we are doing --really-refresh that
				 * means the index is not valid anymore.
				 */
				ce->ce_flags &= ~htons(CE_VALID);
				active_cache_changed = 1;
			}
			if (quiet)
				continue;
			printf("%s: needs update\n", ce->name);
			has_errors = 1;
			continue;
		}
		active_cache_changed = 1;
		/* You can NOT just free active_cache[i] here, since it
		 * might not be necessarily malloc()ed but can also come
		 * from mmap(). */
		active_cache[i] = new;
	}
	return has_errors;
}

697
static int verify_hdr(struct cache_header *hdr, unsigned long size)
698 699
{
	SHA_CTX c;
700
	unsigned char sha1[20];
701

702
	if (hdr->hdr_signature != htonl(CACHE_SIGNATURE))
703
		return error("bad signature");
704 705
	if (hdr->hdr_version != htonl(2))
		return error("bad index version");
706
	SHA1_Init(&c);
707
	SHA1_Update(&c, hdr, size - 20);
708
	SHA1_Final(sha1, &c);
709
	if (memcmp(sha1, (char *) hdr + size - 20, 20))
710
		return error("bad index file sha1 signature");
711 712 713
	return 0;
}

714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730
static int read_index_extension(const char *ext, void *data, unsigned long sz)
{
	switch (CACHE_EXT(ext)) {
	case CACHE_EXT_TREE:
		active_cache_tree = cache_tree_read(data, sz);
		break;
	default:
		if (*ext < 'A' || 'Z' < *ext)
			return error("index uses %.4s extension, which we do not understand",
				     ext);
		fprintf(stderr, "ignoring %.4s extension\n", ext);
		break;
	}
	return 0;
}

int read_cache(void)
731 732 733 734 735 736 737 738 739
{
	int fd, i;
	struct stat st;
	unsigned long size, offset;
	void *map;
	struct cache_header *hdr;

	errno = EBUSY;
	if (active_cache)
740 741
		return active_nr;

742
	errno = ENOENT;
Junio C Hamano's avatar
Junio C Hamano committed
743
	index_file_timestamp = 0;
744
	fd = open(get_index_file(), O_RDONLY);
745 746 747 748 749
	if (fd < 0) {
		if (errno == ENOENT)
			return 0;
		die("index file open failed (%s)", strerror(errno));
	}
750

751
	size = 0; /* avoid gcc warning */
Pavel Roskin's avatar
Pavel Roskin committed
752
	map = MAP_FAILED;
753 754 755
	if (!fstat(fd, &st)) {
		size = st.st_size;
		errno = EINVAL;
756
		if (size >= sizeof(struct cache_header) + 20)
757
			map = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
758 759
	}
	close(fd);
Pavel Roskin's avatar
Pavel Roskin committed
760
	if (map == MAP_FAILED)
761
		die("index file mmap failed (%s)", strerror(errno));
762 763

	hdr = map;
764
	if (verify_hdr(hdr, size) < 0)
765 766
		goto unmap;

767
	active_nr = ntohl(hdr->hdr_entries);
768
	active_alloc = alloc_nr(active_nr);
769
	active_cache = xcalloc(active_alloc, sizeof(struct cache_entry *));
770 771

	offset = sizeof(*hdr);
772
	for (i = 0; i < active_nr; i++) {
773
		struct cache_entry *ce = (struct cache_entry *) ((char *) map + offset);
774 775 776
		offset = offset + ce_size(ce);
		active_cache[i] = ce;
	}
Junio C Hamano's avatar
Junio C Hamano committed
777
	index_file_timestamp = st.st_mtime;
778 779 780 781 782 783 784 785
	while (offset <= size - 20 - 8) {
		/* After an array of active_nr index entries,
		 * there can be arbitrary number of extended
		 * sections, each of which is prefixed with
		 * extension name (4-byte) and section length
		 * in 4-byte network byte order.
		 */
		unsigned long extsize;
786
		memcpy(&extsize, (char *) map + offset + 4, 4);
787
		extsize = ntohl(extsize);
788 789 790
		if (read_index_extension(((const char *) map) + offset,
					 (char *) map + offset + 8,
					 extsize) < 0)
791 792 793 794
			goto unmap;
		offset += 8;
		offset += extsize;
	}
795 796 797 798 799
	return active_nr;

unmap:
	munmap(map, size);
	errno = EINVAL;
800
	die("index file corrupt");
801 802
}

803
#define WRITE_BUFFER_SIZE 8192
804
static unsigned char write_buffer[WRITE_BUFFER_SIZE];
805 806
static unsigned long write_buffer_len;

807
static int ce_write(SHA_CTX *context, int fd, void *data, unsigned int len)
808 809 810 811 812 813 814 815 816
{
	while (len) {
		unsigned int buffered = write_buffer_len;
		unsigned int partial = WRITE_BUFFER_SIZE - buffered;
		if (partial > len)
			partial = len;
		memcpy(write_buffer + buffered, data, partial);
		buffered += partial;
		if (buffered == WRITE_BUFFER_SIZE) {
817
			SHA1_Update(context, write_buffer, WRITE_BUFFER_SIZE);
818 819 820 821 822 823
			if (write(fd, write_buffer, WRITE_BUFFER_SIZE) != WRITE_BUFFER_SIZE)
				return -1;
			buffered = 0;
		}
		write_buffer_len = buffered;
		len -= partial;
824
		data = (char *) data + partial;
825 826 827 828
 	}
 	return 0;
}

