2007-05-02 18:13:14 +02:00
|
|
|
#include "cache.h"
|
|
|
|
#include "pack.h"
|
2007-06-01 21:18:05 +02:00
|
|
|
#include "csum-file.h"
|
|
|
|
|
2011-02-26 00:43:25 +01:00
|
|
|
void reset_pack_idx_option(struct pack_idx_option *opts)
|
|
|
|
{
|
|
|
|
memset(opts, 0, sizeof(*opts));
|
|
|
|
opts->version = 2;
|
|
|
|
opts->off32_limit = 0x7fffffff;
|
|
|
|
}
|
2007-06-01 21:18:05 +02:00
|
|
|
|
|
|
|
static int sha1_compare(const void *_a, const void *_b)
|
|
|
|
{
|
|
|
|
struct pack_idx_entry *a = *(struct pack_idx_entry **)_a;
|
|
|
|
struct pack_idx_entry *b = *(struct pack_idx_entry **)_b;
|
|
|
|
return hashcmp(a->sha1, b->sha1);
|
|
|
|
}
|
|
|
|
|
2011-02-26 01:55:26 +01:00
|
|
|
static int cmp_uint32(const void *a_, const void *b_)
|
|
|
|
{
|
|
|
|
uint32_t a = *((uint32_t *)a_);
|
|
|
|
uint32_t b = *((uint32_t *)b_);
|
|
|
|
|
|
|
|
return (a < b) ? -1 : (a != b);
|
|
|
|
}
|
|
|
|
|
2011-02-26 01:54:00 +01:00
|
|
|
static int need_large_offset(off_t offset, const struct pack_idx_option *opts)
|
|
|
|
{
|
2011-02-26 01:55:26 +01:00
|
|
|
uint32_t ofsval;
|
|
|
|
|
|
|
|
if ((offset >> 31) || (opts->off32_limit < offset))
|
|
|
|
return 1;
|
|
|
|
if (!opts->anomaly_nr)
|
|
|
|
return 0;
|
|
|
|
ofsval = offset;
|
|
|
|
return !!bsearch(&ofsval, opts->anomaly, opts->anomaly_nr,
|
|
|
|
sizeof(ofsval), cmp_uint32);
|
2011-02-26 01:54:00 +01:00
|
|
|
}
|
|
|
|
|
2007-06-01 21:18:05 +02:00
|
|
|
/*
|
|
|
|
* On entry *sha1 contains the pack content SHA1 hash, on exit it is
|
|
|
|
* the SHA1 hash of sorted object names. The objects array passed in
|
|
|
|
* will be sorted by SHA1 on exit.
|
|
|
|
*/
|
2010-01-22 16:55:19 +01:00
|
|
|
const char *write_idx_file(const char *index_name, struct pack_idx_entry **objects,
|
2011-02-26 00:43:25 +01:00
|
|
|
int nr_objects, const struct pack_idx_option *opts,
|
|
|
|
unsigned char *sha1)
|
2007-06-01 21:18:05 +02:00
|
|
|
{
|
|
|
|
struct sha1file *f;
|
|
|
|
struct pack_idx_entry **sorted_by_sha, **list, **last;
|
|
|
|
off_t last_obj_offset = 0;
|
|
|
|
uint32_t array[256];
|
|
|
|
int i, fd;
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA_CTX ctx;
|
2007-06-01 21:18:05 +02:00
|
|
|
uint32_t index_version;
|
|
|
|
|
|
|
|
if (nr_objects) {
|
|
|
|
sorted_by_sha = objects;
|
|
|
|
list = sorted_by_sha;
|
|
|
|
last = sorted_by_sha + nr_objects;
|
|
|
|
for (i = 0; i < nr_objects; ++i) {
|
|
|
|
if (objects[i]->offset > last_obj_offset)
|
|
|
|
last_obj_offset = objects[i]->offset;
|
|
|
|
}
|
|
|
|
qsort(sorted_by_sha, nr_objects, sizeof(sorted_by_sha[0]),
|
|
|
|
sha1_compare);
|
|
|
|
}
|
|
|
|
else
|
|
|
|
sorted_by_sha = list = last = NULL;
|
|
|
|
|
2011-02-03 02:29:01 +01:00
|
|
|
if (opts->flags & WRITE_IDX_VERIFY) {
|
|
|
|
assert(index_name);
|
|
|
|
f = sha1fd_check(index_name);
|
2007-06-01 21:18:05 +02:00
|
|
|
} else {
|
2011-02-03 02:29:01 +01:00
|
|
|
if (!index_name) {
|
|
|
|
static char tmpfile[PATH_MAX];
|
|
|
|
fd = odb_mkstemp(tmpfile, sizeof(tmpfile), "pack/tmp_idx_XXXXXX");
|
|
|
|
index_name = xstrdup(tmpfile);
|
|
|
|
} else {
|
|
|
|
unlink(index_name);
|
|
|
|
fd = open(index_name, O_CREAT|O_EXCL|O_WRONLY, 0600);
|
|
|
|
}
|
|
|
|
if (fd < 0)
|
|
|
|
