read-cache.c 11.4 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 7 8
#include "cache.h"

struct cache_entry **active_cache = NULL;
9
unsigned int active_nr = 0, active_alloc = 0, active_cache_changed = 0;
10

11 12 13 14 15 16 17 18 19
/*
 * 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);
20
#ifdef USE_NSEC
21 22 23 24 25 26 27 28 29 30
	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);
}

31
int ce_match_stat(struct cache_entry *ce, struct stat *st)
32 33 34
{
	unsigned int changed = 0;

35 36 37
	switch (ntohl(ce->ce_mode) & S_IFMT) {
	case S_IFREG:
		changed |= !S_ISREG(st->st_mode) ? TYPE_CHANGED : 0;
38 39 40
		/* We consider only the owner x bit to be relevant for "mode changes" */
		if (0100 & (ntohl(ce->ce_mode) ^ st->st_mode))
			changed |= MODE_CHANGED;
41 42 43 44 45 46 47
		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));
	}
48
	if (ce->ce_mtime.sec != htonl(st->st_mtime))
49
		changed |= MTIME_CHANGED;
50 51 52
	if (ce->ce_ctime.sec != htonl(st->st_ctime))
		changed |= CTIME_CHANGED;

53
#ifdef USE_NSEC
54 55 56 57 58
	/*
	 * 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.
	 */
59
	if (ce->ce_mtime.nsec != htonl(st->st_mtim.tv_nsec))
60
		changed |= MTIME_CHANGED;
61
	if (ce->ce_ctime.nsec != htonl(st->st_ctim.tv_nsec))
62
		changed |= CTIME_CHANGED;
63 64 65 66
#endif	

	if (ce->ce_uid != htonl(st->st_uid) ||
	    ce->ce_gid != htonl(st->st_gid))
67
		changed |= OWNER_CHANGED;
68
	if (ce->ce_ino != htonl(st->st_ino))
69
		changed |= INODE_CHANGED;
70 71 72 73 74 75 76 77 78 79 80

#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

81
	if (ce->ce_size != htonl(st->st_size))
82 83 84 85
		changed |= DATA_CHANGED;
	return changed;
}

86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104
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;
}

105
int cache_name_compare(const char *name1, int flags1, const char *name2, int flags2)
106
{
107 108
	int len1 = flags1 & CE_NAMEMASK;
	int len2 = flags2 & CE_NAMEMASK;
109 110 111 112 113 114 115 116 117 118
	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;
119 120 121 122
	if (flags1 < flags2)
		return -1;
	if (flags1 > flags2)
		return 1;
123 124 125 126 127 128 129 130 131 132 133 134
	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];
135
		int cmp = cache_name_compare(name, namelen, ce->name, htons(ce->ce_flags));
136
		if (!cmp)
137
			return next;
138 139 140 141 142 143
		if (cmp < 0) {
			last = next;
			continue;
		}
		first = next+1;
	}
144
	return -first-1;
145 146
}

147
/* Remove entry, return true if there are more entries to go.. */
148
int remove_cache_entry_at(int pos)
149
{
150
	active_cache_changed = 1;
151 152 153 154 155 156 157
	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;
}

158 159 160
int remove_file_from_cache(char *path)
{
	int pos = cache_name_pos(path, strlen(path));
161 162 163
	if (pos < 0)
		pos = -pos-1;
	while (pos < active_nr && !strcmp(active_cache[pos]->name, path))
164
		remove_cache_entry_at(pos);
165 166 167
	return 0;
}

168
int ce_same_name(struct cache_entry *a, struct cache_entry *b)
169 170 171 172 173
{
	int len = ce_namelen(a);
	return ce_namelen(b) == len && !memcmp(a->name, b->name, len);
}

174 175 176 177
/* 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.
178 179 180 181
 * 
 * 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.
182
 */
183 184
static int check_file_directory_conflict(const struct cache_entry *ce,
					 int ok_to_replace)
185
{
186
	int pos, replaced = 0;
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202
	const char *path = ce->name;
	int namelen = strlen(path);
	int stage = ce_stage(ce);
	char *pathbuf = xmalloc(namelen + 1);
	char *cp;

	memcpy(pathbuf, path, namelen + 1);

