From: Glen Choo <chooglen@google.com>
To: git@vger.kernel.org
Cc: Glen Choo <chooglen@google.com>, Jonathan Tan <jonathantanmy@google.com>
Subject: [PATCH 8/8] submodule: fix bug and remove add_submodule_odb()
Date: Thu, 10 Feb 2022 12:41:52 +0800 [thread overview]
Message-ID: <20220210044152.78352-9-chooglen@google.com> (raw)
In-Reply-To: <20220210044152.78352-1-chooglen@google.com>
add_submodule_odb() is a hack - it adds a submodule's odb as an
alternate, allowing the submodule's objects to be read via
the_repository. Its last caller is submodule_has_commits(), which calls
add_submodule_odb() to prepare for check_has_commit(). This used to be
necessary because check_has_commit() used the_repository's odb, but this
is longer true as of 13a2f620b2 (submodule: pass repo to
check_has_commit(), 2021-10-08).
Removing add_submodule_odb() reveals a bug in check_has_commit(), where
check_has_commit() will segfault if the submodule is missing (e.g. the
user has not init-ed the submodule). This happens because the
submodule's struct repository cannot be initialized, but
check_has_commit() tries to cleanup the uninitialized struct anyway.
This was masked by add_submodule_odb(), because add_submodule_odb()
fails when the submodule is missing, causing the caller to return early
and avoid calling check_has_commit().
Fix the bug and remove the call to add_submodule_odb(). Since
add_submodule_odb() has no more callers, remove it too.
Note that submodule odbs can still by added as alternates via
add_submodule_odb_by_path().
Signed-off-by: Glen Choo <chooglen@google.com>
---
This bug only exists because we can't call repo_clear() twice on the
same struct repository. So instead of just fixing this site, an
alternative (and maybe better) fix would be to fix repo_clear(). If
others think that's a good idea, I'll do that instead.
submodule.c | 35 ++---------------------------------
submodule.h | 9 ++++-----
2 files changed, 6 insertions(+), 38 deletions(-)
diff --git a/submodule.c b/submodule.c
index 0c02bbc9c3..fdfddd3aac 100644
--- a/submodule.c
+++ b/submodule.c
@@ -168,26 +168,6 @@ void stage_updated_gitmodules(struct index_state *istate)
static struct string_list added_submodule_odb_paths = STRING_LIST_INIT_NODUP;
-/* TODO: remove this function, use repo_submodule_init instead. */
-int add_submodule_odb(const char *path)
-{
- struct strbuf objects_directory = STRBUF_INIT;
- int ret = 0;
-
- ret = strbuf_git_path_submodule(&objects_directory, path, "objects/");
- if (ret)
- goto done;
- if (!is_directory(objects_directory.buf)) {
- ret = -1;
- goto done;
- }
- string_list_insert(&added_submodule_odb_paths,
- strbuf_detach(&objects_directory, NULL));
-done:
- strbuf_release(&objects_directory);
- return ret;
-}
-
void add_submodule_odb_by_path(const char *path)
{
string_list_insert(&added_submodule_odb_paths, xstrdup(path));
@@ -972,7 +952,8 @@ static int check_has_commit(const struct object_id *oid, void *data)
if (repo_submodule_init(&subrepo, cb->repo, cb->path, cb->super_oid)) {
cb->result = 0;
- goto cleanup;
+ /* subrepo failed to init, so don't clean it up. */
+ return 0;
}
type = oid_object_info(&subrepo, oid, NULL);
@@ -1003,18 +984,6 @@ static int submodule_has_commits(struct repository *r,
{
struct has_commit_data has_commit = { r, 1, path, super_oid };
- /*
- * Perform a cheap, but incorrect check for the existence of 'commits'.
- * This is done by adding the submodule's object store to the in-core
- * object store, and then querying for each commit's existence. If we
- * do not have the commit object anywhere, there is no chance we have
- * it in the object store of the correct submodule and have it
- * reachable from a ref, so we can fail early without spawning rev-list
- * which is expensive.
- */
- if (add_submodule_odb(path))
- return 0;
-
oid_array_for_each_unique(commits, check_has_commit, &has_commit);
if (has_commit.result) {
diff --git a/submodule.h b/submodule.h
index 784ceffc0e..ca1f12b78b 100644
--- a/submodule.h
+++ b/submodule.h
@@ -103,12 +103,11 @@ int submodule_uses_gitfile(const char *path);
int bad_to_remove_submodule(const char *path, unsigned flags);
/*
- * Call add_submodule_odb() to add the submodule at the given path to a list.