die_errno("unable to create '%s'", index_name);
|
|
|
|
f = sha1fd(fd, index_name);
|
2007-06-01 21:18:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/* if last object's offset is >= 2^31 we should use index V2 */
|
2011-02-26 01:54:00 +01:00
|
|
|
index_version = need_large_offset(last_obj_offset, opts) ? 2 : opts->version;
|
2007-06-01 21:18:05 +02:00
|
|
|
|
|
|
|
/* index versions 2 and above need a header */
|
|
|
|
if (index_version >= 2) {
|
|
|
|
struct pack_idx_header hdr;
|
|
|
|
hdr.idx_signature = htonl(PACK_IDX_SIGNATURE);
|
|
|
|
hdr.idx_version = htonl(index_version);
|
|
|
|
sha1write(f, &hdr, sizeof(hdr));
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Write the first-level table (the list is sorted,
|
|
|
|
* but we use a 256-entry lookup to be able to avoid
|
|
|
|
* having to do eight extra binary search iterations).
|
|
|
|
*/
|
|
|
|
for (i = 0; i < 256; i++) {
|
|
|
|
struct pack_idx_entry **next = list;
|
|
|
|
while (next < last) {
|
|
|
|
struct pack_idx_entry *obj = *next;
|
|
|
|
if (obj->sha1[0] != i)
|
|
|
|
break;
|
|
|
|
next++;
|
|
|
|
}
|
|
|
|
array[i] = htonl(next - sorted_by_sha);
|
|
|
|
list = next;
|
|
|
|
}
|
|
|
|
sha1write(f, array, 256 * 4);
|
|
|
|
|
|
|
|
/* compute the SHA1 hash of sorted object names. */
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Init(&ctx);
|
2007-06-01 21:18:05 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Write the actual SHA1 entries..
|
|
|
|
*/
|
|
|
|
list = sorted_by_sha;
|
|
|
|
for (i = 0; i < nr_objects; i++) {
|
|
|
|
struct pack_idx_entry *obj = *list++;
|
|
|
|
if (index_version < 2) {
|
|
|
|
uint32_t offset = htonl(obj->offset);
|
|
|
|
sha1write(f, &offset, 4);
|
|
|
|
}
|
|
|
|
sha1write(f, obj->sha1, 20);
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Update(&ctx, obj->sha1, 20);
|
2011-11-17 07:04:13 +01:00
|
|
|
if ((opts->flags & WRITE_IDX_STRICT) &&
|
|
|
|
(i && !hashcmp(list[-2]->sha1, obj->sha1)))
|
|
|
|
die("The same object %s appears twice in the pack",
|
|
|
|
sha1_to_hex(obj->sha1));
|
2007-06-01 21:18:05 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
if (index_version >= 2) {
|
|
|
|
unsigned int nr_large_offset = 0;
|
|
|
|
|
|
|
|
/* write the crc32 table */
|
|
|
|
list = sorted_by_sha;
|
|
|
|
for (i = 0; i < nr_objects; i++) {
|
|
|
|
struct pack_idx_entry *obj = *list++;
|
|
|
|
uint32_t crc32_val = htonl(obj->crc32);
|
|
|
|
sha1write(f, &crc32_val, 4);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* write the 32-bit offset table */
|
|
|
|
list = sorted_by_sha;
|
|
|
|
for (i = 0; i < nr_objects; i++) {
|
|
|
|
struct pack_idx_entry *obj = *list++;
|
2011-02-26 01:54:00 +01:00
|
|
|
uint32_t offset;
|
|
|
|
|
|
|
|
offset = (need_large_offset(obj->offset, opts)
|
|
|
|
? (0x80000000 | nr_large_offset++)
|
|
|
|
: obj->offset);
|
2007-06-01 21:18:05 +02:00
|
|
|
offset = htonl(offset);
|
|
|
|
sha1write(f, &offset, 4);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* write the large offset table */
|
|
|
|
list = sorted_by_sha;
|
|
|
|
while (nr_large_offset) {
|
|
|
|
struct pack_idx_entry *obj = *list++;
|
|
|
|
uint64_t offset = obj->offset;
|
2011-02-26 01:54:00 +01:00
|
|
|
uint32_t split[2];
