git/pack-check.c

203 строки
5.5 KiB
C
Исходник Обычный вид История

#include "cache.h"
#include "repository.h"
#include "pack.h"
#include "pack-revindex.h"
#include "progress.h"
#include "packfile.h"
#include "object-store.h"
struct idx_entry {
off_t offset;
unsigned int nr;
};
static int compare_entries(const void *e1, const void *e2)
{
const struct idx_entry *entry1 = e1;
const struct idx_entry *entry2 = e2;
if (entry1->offset < entry2->offset)
return -1;
if (entry1->offset > entry2->offset)
return 1;
return 0;
}
int check_pack_crc(struct packed_git *p, struct pack_window **w_curs,
off_t offset, off_t len, unsigned int nr)
{
const uint32_t *index_crc;
uint32_t data_crc = crc32(0, NULL, 0);
do {
2011-06-10 22:52:15 +04:00
unsigned long avail;
void *data = use_pack(p, w_curs, offset, &avail);
if (avail > len)
avail = len;
data_crc = crc32(data_crc, data, avail);
offset += avail;
len -= avail;
} while (len);
index_crc = p->index_data;
index_crc += 2 + 256 + p->num_objects * (the_hash_algo->rawsz/4) + nr;
return data_crc != ntohl(*index_crc);
}
static int verify_packfile(struct repository *r,
struct packed_git *p,
struct pack_window **w_curs,
verify_fn fn,
struct progress *progress, uint32_t base_count)
{
off_t index_size = p->index_size;
const unsigned char *index_base = p->index_data;
git_hash_ctx ctx;
unsigned char hash[GIT_MAX_RAWSZ], *pack_sig;
off_t offset = 0, pack_sig_ofs = 0;
uint32_t nr_objects, i;
int err = 0;
struct idx_entry *entries;
if (!is_pack_valid(p))
return error("packfile %s cannot be accessed", p->pack_name);
r->hash_algo->init_fn(&ctx);
do {
2011-06-10 22:52:15 +04:00
unsigned long remaining;
unsigned char *in = use_pack(p, w_curs, offset, &remaining);
offset += remaining;
if (!pack_sig_ofs)
pack_sig_ofs = p->pack_size - r->hash_algo->rawsz;
if (offset > pack_sig_ofs)
remaining -= (unsigned int)(offset - pack_sig_ofs);
r->hash_algo->update_fn(&ctx, in, remaining);
} while (offset < pack_sig_ofs);
r->hash_algo->final_fn(hash, &ctx);
pack_sig = use_pack(p, w_curs, pack_sig_ofs, NULL);
if (!hasheq(hash, pack_sig))
err = error("%s pack checksum mismatch",
p->pack_name);
if (!hasheq(index_base + index_size - r->hash_algo->hexsz, pack_sig))
err = error("%s pack checksum does not match its index",
p->pack_name);
unuse_pack(w_curs);
/* Make sure everything reachable from idx is valid. Since we
* have verified that nr_objects matches between idx and pack,
* we do not do scan-streaming check on the pack file.