- * When register_all_submodule_odb_as_alternates() is called, the object stores
- * of all submodules in that list will be added as alternates in
- * the_repository.
+ * Call add_submodule_odb_by_path() to add the submodule at the given
+ * path to a list. When register_all_submodule_odb_as_alternates() is
+ * called, the object stores of all submodules in that list will be
+ * added as alternates in the_repository.
*/
-int add_submodule_odb(const char *path);
void add_submodule_odb_by_path(const char *path);
int register_all_submodule_odb_as_alternates(void);
--
2.33.GIT
next prev parent reply other threads:[~2022-02-10 4:42 UTC|newest]
Thread overview: 149+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-10 4:41 [PATCH 0/8] fetch --recurse-submodules: fetch unpopulated submodules Glen Choo
2022-02-10 4:41 ` [PATCH 1/8] submodule: inline submodule_commits() into caller Glen Choo
2022-02-10 4:41 ` [PATCH 2/8] submodule: store new submodule commits oid_array in a struct Glen Choo
2022-02-10 19:00 ` Jonathan Tan
2022-02-10 22:05 ` Junio C Hamano
2022-02-10 4:41 ` [PATCH 3/8] submodule: make static functions read submodules from commits Glen Choo
2022-02-10 19:15 ` Jonathan Tan
2022-02-11 10:07 ` Glen Choo
2022-02-11 10:09 ` Glen Choo
2022-02-10 4:41 ` [PATCH 4/8] t5526: introduce test helper to assert on fetches Glen Choo
2022-02-10 4:41 ` [PATCH 5/8] t5526: use grep " Glen Choo
2022-02-10 19:17 ` Jonathan Tan
2022-02-10 4:41 ` [PATCH 6/8] submodule: extract get_fetch_task() Glen Choo
2022-02-10 19:33 ` Jonathan Tan
2022-02-10 4:41 ` [PATCH 7/8] fetch: fetch unpopulated, changed submodules Glen Choo
2022-02-10 22:49 ` Junio C Hamano
2022-02-11 7:15 ` Glen Choo
2022-02-11 17:07 ` Junio C Hamano
2022-02-10 22:51 ` Jonathan Tan
2022-02-14 4:24 ` Glen Choo
2022-02-14 18:04 ` Glen Choo
2022-02-14 10:17 ` Glen Choo
2022-02-10 4:41 ` Glen Choo [this message]
2022-02-10 22:54 ` [PATCH 8/8] submodule: fix bug and remove add_submodule_odb() Junio C Hamano
2022-02-11 3:13 ` Glen Choo
2022-02-10 23:04 ` Jonathan Tan
2022-02-11 3:18 ` Glen Choo
2022-02-11 17:19 ` Junio C Hamano
2022-02-14 2:52 ` Glen Choo
2022-02-10 7:07 ` [PATCH 0/8] fetch --recurse-submodules: fetch unpopulated submodules Junio C Hamano
2022-02-10 8:51 ` Glen Choo
2022-02-10 17:40 ` Junio C Hamano
2022-02-11 2:39 ` Glen Choo
2022-02-15 17:23 ` [PATCH v2 0/9] " Glen Choo
2022-02-15 17:23 ` [PATCH v2 1/9] t5526: introduce test helper to assert on fetches Glen Choo
2022-02-15 21:37 ` Ævar Arnfjörð Bjarmason
2022-02-15 17:23 ` [PATCH v2 2/9] t5526: use grep " Glen Choo
2022-02-15 21:53 ` Ævar Arnfjörð Bjarmason
2022-02-16 3:09 ` Glen Choo
2022-02-16 10:02 ` Ævar Arnfjörð Bjarmason
2022-02-17 4:04 ` Glen Choo
2022-02-17 9:25 ` Ævar Arnfjörð Bjarmason
2022-02-17 16:16 ` Glen Choo
2022-02-15 17:23 ` [PATCH v2 3/9] submodule: make static functions read submodules from commits Glen Choo
2022-02-15 21:18 ` Jonathan Tan
2022-02-16 6:59 ` Glen Choo
2022-02-15 22:00 ` Ævar Arnfjörð Bjarmason
