Introduce `USE_THE_REPOSITORY_VARIABLE` macro
Quoted 11 lines · Ramsay Jones
Quoted 9 lines · Junio C Hamano
Quoted 3 lines · Ramsay Jones
Hmm, odd; isn't the declaration of '
the_repository' from the "repository.h" header file visible at this point?No. The declaration is guarded with
USE_THE_REPOSITORY_VARIABLECPP macro in the header, andrepository.cdoes not define it.
Quoted 6 lines
Ah, OK. I haven't been following too closely and didn't notice that the declaration in the header file was now conditional. :(
But that does beg the question - why is
repository.cnot defining theUSE_THE_REPOSITORY_VARIABLE?
I think the goal of the series is to eventually get to the point where nobody uses the_repository variable. If repository.c, which consists of a set of service routines that work on a repository instance, defined it, showing willingness to implicitly rely on the_repository through things like get_oid_hex() (which would rely on the_repository->hash_algo), that would go the opposite direction, so everything, other than the definition of the_repository variable itself that allows other files that still do rely implicitly on the variable to link with it, in repository.c would actively want to refuse to use services only available to those who define USE_THE_* macro.
It is a similar pattern we took when we weaned ourselves off of the_index compatibility macros. The read-cache.c was the central thing that defined the in-core index services, and it was the first thing that lost the_index compatibility macros like read_cache(), add_file_to_cache(), etc. that implicitly worked on the singleton the_index instance, while everybody else still relied on the singleton and services that implicitly worked on the singleton.
Quoted 20 lines · Junio C Hamano
Quoted 11 lines · Ramsay Jones
Quoted 9 lines · Junio C Hamano
Quoted 3 lines · Ramsay Jones
Hmm, odd; isn't the declaration of '
the_repository' from the "repository.h" header file visible at this point?No. The declaration is guarded with
USE_THE_REPOSITORY_VARIABLECPP macro in the header, andrepository.cdoes not define it.Quoted 6 lines
Ah, OK. I haven't been following too closely and didn't notice that the declaration in the header file was now conditional. :(
But that does beg the question - why is
repository.cnot defining theUSE_THE_REPOSITORY_VARIABLE?
Quoted 11 lines
I think the goal of the series is to eventually get to the point where nobody uses
the_repositoryvariable. Ifrepository.c, which consists of a set of service routines that work on a repository instance, defined it, showing willingness to implicitly rely onthe_repositorythrough things likeget_oid_hex()(which would rely onthe_repository->hash_algo), that would go the opposite direction, so everything, other than the definition ofthe_repositoryvariable itself that allows other files that still do rely implicitly on the variable to link with it, inrepository.cwould actively want to refuse to use services only available to those who define USE_THE_* macro.
Exactly, that's why it doesn't declare USE_THE_REPOSITORY_VARIABLE. The macro doesn't only guard use of the_repository, but does also guards other functions that implicitly relies on it, and we do not want to use these in "repository.c". So even though the added extern declaration is somewhat ugly, I think it is preferable over defining the macro.
Patrick
Quoted 15 lines · Patrick Steinhardt
Quoted 14 lines
Quoted 2 lines
But that does beg the question - why is
repository.cnot defining theUSE_THE_REPOSITORY_VARIABLE?I think the goal of the series is to eventually get to the point where nobody uses
the_repositoryvariable. Ifrepository.c, which consists of a set of service routines that work on a repository instance, defined it, showing willingness to implicitly rely onthe_repositorythrough things likeget_oid_hex()(which would rely onthe_repository->hash_algo), that would go the opposite direction, so everything, other than the definition ofthe_repositoryvariable itself that allows other files that still do rely implicitly on the variable to link with it, inrepository.cwould actively want to refuse to use services only available to those who define USE_THE_* macro.