829
static int write_index_ext_header(SHA_CTX *context, int fd,
830
				  unsigned int ext, unsigned int sz)
831 832 833 834 835 836 837 838 839 840
{
	ext = htonl(ext);
	sz = htonl(sz);
	if ((ce_write(context, fd, &ext, 4) < 0) ||
	    (ce_write(context, fd, &sz, 4) < 0))
		return -1;
	return 0;
}

static int ce_flush(SHA_CTX *context, int fd)
841 842
{
	unsigned int left = write_buffer_len;
843

844 845
	if (left) {
		write_buffer_len = 0;
846
		SHA1_Update(context, write_buffer, left);
847
	}
848

849 850 851 852 853 854 855
	/* Flush first if not enough space for SHA1 signature */
	if (left + 20 > WRITE_BUFFER_SIZE) {
		if (write(fd, write_buffer, left) != left)
			return -1;
		left = 0;
	}

856
	/* Append the SHA1 signature at the end */
857
	SHA1_Final(write_buffer + left, context);
858 859 860
	left += 20;
	if (write(fd, write_buffer, left) != left)
		return -1;
861 862 863
	return 0;
}

Junio C Hamano's avatar
Junio C Hamano committed
864 865 866 867 868 869 870 871 872 873 874 875 876 877 878
static void ce_smudge_racily_clean_entry(struct cache_entry *ce)
{
	/*
	 * The only thing we care about in this function is to smudge the
	 * falsely clean entry due to touch-update-touch race, so we leave
	 * everything else as they are.  We are called for entries whose
	 * ce_mtime match the index file mtime.
	 */
	struct stat st;

	if (lstat(ce->name, &st) < 0)
		return;
	if (ce_match_stat_basic(ce, &st))
		return;
	if (ce_modified_check_fs(ce, &st)) {
879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903
		/* This is "racily clean"; smudge it.  Note that this
		 * is a tricky code.  At first glance, it may appear
		 * that it can break with this sequence:
		 *
		 * $ echo xyzzy >frotz
		 * $ git-update-index --add frotz
		 * $ : >frotz
		 * $ sleep 3
		 * $ echo filfre >nitfol
		 * $ git-update-index --add nitfol
		 *
		 * but it does not.  Whe the second update-index runs,
		 * it notices that the entry "frotz" has the same timestamp
		 * as index, and if we were to smudge it by resetting its
		 * size to zero here, then the object name recorded
		 * in index is the 6-byte file but the cached stat information
		 * becomes zero --- which would then match what we would
		 * obtain from the filesystem next time we stat("frotz"). 
		 *
		 * However, the second update-index, before calling
		 * this function, notices that the cached size is 6
		 * bytes and what is on the filesystem is an empty
		 * file, and never calls us, so the cached size information
		 * for "frotz" stays 6 which does not match the filesystem.
		 */
Junio C Hamano's avatar
Junio C Hamano committed
904 905 906 907
		ce->ce_size = htonl(0);
	}
}

908
int write_cache(int newfd, struct cache_entry **cache, int entries)
909 910 911
{
	SHA_CTX c;
	struct cache_header hdr;
912 913 914 915 916
	int i, removed;

	for (i = removed = 0; i < entries; i++)
		if (!cache[i]->ce_mode)
			removed++;
917

918
	hdr.hdr_signature = htonl(CACHE_SIGNATURE);
919
	hdr.hdr_version = htonl(2);
920
	hdr.hdr_entries = htonl(entries - removed);
921 922

	SHA1_Init(&c);
923
	if (ce_write(&c, newfd, &hdr, sizeof(hdr)) < 0)
924 925 926 927
		return -1;

	for (i = 0; i < entries; i++) {
		struct cache_entry *ce = cache[i];
928 929
		if (!ce->ce_mode)
			continue;
Junio C Hamano's avatar
Junio C Hamano committed
930 931 932
		if (index_file_timestamp &&
		    index_file_timestamp <= ntohl(ce->ce_mtime.sec))
			ce_smudge_racily_clean_entry(ce);
933
		if (ce_write(&c, newfd, ce, ce_size(ce)) < 0)
934 935
			return -1;
	}
936

937 938 939 940 941 942 943 944 945 946 947 948 949 950
	/* Write extension data here */
	if (active_cache_tree) {
		unsigned long sz;
		void *data = cache_tree_write(active_cache_tree, &sz);
		if (data &&
		    !write_index_ext_header(&c, newfd, CACHE_EXT_TREE, sz) &&
		    !ce_write(&c, newfd, data, sz))
			;
		else {
			free(data);
			return -1;
		}
	}
	return ce_flush(&c, newfd);
951
}