	/*
	 * We are inserting path/file.  Do they have path registered at
	 * the same stage?  We need to do this for all the levels of our
	 * subpath.
	 */
	cp = pathbuf;
	while (1) {
		char *ep = strchr(cp, '/');
203
		int len;
204
		if (!ep)
205 206
			break;
		*ep = 0;    /* first cut it at slash */
207
		len = ep - pathbuf;
208
		pos = cache_name_pos(pathbuf,
209
				     htons(create_ce_flags(len, stage)));
210 211 212 213 214
		if (0 <= pos) {
			/* Our leading path component is registered as a file,
			 * and we are trying to make it a directory.  This is
			 * bad.
			 */
215 216 217 218 219
			if (!ok_to_replace) {
				free(pathbuf);
				return -1;
			}
			fprintf(stderr, "removing file '%s' to replace it with a directory to create '%s'.\n", pathbuf, path);
220
			remove_cache_entry_at(pos);
221
			replaced = 1;
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
		}
		*ep = '/';  /* then restore it and go downwards */
		cp = ep + 1;
	}
	free(pathbuf);

	/* Do we have an entry in the cache that makes our path a prefix
	 * of it?  That is, are we creating a file where they already expect
	 * a directory there?
	 */
	pos = cache_name_pos(path,
			     htons(create_ce_flags(namelen, stage)));

	/* (0 <= pos) cannot happen because add_cache_entry()
	 * should have taken care of that case.
	 */
	pos = -pos-1;

	/* pos would point at an existing entry that would come immediately
	 * after our path.  It could be the same as our path in higher stage,
	 * or different path but in a lower stage.
	 *
	 * E.g. when we are inserting path at stage 2,
	 *
	 *        1 path
	 * pos->  3 path
248 249 250 251
	 *        2 path/file1
	 *        3 path/file1
	 *        2 path/file2
	 *        2 patho
252 253 254
	 *
	 * We need to examine pos, ignore it because it is at different
	 * stage, examine next to find the path/file at stage 2, and
255 256
	 * complain.  We need to do this until we are not the leading
	 * path of an existing entry anymore.
257 258 259 260 261 262
	 */

	while (pos < active_nr) {
		struct cache_entry *other = active_cache[pos];
		if (strncmp(other->name, path, namelen))
			break; /* it is not our "subdirectory" anymore */
263 264 265 266 267
		if ((ce_stage(other) == stage) &&
		    other->name[namelen] == '/') {
			if (!ok_to_replace)
				return -1;
			fprintf(stderr, "removing file '%s' under '%s' to be replaced with a file\n", other->name, path);
268
			remove_cache_entry_at(pos);
269 270 271
			replaced = 1;
			continue; /* cycle without updating pos */
		}
272 273
		pos++;
	}
274
	return replaced;
275 276
}

277
int add_cache_entry(struct cache_entry *ce, int option)
278 279
{
	int pos;
280 281
	int ok_to_add = option & ADD_CACHE_OK_TO_ADD;
	int ok_to_replace = option & ADD_CACHE_OK_TO_REPLACE;
282
	pos = cache_name_pos(ce->name, htons(ce->ce_flags));
283 284

	/* existing match? Just replace it */
285
	if (pos >= 0) {
286
		active_cache_changed = 1;
287
		active_cache[pos] = ce;
288 289
		return 0;
	}
290
	pos = -pos-1;
291

292 293 294 295 296
	/*
	 * Inserting a merged entry ("stage 0") into the index
	 * will always replace all non-merged entries..
	 */
	if (pos < active_nr && ce_stage(ce) == 0) {
297
		while (ce_same_name(active_cache[pos], ce)) {
298
			ok_to_add = 1;
299
			if (!remove_cache_entry_at(pos))
300 301 302 303
				break;
		}
	}

304 305 306
	if (!ok_to_add)
		return -1;

307 308 309 310 311 312
	if (check_file_directory_conflict(ce, ok_to_replace)) {
		if (!ok_to_replace)
			return -1;
		pos = cache_name_pos(ce->name, htons(ce->ce_flags));
		pos = -pos-1;
	}
313

314 315 316
	/* Make sure the array is big enough .. */
	if (active_nr == active_alloc) {
		active_alloc = alloc_nr(active_alloc);
317
		active_cache = xrealloc(active_cache, active_alloc * sizeof(struct cache_entry *));
318 319 320 321 322 323 324
	}

	/* 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;
325
	active_cache_changed = 1;
326 327 328
	return 0;
}

329 330 331 332 333
static int verify_hdr(struct cache_header *hdr, unsigned long size)
{
	SHA_CTX c;
	unsigned char sha1[20];

334
	if (hdr->hdr_signature != htonl(CACHE_SIGNATURE))
335
		return error("bad signature");
336 337
	if (hdr->hdr_version != htonl(2))
		return error("bad index version");
338
	SHA1_Init(&c);
339
	SHA1_Update(&c, hdr, size - 20);
340
	SHA1_Final(sha1, &c);
341 342
	if (memcmp(sha1, (void *)hdr + size - 20, 20))
		return error("bad index file sha1 signature");
343 344 345 346 347 348 349 350 351 352 353 354 355 356 357
	return 0;
}

int read_cache(void)
{
	int fd, i;
	struct stat st;
	unsigned long size, offset;
	void *map;
	struct cache_header *hdr;

	errno = EBUSY;
	if (active_cache)
		return error("more than one cachefile");
	errno = ENOENT;
358
	fd = open(get_index_file(), O_RDONLY);
359 360 361
	if (fd < 0)
		return (errno == ENOENT) ? 0 : error("open failed");

362
	size = 0; // avoid gcc warning
363 364 365 366
	map = (void *)-1;
	if (!fstat(fd, &st)) {
		size = st.st_size;
		errno = EINVAL;
367
		if (size >= sizeof(struct cache_header) + 20)
368
			map = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
369 370 371 372 373 374 375 376 377
	}
	close(fd);
	if (-1 == (int)(long)map)
		return error("mmap failed");

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

378
	active_nr = ntohl(hdr->hdr_entries);
379 380 381 382
	active_alloc = alloc_nr(active_nr);
	active_cache = calloc(active_alloc, sizeof(struct cache_entry *));

	offset = sizeof(*hdr);
383
	for (i = 0; i < active_nr; i++) {
384 385 386 387 388 389 390 391 392 393 394 395
		struct cache_entry *ce = map + offset;
		offset = offset + ce_size(ce);
		active_cache[i] = ce;
	}
	return active_nr;

unmap:
	munmap(map, size);
	errno = EINVAL;
	return error("verify header failed");
}

396
#define WRITE_BUFFER_SIZE 8192
397
static unsigned char write_buffer[WRITE_BUFFER_SIZE];
398 399
static unsigned long write_buffer_len;

400
static int ce_write(SHA_CTX *context, int fd, void *data, unsigned int len)
401 402 403 404 405 406 407 408 409
{
	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) {
410
			SHA1_Update(context, write_buffer, WRITE_BUFFER_SIZE);
411 412 413 414 415 416 417 418 419 420 421
			if (write(fd, write_buffer, WRITE_BUFFER_SIZE) != WRITE_BUFFER_SIZE)
				return -1;
			buffered = 0;
		}
		write_buffer_len = buffered;
		len -= partial;
		data += partial;
 	}
 	return 0;
}

422
static int ce_flush(SHA_CTX *context, int fd)
423 424
{
	unsigned int left = write_buffer_len;
425

426 427
	if (left) {
		write_buffer_len = 0;
428
		SHA1_Update(context, write_buffer, left);
429
	}
430 431 432 433 434 435

	/* Append the SHA1 signature at the end */
	SHA1_Final(write_buffer + left, context);
	left += 20;
	if (write(fd, write_buffer, left) != left)
		return -1;
436 437 438
	return 0;
}

439 440 441 442 443 444
int write_cache(int newfd, struct cache_entry **cache, int entries)
{
	SHA_CTX c;
	struct cache_header hdr;
	int i;

445
	hdr.hdr_signature = htonl(CACHE_SIGNATURE);
446
	hdr.hdr_version = htonl(2);
447
	hdr.hdr_entries = htonl(entries);
448 449

	SHA1_Init(&c);
450
	if (ce_write(&c, newfd, &hdr, sizeof(hdr)) < 0)
451 452 453 454
		return -1;

	for (i = 0; i < entries; i++) {
		struct cache_entry *ce = cache[i];
455
		if (ce_write(&c, newfd, ce, ce_size(ce)) < 0)
456 457
			return -1;
	}
458
	return ce_flush(&c, newfd);
459
}