diff options
author | Alexandre Julliard <julliard@winehq.org> | 2007-06-03 20:21:41 +0200 |
---|---|---|
committer | Junio C Hamano <gitster@pobox.com> | 2007-06-06 16:04:02 -0700 |
commit | 3af51928ab7cefa35048e3a5a9e78a127749d405 (patch) | |
tree | 1f0f14e68a5f6c95237b6ed5d94c767a2341e95f | |
parent | d44c782bbd6b0e806e056f9e8ff8cd8e426e67a3 (diff) | |
download | git-3af51928ab7cefa35048e3a5a9e78a127749d405.tar.gz git-3af51928ab7cefa35048e3a5a9e78a127749d405.tar.xz |
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>
-rw-r--r-- | pack-check.c | 47 |
1 files changed, 35 insertions, 12 deletions
diff --git a/pack-check.c b/pack-check.c index 3623c716e..d7dd62bb8 100644 --- a/pack-check.c +++ b/pack-check.c @@ -1,6 +1,23 @@ #include "cache.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, 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; uint32_t nr_objects, i; int err; + struct idx_entry *entries; /* Note that the pack header checks are actually performed by * 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. */ 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++) { - const unsigned char *sha1; void *data; enum object_type type; unsigned long size; - off_t offset; - sha1 = nth_packed_object_sha1(p, i); - 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); + data = unpack_entry(p, entries[i].offset, &type, &size); if (!data) { err = error("cannot unpack %s from %s", - sha1_to_hex(sha1), p->pack_name); + sha1_to_hex(entries[i].sha1), p->pack_name); 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", - sha1_to_hex(sha1), p->pack_name); + sha1_to_hex(entries[i].sha1), p->pack_name); free(data); continue; } free(data); } + free(entries); return err; } |