*/
nr_objects = p->num_objects;
ALLOC_ARRAY(entries, nr_objects + 1);
entries[nr_objects].offset = pack_sig_ofs;
/* first sort entries by pack offset, since unpacking them is more efficient that way */
for (i = 0; i < nr_objects; i++) {
entries[i].offset = nth_packed_object_offset(p, i);
entries[i].nr = i;
}
QSORT(entries, nr_objects, compare_entries);
for (i = 0; i < nr_objects; i++) {
void *data;
pack-check: push oid lookup into loop When we're checking a pack with fsck or verify-pack, we first sort the idx entries by offset, since accessing them in pack order is more efficient. To do so, we loop over them and fill in an array of structs with the offset, object_id, and index position of each, sort the result, and only then do we iterate over the sorted array and process each entry. In order to avoid the memory cost of storing the hash of each object, we just store a pointer into the copy in the mmap'd pack index file. To keep that property even as the rest of the code converted to "struct object_id", commit 9fd750461b (Convert the verify_pack callback to struct object_id, 2017-05-06) introduced a union in order to type-pun the pointer-to-hash into an object_id struct. But we can make this even simpler by observing that the sort operation doesn't need the object id at all! We only need them one at a time while we actually process each entry. So we can just omit the oid from the struct entirely and load it on the fly into a local variable in the second loop. This gets rid of the type-punning, and lets us directly use the more type-safe nth_packed_object_id(), simplifying the code. And as a bonus, it saves 8 bytes of memory per object. Note that this does mean we'll do the offset lookup for each object before the oid lookup. The oid lookup has more safety checks in it (e.g., for looking past p->num_objects) which in theory protected the offset lookup. But since violating those checks was already a BUG() condition (as described in the previous commit), it's not worth worrying about. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-02-24 07:36:31 +03:00
struct object_id oid;
enum object_type type;
unsigned long size;
off_t curpos;
int data_valid;
pack-check: push oid lookup into loop When we're checking a pack with fsck or verify-pack, we first sort the idx entries by offset, since accessing them in pack order is more efficient. To do so, we loop over them and fill in an array of structs with the offset, object_id, and index position of each, sort the result, and only then do we iterate over the sorted array and process each entry. In order to avoid the memory cost of storing the hash of each object, we just store a pointer into the copy in the mmap'd pack index file. To keep that property even as the rest of the code converted to "struct object_id", commit 9fd750461b (Convert the verify_pack callback to struct object_id, 2017-05-06) introduced a union in order to type-pun the pointer-to-hash into an object_id struct. But we can make this even simpler by observing that the sort operation doesn't need the object id at all! We only need them one at a time while we actually process each entry. So we can just omit the oid from the struct entirely and load it on the fly into a local variable in the second loop. This gets rid of the type-punning, and lets us directly use the more type-safe nth_packed_object_id(), simplifying the code. And as a bonus, it saves 8 bytes of memory per object. Note that this does mean we'll do the offset lookup for each object before the oid lookup. The oid lookup has more safety checks in it (e.g., for looking past p->num_objects) which in theory protected the offset lookup. But since violating those checks was already a BUG() condition (as described in the previous commit), it's not worth worrying about. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-02-24 07:36:31 +03:00
if (nth_packed_object_id(&oid, p, entries[i].nr) < 0)
BUG("unable to get oid of object %lu from %s",
(unsigned long)entries[i].nr, p->pack_name);
if (p->index_version > 1) {
off_t offset = entries[i].offset;
off_t len = entries[i+1].offset - offset;
unsigned int nr = entries[i].nr;
if (check_pack_crc(p, w_curs, offset, len, nr))
err = error("index CRC mismatch for object %s "
"from %s at offset %"PRIuMAX"",
pack-check: push oid lookup into loop When we're checking a pack with fsck or verify-pack, we first sort the idx entries by offset, since accessing them in pack order is more efficient. To do so, we loop over them and fill in an array of structs with the offset, object_id, and index position of each, sort the result, and only then do we iterate over the sorted array and process each entry. In order to avoid the memory cost of storing the hash of each object, we just store a pointer into the copy in the mmap'd pack index file. To keep that property even as the rest of the code converted to "struct object_id", commit 9fd750461b (Convert the verify_pack callback to struct object_id, 2017-05-06) introduced a union in order to type-pun the pointer-to-hash into an object_id struct. But we can make this even simpler by observing that the sort operation doesn't need the object id at all! We only need them one at a time while we actually process each entry. So we can just omit the oid from the struct entirely and load it on the fly into a local variable in the second loop. This gets rid of the type-punning, and lets us directly use the more type-safe nth_packed_object_id(), simplifying the code. And as a bonus, it saves 8 bytes of memory per object. Note that this does mean we'll do the offset lookup for each object before the oid lookup. The oid lookup has more safety checks in it (e.g., for looking past p->num_objects) which in theory protected the offset lookup. But since violating those checks was already a BUG() condition (as described in the previous commit), it's not worth worrying about. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-02-24 07:36:31 +03:00
oid_to_hex(&oid),
p->pack_name, (uintmax_t)offset);
}
curpos = entries[i].offset;
type = unpack_object_header(p, w_curs, &curpos, &size);
unuse_pack(w_curs);
if (type == OBJ_BLOB && big_file_threshold <= size) {
/*
* Let check_object_signature() check it with
* the streaming interface; no point slurping
* the data in-core only to discard.
