From: Justin Tobler <jltobler@gmail•com>
To: git@vger•kernel.org
Cc: ps@pks•im, Justin Tobler <jltobler@gmail•com>
Subject: [PATCH 3/3] bulk-checkin: wire repository variable
Date: Wed, 20 Aug 2025 17:55:31 -0500 [thread overview]
Message-ID: <20250820225531.1212935-4-jltobler@gmail.com> (raw)
In-Reply-To: <20250820225531.1212935-1-jltobler@gmail.com>
The bulk-checkin subsystem depends on `the_repository`. Adapt functions
and call sites to wire the repository variable where needed. The
`USE_THE_REPOSITORY_VARIBALE` is still required as the
`pack_compression_level` and `pack_size_limit_cfg` globals are still
used.
Signed-off-by: Justin Tobler <jltobler@gmail•com>
---
bulk-checkin.c | 58 ++++++++++++++++++++++++++++----------------------
bulk-checkin.h | 3 ++-
object-file.c | 3 ++-
3 files changed, 37 insertions(+), 27 deletions(-)
diff --git a/bulk-checkin.c b/bulk-checkin.c
index 53a20a2d92f..a1185883837 100644
--- a/bulk-checkin.c
+++ b/bulk-checkin.c
@@ -38,7 +38,8 @@ struct odb_transaction {
struct bulk_checkin_packfile packfile;
};
-static void finish_tmp_packfile(struct strbuf *basename,
+static void finish_tmp_packfile(struct repository *repo,
+ struct strbuf *basename,
const char *pack_tmp_name,
struct pack_idx_entry **written_list,
uint32_t nr_written,
@@ -47,15 +48,16 @@ static void finish_tmp_packfile(struct strbuf *basename,
{
char *idx_tmp_name = NULL;
- stage_tmp_packfiles(the_repository, basename, pack_tmp_name,
+ stage_tmp_packfiles(repo, basename, pack_tmp_name,
written_list, nr_written, NULL, pack_idx_opts, hash,
&idx_tmp_name);
- rename_tmp_packfile_idx(the_repository, basename, &idx_tmp_name);
+ rename_tmp_packfile_idx(repo, basename, &idx_tmp_name);
free(idx_tmp_name);
}
-static void flush_bulk_checkin_packfile(struct bulk_checkin_packfile *state)
+static void flush_bulk_checkin_packfile(struct bulk_checkin_packfile *state,
+ struct repository *repo)
{
unsigned char hash[GIT_MAX_RAWSZ];
struct strbuf packname = STRBUF_INIT;
@@ -73,15 +75,15 @@ static void flush_bulk_checkin_packfile(struct bulk_checkin_packfile *state)
CSUM_HASH_IN_STREAM | CSUM_FSYNC | CSUM_CLOSE);
} else {
int fd = finalize_hashfile(state->f, hash, FSYNC_COMPONENT_PACK, 0);
- fixup_pack_header_footer(the_hash_algo, fd, hash, state->pack_tmp_name,
+ fixup_pack_header_footer(repo->hash_algo, fd, hash, state->pack_tmp_name,
state->nr_written, hash,
state->offset);
close(fd);
}
- strbuf_addf(&packname, "%s/pack/pack-%s.", repo_get_object_directory(the_repository),
- hash_to_hex(hash));
- finish_tmp_packfile(&packname, state->pack_tmp_name,
+ strbuf_addf(&packname, "%s/pack/pack-%s.", repo_get_object_directory(repo),
+ hash_to_hex_algop(hash, repo->hash_algo));
+ finish_tmp_packfile(repo, &packname, state->pack_tmp_name,
state->written, state->nr_written,
&state->pack_idx_opts, hash);
for (uint32_t i = 0; i < state->nr_written; i++)
@@ -94,7 +96,7 @@ static void flush_bulk_checkin_packfile(struct bulk_checkin_packfile *state)
strbuf_release(&packname);
/* Make objects we just wrote available to ourselves */
- reprepare_packed_git(the_repository);
+ reprepare_packed_git(repo);
}
/*
@@ -117,7 +119,8 @@ static void flush_batch_fsync(struct odb_transaction *transaction)
* to ensure that the data in each new object file is durable before
* the final name is visible.