2022-02-16 7:08 ` Glen Choo
2022-02-15 17:23 ` [PATCH v2 4/9] submodule: inline submodule_commits() into caller Glen Choo
2022-02-15 22:02 ` Ævar Arnfjörð Bjarmason
2022-02-15 17:23 ` [PATCH v2 5/9] submodule: store new submodule commits oid_array in a struct Glen Choo
2022-02-15 21:33 ` Ævar Arnfjörð Bjarmason
2022-02-15 17:23 ` [PATCH v2 6/9] submodule: extract get_fetch_task() Glen Choo
2022-02-15 17:23 ` [PATCH v2 7/9] fetch: fetch unpopulated, changed submodules Glen Choo
2022-02-15 22:02 ` Jonathan Tan
2022-02-16 5:46 ` Glen Choo
2022-02-16 9:11 ` Glen Choo
2022-02-16 9:39 ` Ævar Arnfjörð Bjarmason
2022-02-16 17:33 ` Glen Choo
2022-02-15 22:06 ` Ævar Arnfjörð Bjarmason
2022-02-15 17:23 ` [PATCH v2 8/9] submodule: read shallows when finding " Glen Choo
2022-02-15 22:03 ` Jonathan Tan
2022-02-15 17:23 ` [PATCH v2 9/9] submodule: fix latent check_has_commit() bug Glen Choo
2022-02-15 22:04 ` Jonathan Tan
2022-02-24 10:08 ` [PATCH v3 00/10] fetch --recurse-submodules: fetch unpopulated submodules Glen Choo
2022-02-24 10:08 ` [PATCH v3 01/10] t5526: introduce test helper to assert on fetches Glen Choo
2022-02-25 0:34 ` Junio C Hamano
2022-02-24 10:08 ` [PATCH v3 02/10] t5526: stop asserting on stderr literally Glen Choo
2022-02-24 11:52 ` Ævar Arnfjörð Bjarmason
2022-02-24 16:15 ` Glen Choo
2022-02-24 18:13 ` Eric Sunshine
2022-02-24 23:05 ` Jonathan Tan
2022-02-25 2:26 ` Glen Choo
2022-02-24 10:08 ` [PATCH v3 03/10] t5526: create superproject commits with test helper Glen Choo
2022-02-24 23:14 ` Jonathan Tan
2022-02-25 2:52 ` Glen Choo
2022-02-25 11:42 ` Ævar Arnfjörð Bjarmason
2022-02-28 18:11 ` Glen Choo
2022-02-24 10:08 ` [PATCH v3 04/10] submodule: make static functions read submodules from commits Glen Choo
2022-02-24 10:08 ` [PATCH v3 05/10] submodule: inline submodule_commits() into caller Glen Choo
2022-02-24 10:08 ` [PATCH v3 06/10] submodule: store new submodule commits oid_array in a struct Glen Choo
2022-02-24 10:08 ` [PATCH v3 07/10] submodule: extract get_fetch_task() Glen Choo
2022-02-24 23:26 ` Jonathan Tan
2022-02-24 10:08 ` [PATCH v3 08/10] submodule: move logic into fetch_task_create() Glen Choo
2022-02-24 23:36 ` Jonathan Tan
2022-02-24 10:08 ` [PATCH v3 09/10] fetch: fetch unpopulated, changed submodules Glen Choo
2022-02-24 21:30 ` Junio C Hamano
2022-02-25 3:04 ` Glen Choo
2022-02-25 0:33 ` Junio C Hamano
2022-02-25 3:07 ` Glen Choo
2022-02-25 0:39 ` Jonathan Tan
2022-02-25 3:46 ` Glen Choo
2022-03-04 23:46 ` Jonathan Tan
2022-03-05 0:22 ` Glen Choo
2022-03-04 23:53 ` Jonathan Tan
2022-02-26 18:53 ` Junio C Hamano
2022-03-01 20:24 ` Johannes Schindelin
2022-03-01 20:33 ` Junio C Hamano
2022-03-02 23:25 ` Glen Choo
2022-03-01 20:32 ` Junio C Hamano
2022-02-24 10:08 ` [PATCH v3 10/10] submodule: fix latent check_has_commit() bug Glen Choo
2022-03-04 0:57 ` [PATCH v4 00/10] fetch --recurse-submodules: fetch unpopulated submodules Glen Choo
2022-03-04 0:57 ` [PATCH v4 01/10] t5526: introduce test helper to assert on fetches Glen Choo