*/
data = NULL;
data_valid = 0;
} else {
data = unpack_entry(r, p, entries[i].offset, &type, &size);
data_valid = 1;
}
if (data_valid && !data)
err = error("cannot unpack %s from %s at offset %"PRIuMAX"",
pack-check: push oid lookup into loop When we're checking a pack with fsck or verify-pack, we first sort the idx entries by offset, since accessing them in pack order is more efficient. To do so, we loop over them and fill in an array of structs with the offset, object_id, and index position of each, sort the result, and only then do we iterate over the sorted array and process each entry. In order to avoid the memory cost of storing the hash of each object, we just store a pointer into the copy in the mmap'd pack index file. To keep that property even as the rest of the code converted to "struct object_id", commit 9fd750461b (Convert the verify_pack callback to struct object_id, 2017-05-06) introduced a union in order to type-pun the pointer-to-hash into an object_id struct. But we can make this even simpler by observing that the sort operation doesn't need the object id at all! We only need them one at a time while we actually process each entry. So we can just omit the oid from the struct entirely and load it on the fly into a local variable in the second loop. This gets rid of the type-punning, and lets us directly use the more type-safe nth_packed_object_id(), simplifying the code. And as a bonus, it saves 8 bytes of memory per object. Note that this does mean we'll do the offset lookup for each object before the oid lookup. The oid lookup has more safety checks in it (e.g., for looking past p->num_objects) which in theory protected the offset lookup. But since violating those checks was already a BUG() condition (as described in the previous commit), it's not worth worrying about. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-02-24 07:36:31 +03:00
oid_to_hex(&oid), p->pack_name,
(uintmax_t)entries[i].offset);
pack-check: push oid lookup into loop When we're checking a pack with fsck or verify-pack, we first sort the idx entries by offset, since accessing them in pack order is more efficient. To do so, we loop over them and fill in an array of structs with the offset, object_id, and index position of each, sort the result, and only then do we iterate over the sorted array and process each entry. In order to avoid the memory cost of storing the hash of each object, we just store a pointer into the copy in the mmap'd pack index file. To keep that property even as the rest of the code converted to "struct object_id", commit 9fd750461b (Convert the verify_pack callback to struct object_id, 2017-05-06) introduced a union in order to type-pun the pointer-to-hash into an object_id struct. But we can make this even simpler by observing that the sort operation doesn't need the object id at all! We only need them one at a time while we actually process each entry. So we can just omit the oid from the struct entirely and load it on the fly into a local variable in the second loop. This gets rid of the type-punning, and lets us directly use the more type-safe nth_packed_object_id(), simplifying the code. And as a bonus, it saves 8 bytes of memory per object. Note that this does mean we'll do the offset lookup for each object before the oid lookup. The oid lookup has more safety checks in it (e.g., for looking past p->num_objects) which in theory protected the offset lookup. But since violating those checks was already a BUG() condition (as described in the previous commit), it's not worth worrying about. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-02-24 07:36:31 +03:00
else if (check_object_signature(r, &oid, data, size, type_name(type)))
err = error("packed %s from %s is corrupt",
