kernel
/
git
mirror of https://git.kernel.org/pub/scm/git/git.git
1
0
Fork 0
Code Releases Activity

pack-check: Sort entries by pack offset before unpacking them. 

Because of the way objects are sorted in a pack, unpacking them in
disk order is much more efficient than random access. Tests on the
Wine repository show a gain in pack validation time of about 35%.

Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
maint
Alexandre Julliard 2007-06-03 20:21:41 +02:00 committed by Junio C Hamano
parent d44c782bbd
commit 3af51928ab
1 changed files with 35 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

47
pack-check.c
View File

@ -1,6 +1,23 @@
#include "cache.h" #include "cache.h"
#include "pack.h" #include "pack.h"


struct idx_entry
{
const unsigned char *sha1;
off_t offset;
};

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;
}

static int verify_packfile(struct packed_git *p, static int verify_packfile(struct packed_git *p,
struct pack_window **w_curs) struct pack_window **w_curs)
{ {
@ -11,6 +28,7 @@ static int verify_packfile(struct packed_git *p,
off_t offset = 0, pack_sig = p->pack_size - 20; off_t offset = 0, pack_sig = p->pack_size - 20;
uint32_t nr_objects, i; uint32_t nr_objects, i;
int err; int err;
struct idx_entry *entries;


/* Note that the pack header checks are actually performed by /* Note that the pack header checks are actually performed by
* use_pack when it first opens the pack file. If anything * use_pack when it first opens the pack file. If anything
@ -41,33 +59,38 @@ static int verify_packfile(struct packed_git *p,
* we do not do scan-streaming check on the pack file. * we do not do scan-streaming check on the pack file.
*/ */
nr_objects = p->num_objects; nr_objects = p->num_objects;
entries = xmalloc(nr_objects * sizeof(*entries));
/* first sort entries by pack offset, since unpacking them is more efficient that way */
for (i = 0; i < nr_objects; i++) {
entries[i].sha1 = nth_packed_object_sha1(p, i);
if (!entries[i].sha1)
die("internal error pack-check nth-packed-object");
entries[i].offset = find_pack_entry_one(entries[i].sha1, p);
if (!entries[i].offset)
die("internal error pack-check find-pack-entry-one");
}
qsort(entries, nr_objects, sizeof(*entries), compare_entries);

for (i = 0, err = 0; i < nr_objects; i++) { for (i = 0, err = 0; i < nr_objects; i++) {
const unsigned char *sha1;
void *data; void *data;
enum object_type type; enum object_type type;
unsigned long size; unsigned long size;
off_t offset;


sha1 = nth_packed_object_sha1(p, i); data = unpack_entry(p, entries[i].offset, &type, &size);
if (!sha1)
die("internal error pack-check nth-packed-object");
offset = find_pack_entry_one(sha1, p);
if (!offset)
die("internal error pack-check find-pack-entry-one");
data = unpack_entry(p, offset, &type, &size);
if (!data) { if (!data) {
err = error("cannot unpack %s from %s", err = error("cannot unpack %s from %s",
sha1_to_hex(sha1), p->pack_name); sha1_to_hex(entries[i].sha1), p->pack_name);
continue; continue;
} }
if (check_sha1_signature(sha1, data, size, typename(type))) { if (check_sha1_signature(entries[i].sha1, data, size, typename(type))) {
err = error("packed %s from %s is corrupt", err = error("packed %s from %s is corrupt",
sha1_to_hex(sha1), p->pack_name); sha1_to_hex(entries[i].sha1), p->pack_name);
free(data); free(data);
continue; continue;
} }
free(data); free(data);
} }
free(entries);


return err; return err;
} }