git/tmp-objdir.c
Taylor Blau b1b8dfde69 finalize_object_file(): implement collision check
We've had "FIXME!!! Collision check here ?" in finalize_object_file()
since aac1794132 (Improve sha1 object file writing., 2005-05-03). That
is, when we try to write a file with the same name, we assume the
on-disk contents are the same and blindly throw away the new copy.

One of the reasons we never implemented this is because the files it
moves are all named after the cryptographic hash of their contents
(either loose objects, or packs which have their hash in the name these
days). So we are unlikely to see such a collision by accident. And even
though there are weaknesses in sha1, we assume they are mitigated by our
use of sha1dc.

So while it's a theoretical concern now, it hasn't been a priority.
However, if we start using weaker hashes for pack checksums and names,
this will become a practical concern. So in preparation, let's actually
implement a byte-for-byte collision check.

The new check will cause the write of new differing content to be a
failure, rather than a silent noop, and we'll retain the temporary file
on disk. If there's no collision present, we'll clean up the temporary
file as usual after either rename()-ing or link()-ing it into place.

Note that this may cause some extra computation when the files are in
fact identical, but this should happen rarely.

Loose objects are exempt from this check, and the collision check may be
skipped by calling the _flags variant of this function with the
FOF_SKIP_COLLISION_CHECK bit set. This is done for a couple of reasons:

  - We don't treat the hash of the loose object file's contents as a
    checksum, since the same loose object can be stored using different
    bytes on disk (e.g., when adjusting core.compression, using a
    different version of zlib, etc.).

    This is fundamentally different from cases where
    finalize_object_file() is operating over a file which uses the hash
    value as a checksum of the contents. In other words, a pair of
    identical loose objects can be stored using different bytes on disk,
    and that should not be treated as a collision.

  - We already use the path of the loose object as its hash value /
    object name, so checking for collisions at the content level doesn't
    add anything.

    Adding a content-level collision check would have to happen at a
    higher level than in finalize_object_file(), since (avoiding race
    conditions) writing an object loose which already exists in the
    repository will prevent us from even reaching finalize_object_file()
    via the object freshening code.

    There is a collision check in index-pack via its `check_collision()`
    function, but there isn't an analogous function in unpack-objects,
    which just feeds the result to write_object_file().

    So skipping the collision check here does not change for better or
    worse the hardness of loose object writes.

As a small note related to the latter bullet point above, we must teach
the tmp-objdir routines to similarly skip the content-level collision
checks when calling migrate_one() on a loose object file, which we do by
setting the FOF_SKIP_COLLISION_CHECK bit when we are inside of a loose
object shard.

Co-authored-by: Jeff King <peff@peff.net>
Signed-off-by: Jeff King <peff@peff.net>
Helped-by: Elijah Newren <newren@gmail.com>
Signed-off-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
2024-09-27 11:27:47 -07:00