*/
- strbuf_addf(&temp_path, "%s/bulk_fsync_XXXXXX", repo_get_object_directory(the_repository));
+ strbuf_addf(&temp_path, "%s/bulk_fsync_XXXXXX",
+ repo_get_object_directory(transaction->odb->repo));
temp = xmks_tempfile(temp_path.buf);
fsync_or_die(get_tempfile_fd(temp), get_tempfile_path(temp));
delete_tempfile(&temp);
@@ -131,10 +134,11 @@ static void flush_batch_fsync(struct odb_transaction *transaction)
transaction->objdir = NULL;
}
-static int already_written(struct bulk_checkin_packfile *state, struct object_id *oid)
+static int already_written(struct bulk_checkin_packfile *state,
+ struct repository *repo, struct object_id *oid)
{
/* The object may already exist in the repository */
- if (odb_has_object(the_repository->objects, oid,
+ if (odb_has_object(repo->objects, oid,
HAS_OBJECT_RECHECK_PACKED | HAS_OBJECT_FETCH_PROMISOR))
return 1;
@@ -240,12 +244,13 @@ static int stream_blob_to_pack(struct bulk_checkin_packfile *state,
/* Lazily create backing packfile for the state */
static void prepare_to_stream(struct bulk_checkin_packfile *state,
+ struct repository *repo,
unsigned flags)
{
if (!(flags & INDEX_WRITE_OBJECT) || state->f)
return;
- state->f = create_tmp_packfile(the_repository, &state->pack_tmp_name);
+ state->f = create_tmp_packfile(repo, &state->pack_tmp_name);
reset_pack_idx_option(&state->pack_idx_opts);
/* Pretend we are going to write only one object */
@@ -255,6 +260,7 @@ static void prepare_to_stream(struct bulk_checkin_packfile *state,
}
static int deflate_blob_to_pack(struct bulk_checkin_packfile *state,
+ struct repository *repo,
struct object_id *result_oid,
int fd, size_t size,
const char *path, unsigned flags)
@@ -272,21 +278,21 @@ static int deflate_blob_to_pack(struct bulk_checkin_packfile *state,
header_len = format_object_header((char *)obuf, sizeof(obuf),
OBJ_BLOB, size);
- the_hash_algo->init_fn(&ctx);
+ repo->hash_algo->init_fn(&ctx);
git_hash_update(&ctx, obuf, header_len);
/* Note: idx is non-NULL when we are writing */
if ((flags & INDEX_WRITE_OBJECT) != 0) {
CALLOC_ARRAY(idx, 1);
- prepare_to_stream(state, flags);
+ prepare_to_stream(state, repo, flags);
hashfile_checkpoint_init(state->f, &checkpoint);
}
already_hashed_to = 0;
while (1) {
- prepare_to_stream(state, flags);
+ prepare_to_stream(state, repo, flags);
if (idx) {
hashfile_checkpoint(state->f, &checkpoint);
idx->offset = state->offset;
@@ -304,7 +310,7 @@ static int deflate_blob_to_pack(struct bulk_checkin_packfile *state,
BUG("should not happen");
hashfile_truncate(state->f, &checkpoint);
state->offset = checkpoint.offset;
- flush_bulk_checkin_packfile(state);
+ flush_bulk_checkin_packfile(state, repo);
if (lseek(fd, seekback, SEEK_SET) == (off_t) -1)
return error("cannot seek back");
}
@@ -313,7 +319,7 @@ static int deflate_blob_to_pack(struct bulk_checkin_packfile *state,
return 0;
idx->crc32 = crc32_end(state->f);
- if (already_written(state, result_oid)) {
+ if (already_written(state, repo, result_oid)) {
hashfile_truncate(state->f, &checkpoint);
state->offset = checkpoint.offset;
free(idx);
@@ -338,7 +344,7 @@ void prepare_loose_object_bulk_checkin(struct odb_transaction *transaction)
if (!transaction || transaction->objdir)
return;
- transaction->objdir = tmp_objdir_create(the_repository, "bulk-fsync");
+ transaction->objdir = tmp_objdir_create(transaction->odb->repo, "bulk-fsync");
if (transaction->objdir)
tmp_objdir_replace_primary_odb(transaction->objdir, 0);
}
@@ -361,20 +367,21 @@ void fsync_loose_object_bulk_checkin(struct odb_transaction *transaction,
}
}
-int index_blob_bulk_checkin(struct odb_transaction *transaction,
+int index_blob_bulk_checkin(struct repository *repo,
+ struct odb_transaction *transaction,
struct object_id *oid, int fd, size_t size,
const char *path, unsigned flags)
{
int status;
if (transaction) {
- status = deflate_blob_to_pack(&transaction->packfile, oid, fd,
- size, path, flags);
+ status = deflate_blob_to_pack(&transaction->packfile,
+ repo, oid, fd, size, path, flags);
} else {
struct bulk_checkin_packfile state = { 0 };
- status = deflate_blob_to_pack(&state, oid, fd, size, path, flags);
- flush_bulk_checkin_packfile(&state);
+ status = deflate_blob_to_pack(&state, repo, oid, fd, size, path, flags);
+ flush_bulk_checkin_packfile(&state, repo);
}
return status;
@@ -398,7 +405,8 @@ void flush_odb_transaction(struct odb_transaction *transaction)
return;
flush_batch_fsync(transaction);
- flush_bulk_checkin_packfile(&transaction->packfile);
+ flush_bulk_checkin_packfile(&transaction->packfile,
+ transaction->odb->repo);
}
void end_odb_transaction(struct odb_transaction *transaction)
diff --git a/bulk-checkin.h b/bulk-checkin.h
index 16254ce6a70..ac8dbf3523f 100644
--- a/bulk-checkin.h
+++ b/bulk-checkin.h
@@ -28,7 +28,8 @@ void fsync_loose_object_bulk_checkin(struct odb_transaction *transaction,
* binary blobs, they generally do not want to get any conversion, and
* callers should avoid this code path when filters are requested.