pack-check: push oid lookup into loop When we're checking a pack with fsck or verify-pack, we first sort the idx entries by offset, since accessing them in pack order is more efficient. To do so, we loop over them and fill in an array of structs with the offset, object_id, and index position of each, sort the result, and only then do we iterate over the sorted array and process each entry. In order to avoid the memory cost of storing the hash of each object, we just store a pointer into the copy in the mmap'd pack index file. To keep that property even as the rest of the code converted to "struct object_id", commit 9fd750461b (Convert the verify_pack callback to struct object_id, 2017-05-06) introduced a union in order to type-pun the pointer-to-hash into an object_id struct. But we can make this even simpler by observing that the sort operation doesn't need the object id at all! We only need them one at a time while we actually process each entry. So we can just omit the oid from the struct entirely and load it on the fly into a local variable in the second loop. This gets rid of the type-punning, and lets us directly use the more type-safe nth_packed_object_id(), simplifying the code. And as a bonus, it saves 8 bytes of memory per object. Note that this does mean we'll do the offset lookup for each object before the oid lookup. The oid lookup has more safety checks in it (e.g., for looking past p->num_objects) which in theory protected the offset lookup. But since violating those checks was already a BUG() condition (as described in the previous commit), it's not worth worrying about. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-02-24 07:36:31 +03:00
oid_to_hex(&oid), p->pack_name);
else if (fn) {
int eaten = 0;
pack-check: push oid lookup into loop When we're checking a pack with fsck or verify-pack, we first sort the idx entries by offset, since accessing them in pack order is more efficient. To do so, we loop over them and fill in an array of structs with the offset, object_id, and index position of each, sort the result, and only then do we iterate over the sorted array and process each entry. In order to avoid the memory cost of storing the hash of each object, we just store a pointer into the copy in the mmap'd pack index file. To keep that property even as the rest of the code converted to "struct object_id", commit 9fd750461b (Convert the verify_pack callback to struct object_id, 2017-05-06) introduced a union in order to type-pun the pointer-to-hash into an object_id struct. But we can make this even simpler by observing that the sort operation doesn't need the object id at all! We only need them one at a time while we actually process each entry. So we can just omit the oid from the struct entirely and load it on the fly into a local variable in the second loop. This gets rid of the type-punning, and lets us directly use the more type-safe nth_packed_object_id(), simplifying the code. And as a bonus, it saves 8 bytes of memory per object. Note that this does mean we'll do the offset lookup for each object before the oid lookup. The oid lookup has more safety checks in it (e.g., for looking past p->num_objects) which in theory protected the offset lookup. But since violating those checks was already a BUG() condition (as described in the previous commit), it's not worth worrying about. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-02-24 07:36:31 +03:00
err |= fn(&oid, type, size, data, &eaten);
if (eaten)
data = NULL;
}
if (((base_count + i) & 1023) == 0)
display_progress(progress, base_count + i);
free(data);
}
display_progress(progress, base_count + i);
free(entries);
return err;
}
int verify_pack_index(struct packed_git *p)
{
off_t index_size;
const unsigned char *index_base;
git_hash_ctx ctx;
unsigned char hash[GIT_MAX_RAWSZ];
int err = 0;
if (open_pack_index(p))
return error("packfile %s index not opened", p->pack_name);
index_size = p->index_size;
index_base = p->index_data;
/* Verify SHA1 sum of the index file */
the_hash_algo->init_fn(&ctx);
the_hash_algo->update_fn(&ctx, index_base, (unsigned int)(index_size - the_hash_algo->rawsz));
the_hash_algo->final_fn(hash, &ctx);
if (!hasheq(hash, index_base + index_size - the_hash_algo->rawsz))
err = error("Packfile index for %s hash mismatch",
p->pack_name);
return err;
}
int verify_pack(struct repository *r, struct packed_git *p, verify_fn fn,
struct progress *progress, uint32_t base_count)
{
int err = 0;
struct pack_window *w_curs = NULL;
err |= verify_pack_index(p);
if (!p->index_data)
return -1;
err |= verify_packfile(r, p, &w_curs, fn, progress, base_count);
unuse_pack(&w_curs);
return err;
}