334 lines
7.6 KiB
C

#define USE_THE_REPOSITORY_VARIABLE
#include "git-compat-util.h"
#include "tmp-objdir.h"
#include "abspath.h"
#include "chdir-notify.h"
#include "dir.h"
#include "environment.h"
#include "object-file.h"
#include "path.h"
#include "string-list.h"
#include "strbuf.h"
#include "strvec.h"
#include "quote.h"
#include "object-store-ll.h"
struct tmp_objdir {
struct strbuf path;
struct strvec env;
struct object_directory *prev_odb;
int will_destroy;
};
/*
* Allow only one tmp_objdir at a time in a running process, which simplifies
* our atexit cleanup routines. It's doubtful callers will ever need
* more than one, and we can expand later if so. You can have many such
* tmp_objdirs simultaneously in many processes, of course.
*/
static struct tmp_objdir *the_tmp_objdir;
static void tmp_objdir_free(struct tmp_objdir *t)
{
strbuf_release(&t->path);
strvec_clear(&t->env);
free(t);
}
int tmp_objdir_destroy(struct tmp_objdir *t)
{
int err;
if (!t)
return 0;
if (t == the_tmp_objdir)
the_tmp_objdir = NULL;
if (t->prev_odb)
restore_primary_odb(t->prev_odb, t->path.buf);
err = remove_dir_recursively(&t->path, 0);
tmp_objdir_free(t);
return err;
}
static void remove_tmp_objdir(void)
{
tmp_objdir_destroy(the_tmp_objdir);
}
void tmp_objdir_discard_objects(struct tmp_objdir *t)
{
remove_dir_recursively(&t->path, REMOVE_DIR_KEEP_TOPLEVEL);
}
/*
* These env_* functions are for setting up the child environment; the
* "replace" variant overrides the value of any existing variable with that
* "key". The "append" variant puts our new value at the end of a list,
* separated by PATH_SEP (which is what separate values in
* GIT_ALTERNATE_OBJECT_DIRECTORIES).
*/
static void env_append(struct strvec *env, const char *key, const char *val)
{
struct strbuf quoted = STRBUF_INIT;
const char *old;
/*
* Avoid quoting if it's not necessary, for maximum compatibility
* with older parsers which don't understand the quoting.
*/
if (*val == '"' || strchr(val, PATH_SEP)) {
strbuf_addch(&quoted, '"');
quote_c_style(val, &quoted, NULL, 1);
strbuf_addch(&quoted, '"');
val = quoted.buf;
}
old = getenv(key);
if (!old)
strvec_pushf(env, "%s=%s", key, val);
else
strvec_pushf(env, "%s=%s%c%s", key, old, PATH_SEP, val);
strbuf_release(&quoted);
}
static void env_replace(struct strvec *env, const char *key, const char *val)
{
strvec_pushf(env, "%s=%s", key, val);
}
static int setup_tmp_objdir(const char *root)
{
char *path;
int ret = 0;
path = xstrfmt("%s/pack", root);
ret = mkdir(path, 0777);
free(path);
return ret;
}
struct tmp_objdir *tmp_objdir_create(const char *prefix)
{
static int installed_handlers;
struct tmp_objdir *t;
if (the_tmp_objdir)
BUG("only one tmp_objdir can be used at a time");
t = xcalloc(1, sizeof(*t));
strbuf_init(&t->path, 0);
strvec_init(&t->env);
/*
* Use a string starting with tmp_ so that the builtin/prune.c code
* can recognize any stale objdirs left behind by a crash and delete
* them.
*/
strbuf_addf(&t->path, "%s/tmp_objdir-%s-XXXXXX", get_object_directory(), prefix);
if (!mkdtemp(t->path.buf)) {
/* free, not destroy, as we never touched the filesystem */
tmp_objdir_free(t);
return NULL;
}
the_tmp_objdir = t;
if (!installed_handlers) {
atexit(remove_tmp_objdir);
installed_handlers++;
}
if (setup_tmp_objdir(t->path.buf)) {
tmp_objdir_destroy(t);
return NULL;
}
env_append(&t->env, ALTERNATE_DB_ENVIRONMENT,
absolute_path(get_object_directory()));
env_replace(&t->env, DB_ENVIRONMENT, absolute_path(t->path.buf));
env_replace(&t->env, GIT_QUARANTINE_ENVIRONMENT,
absolute_path(t->path.buf));
return t;
}
/*
* Make sure we copy packfiles and their associated metafiles in the correct
* order. All of these ends_with checks are slightly expensive to do in
* the midst of a sorting routine, but in practice it shouldn't matter.
* We will have a relatively small number of packfiles to order, and loose
* objects exit early in the first line.
*/
static int pack_copy_priority(const char *name)
{
if (!starts_with(name, "pack"))
return 0;
if (ends_with(name, ".keep"))
return 1;
if (ends_with(name, ".pack"))
return 2;
if (ends_with(name, ".rev"))
return 3;
if (ends_with(name, ".idx"))
return 4;
return 5;
}
static int pack_copy_cmp(const char *a, const char *b)
{
return pack_copy_priority(a) - pack_copy_priority(b);
}
static int read_dir_paths(struct string_list *out, const char *path)
{
DIR *dh;
struct dirent *de;
dh = opendir(path);
if (!dh)
return -1;
while ((de = readdir(dh)))
if (de->d_name[0] != '.')
string_list_append(out, de->d_name);
closedir(dh);
return 0;
}
static int migrate_paths(struct strbuf *src, struct strbuf *dst,
enum finalize_object_file_flags flags);
static int migrate_one(struct strbuf *src, struct strbuf *dst,
enum finalize_object_file_flags flags)
{
struct stat st;
if (stat(src->buf, &st) < 0)
return -1;
if (S_ISDIR(st.st_mode)) {
if (!mkdir(dst->buf, 0777)) {
if (adjust_shared_perm(dst->buf))
return -1;
} else if (errno != EEXIST)
return -1;
return migrate_paths(src, dst, flags);
}
return finalize_object_file_flags(src->buf, dst->buf, flags);
}
static int is_loose_object_shard(const char *name)
{
return strlen(name) == 2 && isxdigit(name[0]) && isxdigit(name[1]);
}
static int migrate_paths(struct strbuf *src, struct strbuf *dst,
enum finalize_object_file_flags flags)
{
size_t src_len = src->len, dst_len = dst->len;
struct string_list paths = STRING_LIST_INIT_DUP;
int i;
int ret = 0;
if (read_dir_paths(&paths, src->buf) < 0)
return -1;
paths.cmp = pack_copy_cmp;
string_list_sort(&paths);
for (i = 0; i < paths.nr; i++) {
const char *name = paths.items[i].string;
enum finalize_object_file_flags flags_copy = flags;
strbuf_addf(src, "/%s", name);
strbuf_addf(dst, "/%s", name);
if (is_loose_object_shard(name))
flags_copy |= FOF_SKIP_COLLISION_CHECK;
ret |= migrate_one(src, dst, flags_copy);
strbuf_setlen(src, src_len);
strbuf_setlen(dst, dst_len);
}
string_list_clear(&paths, 0);
return ret;
}
int tmp_objdir_migrate(struct tmp_objdir *t)
{
struct strbuf src = STRBUF_INIT, dst = STRBUF_INIT;
int ret;
if (!t)
return 0;
if (t->prev_odb) {
if (the_repository->objects->odb->will_destroy)
BUG("migrating an ODB that was marked for destruction");
restore_primary_odb(t->prev_odb, t->path.buf);
t->prev_odb = NULL;
}
strbuf_addbuf(&src, &t->path);
strbuf_addstr(&dst, get_object_directory());
ret = migrate_paths(&src, &dst, 0);
strbuf_release(&src);
strbuf_release(&dst);
tmp_objdir_destroy(t);
return ret;
}
const char **tmp_objdir_env(const struct tmp_objdir *t)
{
if (!t)
return NULL;
return t->env.v;
}
void tmp_objdir_add_as_alternate(const struct tmp_objdir *t)
{
add_to_alternates_memory(t->path.buf);
}
void tmp_objdir_replace_primary_odb(struct tmp_objdir *t, int will_destroy)
{
if (t->prev_odb)
BUG("the primary object database is already replaced");
t->prev_odb = set_temporary_primary_odb(t->path.buf, will_destroy);
t->will_destroy = will_destroy;
}
struct tmp_objdir *tmp_objdir_unapply_primary_odb(void)
{
if (!the_tmp_objdir || !the_tmp_objdir->prev_odb)
return NULL;
restore_primary_odb(the_tmp_objdir->prev_odb, the_tmp_objdir->path.buf);
the_tmp_objdir->prev_odb = NULL;
return the_tmp_objdir;
}
void tmp_objdir_reapply_primary_odb(struct tmp_objdir *t, const char *old_cwd,
const char *new_cwd)
{
char *path;
path = reparent_relative_path(old_cwd, new_cwd, t->path.buf);
strbuf_reset(&t->path);
strbuf_addstr(&t->path, path);
free(path);
tmp_objdir_replace_primary_odb(t, t->will_destroy);
}