2022-03-04 2:06 ` Junio C Hamano
2022-03-04 22:11 ` Glen Choo
2022-03-04 0:57 ` [PATCH v4 02/10] t5526: stop asserting on stderr literally Glen Choo
2022-03-04 2:12 ` Junio C Hamano
2022-03-04 22:41 ` Jonathan Tan
2022-03-04 23:48 ` Junio C Hamano
2022-03-05 0:25 ` Glen Choo
2022-03-04 0:57 ` [PATCH v4 03/10] t5526: create superproject commits with test helper Glen Choo
2022-03-04 22:59 ` Jonathan Tan
2022-03-04 0:57 ` [PATCH v4 04/10] submodule: make static functions read submodules from commits Glen Choo
2022-03-04 0:57 ` [PATCH v4 05/10] submodule: inline submodule_commits() into caller Glen Choo
2022-03-04 0:57 ` [PATCH v4 06/10] submodule: store new submodule commits oid_array in a struct Glen Choo
2022-03-04 0:57 ` [PATCH v4 07/10] submodule: extract get_fetch_task() Glen Choo
2022-03-04 0:57 ` [PATCH v4 08/10] submodule: move logic into fetch_task_create() Glen Choo
2022-03-04 0:57 ` [PATCH v4 09/10] fetch: fetch unpopulated, changed submodules Glen Choo
2022-03-04 2:37 ` Junio C Hamano
2022-03-04 22:59 ` Glen Choo
2022-03-05 0:13 ` Junio C Hamano
2022-03-05 0:37 ` Glen Choo
2022-03-08 0:11 ` Junio C Hamano
2022-03-04 23:56 ` Jonathan Tan
2022-03-04 0:57 ` [PATCH v4 10/10] submodule: fix latent check_has_commit() bug Glen Choo
2022-03-04 2:17 ` Junio C Hamano
2022-03-04 2:22 ` [PATCH v4 00/10] fetch --recurse-submodules: fetch unpopulated submodules Junio C Hamano
2022-03-08 0:14 ` [PATCH v5 " Glen Choo
2022-03-08 0:14 ` [PATCH v5 01/10] t5526: introduce test helper to assert on fetches Glen Choo
2022-03-08 0:14 ` [PATCH v5 02/10] t5526: stop asserting on stderr literally Glen Choo
2022-03-08 0:14 ` [PATCH v5 03/10] t5526: create superproject commits with test helper Glen Choo
2022-03-08 0:14 ` [PATCH v5 04/10] submodule: make static functions read submodules from commits Glen Choo
2022-03-08 0:14 ` [PATCH v5 05/10] submodule: inline submodule_commits() into caller Glen Choo
2022-03-08 0:14 ` [PATCH v5 06/10] submodule: store new submodule commits oid_array in a struct Glen Choo
2022-03-08 0:14 ` [PATCH v5 07/10] submodule: extract get_fetch_task() Glen Choo
2022-03-08 0:14 ` [PATCH v5 08/10] submodule: move logic into fetch_task_create() Glen Choo
2022-03-08 0:14 ` [PATCH v5 09/10] fetch: fetch unpopulated, changed submodules Glen Choo
2022-03-08 0:14 ` [PATCH v5 10/10] submodule: fix latent check_has_commit() bug Glen Choo
2022-03-08 0:50 ` [PATCH v5 00/10] fetch --recurse-submodules: fetch unpopulated submodules Junio C Hamano
2022-03-08 18:24 ` Glen Choo
2022-03-09 19:13 ` Junio C Hamano
2022-03-09 19:49 ` Glen Choo
2022-03-09 20:22 ` Junio C Hamano
2022-03-09 22:11 ` Glen Choo
2022-03-16 21:58 ` Glen Choo
2022-03-16 22:06 ` Junio C Hamano
2022-03-16 22:37 ` Glen Choo
2022-03-16 23:08 ` Junio C Hamano
2022-03-08 21:42 ` Jonathan Tan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20220210044152.78352-9-chooglen@google.com \
--to=chooglen@google.com \
--cc=git@vger.kernel.org \
--cc=jonathantanmy@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.