*/
-int index_blob_bulk_checkin(struct odb_transaction *transaction,
+int index_blob_bulk_checkin(struct repository *repo,
+ struct odb_transaction *transaction,
struct object_id *oid, int fd, size_t size,
const char *path, unsigned flags);
diff --git a/object-file.c b/object-file.c
index 1740aa2b2e3..35f33e466c2 100644
--- a/object-file.c
+++ b/object-file.c
@@ -1263,7 +1263,8 @@ int index_fd(struct index_state *istate, struct object_id *oid,
ret = index_core(istate, oid, fd, xsize_t(st->st_size),
type, path, flags);
else
- ret = index_blob_bulk_checkin(the_repository->objects->transaction,
+ ret = index_blob_bulk_checkin(the_repository,
+ the_repository->objects->transaction,
oid, fd, xsize_t(st->st_size),
path, flags);
close(fd);
--
2.51.0
next prev parent reply other threads:[~2025-08-20 22:55 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-20 22:55 [PATCH 0/3] bulk-checkin: remove global transaction state Justin Tobler
2025-08-20 22:55 ` [PATCH 1/3] bulk-checkin: introduce object database transaction structure Justin Tobler
2025-08-20 22:55 ` [PATCH 2/3] bulk-checkin: remove global transaction state Justin Tobler
2025-08-20 22:55 ` Justin Tobler [this message]
2025-08-21 0:15 ` [PATCH 3/3] bulk-checkin: wire repository variable Junio C Hamano
2025-08-21 20:26 ` Justin Tobler
2025-08-21 20:32 ` Junio C Hamano
2025-08-21 0:00 ` [PATCH 0/3] bulk-checkin: remove global transaction state Junio C Hamano
2025-08-21 23:22 ` [PATCH v2 0/4] " Justin Tobler
2025-08-21 23:22 ` [PATCH v2 1/4] bulk-checkin: introduce object database transaction structure Justin Tobler
2025-08-21 23:22 ` [PATCH v2 2/4] bulk-checkin: remove global transaction state Justin Tobler
2025-08-22 16:37 ` Junio C Hamano
2025-08-22 18:07 ` Justin Tobler
2025-08-22 20:25 ` Junio C Hamano
2025-08-21 23:22 ` [PATCH v2 3/4] bulk-checkin: require transaction for index_blob_bulk_checkin() Justin Tobler
2025-08-22 16:49 ` Junio C Hamano
2025-08-22 19:13 ` Justin Tobler
2025-08-22 20:33 ` Junio C Hamano
2025-08-21 23:22 ` [PATCH v2 4/4] bulk-checkin: use repository variable from transaction Justin Tobler
2025-08-22 17:03 ` Junio C Hamano
2025-08-22 19:38 ` Justin Tobler
2025-08-22 21:34 ` [PATCH v3 0/4] bulk-checkin: remove global transaction state Justin Tobler
2025-08-22 21:34 ` [PATCH v3 1/4] bulk-checkin: introduce object database transaction structure Justin Tobler
2025-08-22 21:34 ` [PATCH v3 2/4] bulk-checkin: remove global transaction state Justin Tobler
2025-08-22 21:34 ` [PATCH v3 3/4] bulk-checkin: require transaction for index_blob_bulk_checkin() Justin Tobler
2025-08-22 21:35 ` [PATCH v3 4/4] bulk-checkin: use repository variable from transaction Justin Tobler
2025-08-25 20:25 ` [PATCH v3 0/4] bulk-checkin: remove global transaction state Junio C Hamano
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=20250820225531.1212935-4-jltobler@gmail.com \
--to=jltobler@gmail$(echo .)com \
--cc=git@vger$(echo .)kernel.org \
--cc=ps@pks$(echo .)im \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox