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
int cache_errno = 0;

static void *cache_mmap = NULL;
static size_t cache_mmap_size = 0;

32 33 34 35 36 37 38 39 40
/*
 * 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);
41
#ifdef USE_NSEC
42 43 44 45 46 47 48 49
	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
50 51 52

	if (assume_unchanged)
		ce->ce_flags |= htons(CE_VALID);
53 54
}

Junio C Hamano's avatar
Junio C Hamano committed
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
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
113
static int ce_match_stat_basic(struct cache_entry *ce, struct stat *st)
114 115 116
{
	unsigned int changed = 0;

117 118 119
	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
120 121 122 123 124
		/* 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)))
125
			changed |= MODE_CHANGED;
126 127 128 129 130 131 132
		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));
	}
133
	if (ce->ce_mtime.sec != htonl(st->st_mtime))
134
		changed |= MTIME_CHANGED;
135 136 137
	if (ce->ce_ctime.sec != htonl(st->st_ctime))
		changed |= CTIME_CHANGED;

138
#ifdef USE_NSEC
139 140 141 142 143
	/*
	 * 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.
	 */
144
	if (ce->ce_mtime.nsec != htonl(st->st_mtim.tv_nsec))
145
		changed |= MTIME_CHANGED;
146
	if (ce->ce_ctime.nsec != htonl(st->st_ctim.tv_nsec))
147
		changed |= CTIME_CHANGED;
148 149 150 151
#endif	

	if (ce->ce_uid != htonl(st->st_uid) ||
	    ce->ce_gid != htonl(st->st_gid))
152
		changed |= OWNER_CHANGED;
153
	if (ce->ce_ino != htonl(st->st_ino))
154
		changed |= INODE_CHANGED;
155 156 157 158 159 160 161 162 163 164 165

#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

166
	if (ce->ce_size != htonl(st->st_size))
167
		changed |= DATA_CHANGED;
168

Junio C Hamano's avatar
Junio C Hamano committed
169 170 171
	return changed;
}

Junio C Hamano's avatar
Junio C Hamano committed
172
int ce_match_stat(struct cache_entry *ce, struct stat *st, int ignore_valid)
Junio C Hamano's avatar
Junio C Hamano committed
173
{
Junio C Hamano's avatar
Junio C Hamano committed
174 175 176 177 178 179 180 181 182 183
	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
184

Junio C Hamano's avatar
Junio C Hamano committed
185 186 187 188 189 190 191 192 193 194 195 196 197 198
	/*
	 * 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
199
	 * carefully than others.
Junio C Hamano's avatar
Junio C Hamano committed
200 201 202 203 204
	 */
	if (!changed &&
	    index_file_timestamp &&
	    index_file_timestamp <= ntohl(ce->ce_mtime.sec))
		changed |= ce_modified_check_fs(ce, st);
205

Junio C Hamano's avatar
Junio C Hamano committed
206
	return changed;
207 208
}

Junio C Hamano's avatar
Junio C Hamano committed
209
int ce_modified(struct cache_entry *ce, struct stat *st, int really)
210
{
Junio C Hamano's avatar
Junio C Hamano committed
211
	int changed, changed_fs;
Junio C Hamano's avatar
Junio C Hamano committed
212
	changed = ce_match_stat(ce, st, really);
213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228
	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
229 230 231
	changed_fs = ce_modified_check_fs(ce, st);
	if (changed_fs)
		return changed | changed_fs;
232 233 234
	return 0;
}

235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
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;
}

254
int cache_name_compare(const char *name1, int flags1, const char *name2, int flags2)
255
{
256 257
	int len1 = flags1 & CE_NAMEMASK;
	int len2 = flags2 & CE_NAMEMASK;
258 259 260 261 262 263 264 265 266 267
	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
268

269 270 271
	/* Compare stages  */
	flags1 &= CE_STAGEMASK;
	flags2 &= CE_STAGEMASK;
Junio C Hamano's avatar
Junio C Hamano committed
272

273 274 275 276
	if (flags1 < flags2)
		return -1;
	if (flags1 > flags2)
		return 1;
277 278 279 280 281 282 283 284 285 286 287 288
	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];
289
		int cmp = cache_name_compare(name, namelen, ce->name, ntohs(ce->ce_flags));
290
		if (!cmp)
291
			return next;
292 293 294 295 296 297
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
298
	return -first-1;
299 300
}

301
/* Remove entry, return true if there are more entries to go.. */
302
int remove_cache_entry_at(int pos)
303
{
304
	active_cache_changed = 1;
305 306 307 308 309 310 311
	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
312
int remove_file_from_cache(const char *path)
313 314
{
	int pos = cache_name_pos(path, strlen(path));
315 316 317
	if (pos < 0)
		pos = -pos-1;
	while (pos < active_nr && !strcmp(active_cache[pos]->name, path))
318
		remove_cache_entry_at(pos);
319 320 321
	return 0;
}

322
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
323 324 325 326 327
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347
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;
348 349
		if (!matchlen)
			return 1;
350 351 352 353
	}
	return 0;
}

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
/*
 * 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++;
	}
}

418 419 420
/*
 * Do we have another file that has the beginning components being a
 * proper superset of the name we're trying to add?
421
 */
422
static int has_file_name(const struct cache_entry *ce, int pos, int ok_to_replace)
423
{
424 425
	int retval = 0;
	int len = ce_namelen(ce);
426
	int stage = ce_stage(ce);
427
	const char *name = ce->name;
428

429 430
	while (pos < active_nr) {
		struct cache_entry *p = active_cache[pos++];
431

432
		if (len >= ce_namelen(p))
433
			break;
434 435
		if (memcmp(name, p->name, len))
			break;
436 437
		if (ce_stage(p) != stage)
			continue;
438 439 440 441 442 443
		if (p->name[len] != '/')
			continue;
		retval = -1;
		if (!ok_to_replace)
			break;
		remove_cache_entry_at(--pos);
444
	}
445 446
	return retval;
}
447

448 449 450 451 452 453 454
/*
 * 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;
455
	int stage = ce_stage(ce);
456 457
	const char *name = ce->name;
	const char *slash = name + ce_namelen(ce);
458

459 460
	for (;;) {
		int len;
461

462 463 464 465 466 467 468
		for (;;) {
			if (*--slash == '/')
				break;
			if (slash <= ce->name)
				return retval;
		}
		len = slash - name;
469

470
		pos = cache_name_pos(name, ntohs(create_ce_flags(len, stage)));
471 472 473 474
		if (pos >= 0) {
			retval = -1;
			if (ok_to_replace)
				break;
475
			remove_cache_entry_at(pos);
476 477 478 479 480 481
			continue;
		}

		/*
		 * Trivial optimization: if we find an entry that
		 * already matches the sub-directory, then we know
482
		 * we're ok, and we can exit.
483 484
		 */
		pos = -pos-1;
485
		while (pos < active_nr) {
486
			struct cache_entry *p = active_cache[pos];
487 488 489 490 491 492 493 494 495 496 497 498
			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++;
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
	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);
526 527
}

528
int add_cache_entry(struct cache_entry *ce, int option)
529 530
{
	int pos;
531 532
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
533
	int skip_df_check = option & ADD_CACHE_SKIP_DFCHECK;
Junio C Hamano's avatar
Junio C Hamano committed
534

535
	pos = cache_name_pos(ce->name, ntohs(ce->ce_flags));
536

Junio C Hamano's avatar
Junio C Hamano committed
537
	/* existing match? Just replace it. */
538
	if (pos >= 0) {
539
		active_cache_changed = 1;
540
		active_cache[pos] = ce;
541 542
		return 0;
	}
543
	pos = -pos-1;
544

545 546 547 548 549
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
	if (pos < active_nr && ce_stage(ce) == 0) {
550
		while (ce_same_name(active_cache[pos], ce)) {
551
			ok_to_add = 1;
552
			if (!remove_cache_entry_at(pos))
553 554 555 556
				break;
		}
	}

557 558
	if (!ok_to_add)
		return -1;
559 560
	if (!verify_path(ce->name))
		return -1;
561

Junio C Hamano's avatar
Junio C Hamano committed
562 563
	if (!skip_df_check &&
	    check_file_directory_conflict(ce, pos, ok_to_replace)) {
564 565
		if (!ok_to_replace)
			return -1;
566
		pos = cache_name_pos(ce->name, ntohs(ce->ce_flags));
567 568
		pos = -pos-1;
	}
569

570 571 572
	/* Make sure the array is big enough .. */
	if (active_nr == active_alloc) {
		active_alloc = alloc_nr(active_alloc);
573
		active_cache = xrealloc(active_cache, active_alloc * sizeof(struct cache_entry *));
574 575 576 577 578 579 580
	}

	/* 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;
581
	active_cache_changed = 1;
582 583 584
	return 0;
}

585 586 587 588 589 590 591 592 593 594 595
/*
 * "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.
 */
596
struct cache_entry *refresh_cache_entry(struct cache_entry *ce, int really)
597 598 599 600 601
{
	struct stat st;
	struct cache_entry *updated;
	int changed, size;

602 603 604 605
	if (lstat(ce->name, &st) < 0) {
		cache_errno = errno;
		return NULL;
	}
606 607 608 609 610 611 612

	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
613
			return ce;
614 615
	}

616 617 618 619
	if (ce_modified(ce, &st, really)) {
		cache_errno = EINVAL;
		return NULL;
	}
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

	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;
		}

662 663
		new = refresh_cache_entry(ce, really);
		if (new == ce)
664
			continue;
665 666
		if (!new) {
			if (not_new && cache_errno == ENOENT)
667
				continue;
668
			if (really && cache_errno == EINVAL) {
669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689
				/* 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;
}

690
static int verify_hdr(struct cache_header *hdr, unsigned long size)
691 692
{
	SHA_CTX c;
693
	unsigned char sha1[20];
694

695
	if (hdr->hdr_signature != htonl(CACHE_SIGNATURE))
696
		return error("bad signature");
697 698
	if (hdr->hdr_version != htonl(2))
		return error("bad index version");
699
	SHA1_Init(&c);
700
	SHA1_Update(&c, hdr, size - 20);
701
	SHA1_Final(sha1, &c);
702
	if (memcmp(sha1, (char *) hdr + size - 20, 20))
703
		return error("bad index file sha1 signature");
704 705 706
	return 0;
}

707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723
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)
724 725 726 727 728 729
{
	return read_cache_from(get_index_file());
}

/* remember to discard_cache() before reading a different cache! */
int read_cache_from(const char *path)
730 731 732
{
	int fd, i;
	struct stat st;
733
	unsigned long offset;
734 735 736
	struct cache_header *hdr;

	errno = EBUSY;
737
	if (cache_mmap)
738 739
		return active_nr;

740
	errno = ENOENT;
Junio C Hamano's avatar
Junio C Hamano committed
741
	index_file_timestamp = 0;
742
	fd = open(path, O_RDONLY);
743 744 745 746 747
	if (fd < 0) {
		if (errno == ENOENT)
			return 0;
		die("index file open failed (%s)", strerror(errno));
	}
748

749
	cache_mmap = MAP_FAILED;
750
	if (!fstat(fd, &st)) {
751
		cache_mmap_size = st.st_size;
752
		errno = EINVAL;
753 754
		if (cache_mmap_size >= sizeof(struct cache_header) + 20)
			cache_mmap = mmap(NULL, cache_mmap_size, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
755 756
	}
	close(fd);
757
	if (cache_mmap == MAP_FAILED)
758
		die("index file mmap failed (%s)", strerror(errno));
759

760 761
	hdr = cache_mmap;
	if (verify_hdr(hdr, cache_mmap_size) < 0)
762 763
		goto unmap;

764
	active_nr = ntohl(hdr->hdr_entries);
765
	active_alloc = alloc_nr(active_nr);
766
	active_cache = xcalloc(active_alloc, sizeof(struct cache_entry *));
767 768

	offset = sizeof(*hdr);
769
	for (i = 0; i < active_nr; i++) {
770
		struct cache_entry *ce = (struct cache_entry *) ((char *) cache_mmap + offset);
771 772 773
		offset = offset + ce_size(ce);
		active_cache[i] = ce;
	}
Junio C Hamano's avatar
Junio C Hamano committed
774
	index_file_timestamp = st.st_mtime;
775
	while (offset <= cache_mmap_size - 20 - 8) {
776 777 778 779 780 781 782
		/* 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;
783
		memcpy(&extsize, (char *) cache_mmap + offset + 4, 4);
784
		extsize = ntohl(extsize);
785 786
		if (read_index_extension(((const char *) cache_mmap) + offset,
					 (char *) cache_mmap + offset + 8,
787
					 extsize) < 0)
788 789 790 791
			goto unmap;
		offset += 8;
		offset += extsize;
	}
792 793 794
	return active_nr;

unmap:
795
	munmap(cache_mmap, cache_mmap_size);
796
	errno = EINVAL;
797
	die("index file corrupt");
798 799
}

800
#define WRITE_BUFFER_SIZE 8192
801
static unsigned char write_buffer[WRITE_BUFFER_SIZE];
802 803
static unsigned long write_buffer_len;

804
static int ce_write(SHA_CTX *context, int fd, void *data, unsigned int len)
805 806 807 808 809 810 811 812 813
{
	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) {
814
			SHA1_Update(context, write_buffer, WRITE_BUFFER_SIZE);
815 816 817 818 819 820
			if (write(fd, write_buffer, WRITE_BUFFER_SIZE) != WRITE_BUFFER_SIZE)
				return -1;
			buffered = 0;
		}
		write_buffer_len = buffered;
		len -= partial;
821
		data = (char *) data + partial;
822 823 824 825
 	}
 	return 0;
}

826
static int write_index_ext_header(SHA_CTX *context, int fd,
827
				  unsigned int ext, unsigned int sz)
828 829 830 831 832 833 834 835 836 837
{
	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)
838 839
{
	unsigned int left = write_buffer_len;
840

841 842
	if (left) {
		write_buffer_len = 0;
843
		SHA1_Update(context, write_buffer, left);
844
	}
845

846 847 848 849 850 851 852
	/* 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;
	}

853
	/* Append the SHA1 signature at the end */
854
	SHA1_Final(write_buffer + left, context);
855 856 857
	left += 20;
	if (write(fd, write_buffer, left) != left)
		return -1;
858 859 860
	return 0;
}

Junio C Hamano's avatar
Junio C Hamano committed
861 862 863 864 865 866 867 868 869 870 871 872 873 874 875
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)) {
876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900
		/* 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
901 902 903 904
		ce->ce_size = htonl(0);
	}
}

905
int write_cache(int newfd, struct cache_entry **cache, int entries)
906 907 908
{
	SHA_CTX c;
	struct cache_header hdr;
909 910 911 912 913
	int i, removed;

	for (i = removed = 0; i < entries; i++)
		if (!cache[i]->ce_mode)
			removed++;
914

915
	hdr.hdr_signature = htonl(CACHE_SIGNATURE);
916
	hdr.hdr_version = htonl(2);
917
	hdr.hdr_entries = htonl(entries - removed);
918 919

	SHA1_Init(&c);
920
	if (ce_write(&c, newfd, &hdr, sizeof(hdr)) < 0)
921 922 923 924
		return -1;

	for (i = 0; i < entries; i++) {
		struct cache_entry *ce = cache[i];
925 926
		if (!ce->ce_mode)
			continue;
Junio C Hamano's avatar
Junio C Hamano committed
927 928 929
		if (index_file_timestamp &&
		    index_file_timestamp <= ntohl(ce->ce_mtime.sec))
			ce_smudge_racily_clean_entry(ce);
930
		if (ce_write(&c, newfd, ce, ce_size(ce)) < 0)
931 932
			return -1;
	}
933

934 935 936 937 938 939 940 941 942 943 944 945 946 947
	/* 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);
948
}