|
|
|
|
|
|
|
|
if (!need_large_offset(offset, opts))
|
|
|
|
continue;
|
|
|
|
split[0] = htonl(offset >> 32);
|
|
|
|
split[1] = htonl(offset & 0xffffffff);
|
|
|
|
sha1write(f, split, 8);
|
|
|
|
nr_large_offset--;
|
2007-06-01 21:18:05 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
sha1write(f, sha1, 20);
|
2011-02-03 02:29:01 +01:00
|
|
|
sha1close(f, NULL, ((opts->flags & WRITE_IDX_VERIFY)
|
|
|
|
? CSUM_CLOSE : CSUM_FSYNC));
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Final(sha1, &ctx);
|
2007-06-01 21:18:05 +02:00
|
|
|
return index_name;
|
|
|
|
}
|
2007-05-02 18:13:14 +02:00
|
|
|
|
2011-10-28 20:40:48 +02:00
|
|
|
off_t write_pack_header(struct sha1file *f, uint32_t nr_entries)
|
|
|
|
{
|
|
|
|
struct pack_header hdr;
|
|
|
|
|
|
|
|
hdr.hdr_signature = htonl(PACK_SIGNATURE);
|
|
|
|
hdr.hdr_version = htonl(PACK_VERSION);
|
|
|
|
hdr.hdr_entries = htonl(nr_entries);
|
|
|
|
if (sha1write(f, &hdr, sizeof(hdr)))
|
|
|
|
return 0;
|
|
|
|
return sizeof(hdr);
|
|
|
|
}
|
|
|
|
|
2008-08-29 22:07:59 +02:00
|
|
|
/*
|
|
|
|
* Update pack header with object_count and compute new SHA1 for pack data
|
|
|
|
* associated to pack_fd, and write that SHA1 at the end. That new SHA1
|
|
|
|
* is also returned in new_pack_sha1.
|
|
|
|
*
|
|
|
|
* If partial_pack_sha1 is non null, then the SHA1 of the existing pack
|
|
|
|
* (without the header update) is computed and validated against the
|
|
|
|
* one provided in partial_pack_sha1. The validation is performed at
|
|
|
|
* partial_pack_offset bytes in the pack file. The SHA1 of the remaining
|
|
|
|
* data (i.e. from partial_pack_offset to the end) is then computed and
|
|
|
|
* returned in partial_pack_sha1.
|
|
|
|
*
|
|
|
|
* Note that new_pack_sha1 is updated last, so both new_pack_sha1 and
|
|
|
|
* partial_pack_sha1 can refer to the same buffer if the caller is not
|
|
|
|
* interested in the resulting SHA1 of pack data above partial_pack_offset.
|
|
|
|
*/
|
2007-05-02 18:13:14 +02:00
|
|
|
void fixup_pack_header_footer(int pack_fd,
|
2008-08-29 22:07:59 +02:00
|
|
|
unsigned char *new_pack_sha1,
|
2007-05-02 18:13:14 +02:00
|
|
|
const char *pack_name,
|
2008-08-29 22:07:59 +02:00
|
|
|
uint32_t object_count,
|
|
|
|
unsigned char *partial_pack_sha1,
|
|
|
|
off_t partial_pack_offset)
|
2007-05-02 18:13:14 +02:00
|
|
|
{
|
2008-08-29 22:08:02 +02:00
|
|
|
int aligned_sz, buf_sz = 8 * 1024;
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA_CTX old_sha1_ctx, new_sha1_ctx;
|
2007-05-02 18:13:14 +02:00
|
|
|
struct pack_header hdr;
|
|
|
|
char *buf;
|
|
|
|
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Init(&old_sha1_ctx);
|
|
|
|
git_SHA1_Init(&new_sha1_ctx);
|
2008-08-29 22:07:59 +02:00
|
|
|
|
2007-05-02 18:13:14 +02:00
|
|
|
if (lseek(pack_fd, 0, SEEK_SET) != 0)
|
2009-06-27 17:58:46 +02:00
|
|
|
die_errno("Failed seeking to start of '%s'", pack_name);
|
2007-05-02 18:13:14 +02:00
|
|
|
if (read_in_full(pack_fd, &hdr, sizeof(hdr)) != sizeof(hdr))
|
2009-06-27 17:58:46 +02:00
|
|
|
die_errno("Unable to reread header of '%s'", pack_name);
|
2007-05-02 18:13:14 +02:00
|
|
|
if (lseek(pack_fd, 0, SEEK_SET) != 0)
|
2009-06-27 17:58:46 +02:00
|
|
|
die_errno("Failed seeking to start of '%s'", pack_name);
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Update(&old_sha1_ctx, &hdr, sizeof(hdr));
|
2007-05-02 18:13:14 +02:00
|
|
|
hdr.hdr_entries = htonl(object_count);
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Update(&new_sha1_ctx, &hdr, sizeof(hdr));
|
2007-05-02 18:13:14 +02:00
|
|
|
write_or_die(pack_fd, &hdr, sizeof(hdr));
|
2008-08-29 22:07:59 +02:00
|
|
|
partial_pack_offset -= sizeof(hdr);
|
2007-05-02 18:13:14 +02:00
|
|
|
|
|
|
|
buf = xmalloc(buf_sz);
|
2008-08-29 22:08:02 +02:00
|
|
|
aligned_sz = buf_sz - sizeof(hdr);
|
2007-05-02 18:13:14 +02:00
|
|
|
for (;;) {
|
2008-08-29 22:07:59 +02:00
|
|
|
ssize_t m, n;
|
2008-08-29 22:08:02 +02:00
|
|
|
m = (partial_pack_sha1 && partial_pack_offset < aligned_sz) ?
|
|
|
|
partial_pack_offset : aligned_sz;
|
2008-08-29 22:07:59 +02:00
|
|
|
n = xread(pack_fd, buf, m);
|
2007-05-02 18:13:14 +02:00
|
|
|
if (!n)
|
|
|
|
break;
|
|
|
|
if (n < 0)
|
2009-06-27 17:58:46 +02:00
|
|
|
die_errno("Failed to checksum '%s'", pack_name);
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Update(&new_sha1_ctx, buf, n);
|
2008-08-29 22:07:59 +02:00
|
|
|
|
2008-08-29 22:08:02 +02:00
|
|
|
aligned_sz -= n;
|
|
|
|
if (!aligned_sz)
|
|
|
|
aligned_sz = buf_sz;
|
|
|
|
|
2008-08-29 22:07:59 +02:00
|
|
|
if (!partial_pack_sha1)
|
|
|
|
continue;
|
|
|
|
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Update(&old_sha1_ctx, buf, n);
|
2008-08-29 22:07:59 +02:00
|
|
|
partial_pack_offset -= n;
|
|
|
|
if (partial_pack_offset == 0) {
|
|
|
|
unsigned char sha1[20];
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Final(sha1, &old_sha1_ctx);
|
2008-08-29 22:07:59 +02:00
|
|
|
if (hashcmp(sha1, partial_pack_sha1) != 0)
|
|
|
|
die("Unexpected checksum for %s "
|
|
|
|
"(disk corruption?)", pack_name);
|
|
|
|
/*
|
|
|
|
* Now let's compute the SHA1 of the remainder of the
|
|
|
|
* pack, which also means making partial_pack_offset
|
|
|
|
* big enough not to matter anymore.
|
|
|
|
*/
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Init(&old_sha1_ctx);
|
2008-08-29 22:07:59 +02:00
|
|
|
partial_pack_offset = ~partial_pack_offset;
|
|
|
|
partial_pack_offset -= MSB(partial_pack_offset, 1);
|
|
|
|
}
|
2007-05-02 18:13:14 +02:00
|
|
|
}
|
|
|
|
free(buf);
|
|
|
|
|
2008-08-29 22:07:59 +02:00
|
|
|
if (partial_pack_sha1)
|
2008-10-01 20:05:20 +02:00
|
|
|
git_SHA1_Final(partial_pack_sha1, &old_sha1_ctx);
|
|
|
|
git_SHA1_Final(new_pack_sha1, &new_sha1_ctx);
|
2008-08-29 22:07:59 +02:00
|
|
|
write_or_die(pack_fd, new_pack_sha1, 20);
|
2008-08-27 21:48:00 +02:00
|
|
|
fsync_or_die(pack_fd, pack_name);
|
2007-05-02 18:13:14 +02:00
|
|
|
}
|
2007-09-14 09:31:16 +02:00
|
|
|
|
|
|
|
char *index_pack_lockfile(int ip_out)
|
|
|
|
{
|
|
|
|
char packname[46];
|
|
|
|
|
|
|
|
/*
|
2009-02-25 08:11:29 +01:00
|
|
|
* The first thing we expect from index-pack's output
|
2007-09-14 09:31:16 +02:00
|
|
|
* is "pack\t%40s\n" or "keep\t%40s\n" (46 bytes) where
|
|
|
|
* %40s is the newly created pack SHA1 name. In the "keep"
|
|
|
|
* case, we need it to remove the corresponding .keep file
|
|
|
|
* later on. If we don't get that then tough luck with it.
|
|
|
|
*/
|
2008-05-03 15:27:26 +02:00
|
|
|
if (read_in_full(ip_out, packname, 46) == 46 && packname[45] == '\n' &&
|
|
|
|
memcmp(packname, "keep\t", 5) == 0) {
|
2007-09-14 09:31:16 +02:00
|
|
|
char path[PATH_MAX];
|
|
|
|
packname[45] = 0;
|
|
|
|
snprintf(path, sizeof(path), "%s/pack/pack-%s.keep",
|
|
|
|
get_object_directory(), packname + 5);
|
|
|
|
return xstrdup(path);
|
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
2010-02-23 21:02:37 +01:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The per-object header is a pretty dense thing, which is
|
|
|
|
* - first byte: low four bits are "size", then three bits of "type",
|
|
|
|
* and the high bit is "size continues".
|
|
|
|
* - each byte afterwards: low seven bits are size continuation,
|
|
|
|
* with the high bit being "size continues"
|
|
|
|
*/
|
|
|
|
int encode_in_pack_object_header(enum object_type type, uintmax_t size, unsigned char *hdr)
|
|
|
|
{
|
|
|
|
int n = 1;
|
|
|
|
unsigned char c;
|
|
|
|
|
|
|
|
if (type < OBJ_COMMIT || type > OBJ_REF_DELTA)
|
|
|
|
die("bad type %d", type);
|
|
|
|
|
|
|
|
c = (type << 4) | (size & 15);
|
|
|
|
size >>= 4;
|
|
|
|
while (size) {
|
|
|
|
*hdr++ = c | 0x80;
|
|
|
|
c = size & 0x7f;
|
|
|
|
size >>= 7;
|
|
|
|
n++;
|
|
|
|
}
|
|
|
|
*hdr = c;
|
|
|
|
return n;
|
|
|
|
}
|
2011-10-28 20:52:14 +02:00
|
|
|
|
|
|
|
struct sha1file *create_tmp_packfile(char **pack_tmp_name)
|
|
|
|
{
|
|
|
|
char tmpname[PATH_MAX];
|
|
|
|
int fd;
|
|
|
|
|
|
|
|
fd = odb_mkstemp(tmpname, sizeof(tmpname), "pack/tmp_pack_XXXXXX");
|
|
|
|
*pack_tmp_name = xstrdup(tmpname);
|
|
|
|
return sha1fd(fd, *pack_tmp_name);
|
|
|
|
}
|
2011-10-28 21:34:09 +02:00
|
|
|
|
|
|
|
void finish_tmp_packfile(char *name_buffer,
|
|
|
|
const char *pack_tmp_name,
|
|
|
|
struct pack_idx_entry **written_list,
|
|
|
|
uint32_t nr_written,
|
|
|
|
struct pack_idx_option *pack_idx_opts,
|
|
|
|
unsigned char sha1[])
|
|
|
|
{
|
|
|
|
const char *idx_tmp_name;
|
|
|
|
char *end_of_name_prefix = strrchr(name_buffer, 0);
|
|
|
|
|
|
|
|
if (adjust_shared_perm(pack_tmp_name))
|
|
|
|
die_errno("unable to make temporary pack file readable");
|
|
|
|
|
|
|
|
idx_tmp_name = write_idx_file(NULL, written_list, nr_written,
|
|
|
|
pack_idx_opts, sha1);
|
|
|
|
if (adjust_shared_perm(idx_tmp_name))
|
|
|
|
die_errno("unable to make temporary index file readable");
|
|
|
|
|
|
|
|
sprintf(end_of_name_prefix, "%s.pack", sha1_to_hex(sha1));
|
|
|
|
free_pack_by_name(name_buffer);
|
|
|
|
|
|
|
|
if (rename(pack_tmp_name, name_buffer))
|
|
|
|
die_errno("unable to rename temporary pack file");
|
|
|
|
|
|
|
|
sprintf(end_of_name_prefix, "%s.idx", sha1_to_hex(sha1));
|
|
|
|
if (rename(idx_tmp_name, name_buffer))
|
|
|
|
die_errno("unable to rename temporary index file");
|
|
|
|
|
|
|
|
free((void *)idx_tmp_name);
|
|
|
|
}
|