Quoted 6 lines
Exactly, that's why it doesn't declare
USE_THE_REPOSITORY_VARIABLE. The macro doesn't only guard use ofthe_repository, but does also guards other functions that implicitly relies on it, and we do not want to use these in "repository.c". So even though the addedexterndeclaration is somewhat ugly, I think it is preferable over defining the macro.
Slightly off-topic, but in retrospect, the approach of marking "this file is done" taken by the very initial version of the "no more implicit access to the_index" series may have been easier to grok easier to grok for reviewers casually looking at it from the sideline, than marking "this file still needs work". When we gradually deprecated the convenience API that implicitly access the_index instance, we marked with NO_THE_INDEX_COMPATIBILITY_MACROS the files that no longer require it (in other words, we are done with the file wrt the transition). We later flipped the polarity and gave USE_THE_INDEX_COMPATIBILITY_MACROS to those that are not marked with NO_THE_INDEX_COMPATIBILITY_MACROS as we progressed and the source files that still used the convenience API has become minorities.
Anyway, I think now people understood what this series aims to achieve and how it does so, hopefully.
Thanks.
v3 cover Introduce `USE_THE_REPOSITORY_VARIABLE` macro
In reply to Patrick SteinhardtHi,
this is the third version of my patch series that introduces the USE_THE_REPOSITORY_VARIABLE macro. When unset, this will cause us to hide the_repository, the_hash_algo and several other functions that implicitly rely on those global variables from our headers. This is a first step towards fully getting rid of this global state in favor of passing it down explicitly via function parameters.
Changes compared to v2:
- Note in a commit message that we aim to have a faithful conversion when introducing a struct git_hash_algo parameter to functions. So even in case the calling context has a struct git_hash_algo available via a local repository, we still use the_repository such that there cannot be a change in behaviour here. Fixing those sites will be left for a future patch series such that we can avoid any kind of regressions caused by this comparatively large refactoring. I also adapted some conversions to fully follow through with this intent.
- Fix an issue with sparse by adding another extern declaration of the_repository to "repository.c".
Thanks!
Patrick
Patrick Steinhardt (20):
hash: drop (mostly) unused is_empty_{blob,tree}_sha1() functions
hash: require hash algorithm in hasheq(), hashcmp() and
`hashclr()`
hash: require hash algorithm in oidread() and oidclr()
global: ensure that object IDs are always padded
hash: convert oidcmp() and oideq() to compare whole hash
hash: make is_null_oid() independent of the_repository
hash: require hash algorithm in is_empty_{blob,tree}_oid()
hash: require hash algorithm in empty_tree_oid_hex()
global: introduce USE_THE_REPOSITORY_VARIABLE macro
refs: avoid include cycle with "repository.h"
hash-ll: merge with "hash.h"
http-fetch: don't crash when parsing packfile without a repo
oidset: pass hash algorithm when parsing file
protocol-caps: use hash algorithm from passed-in repository
replace-object: use hash algorithm from passed-in repository compat/fsmonitor: fix socket path in networked SHA256 repos t/helper: use correct object hash in partial-clone helper t/helper: fix segfault in "oid-array" command without repository t/helper: remove dependency on the_repository in "proc-receive"
hex: guard declarations with USE_THE_REPOSITORY_VARIABLE
add-interactive.c | 4 +- add-patch.c | 4 +- apply.c | 4 +- apply.h | 2 +- archive-tar.c | 3 + archive-zip.c | 3 + archive.c | 2 + attr.c | 2 + bisect.c | 2 + blame.c | 4 +- bloom.c | 1 + branch.c | 2 + builtin.h | 8 + builtin/am.c | 8 +- builtin/blame.c | 3 +- builtin/fast-export.c | 2 +- builtin/fast-import.c | 43 ++- builtin/fetch-pack.c | 4 +- builtin/index-pack.c | 11 +- builtin/log.c | 4 +- builtin/merge.c | 7 +- builtin/notes.c | 2 +- builtin/pack-objects.c | 3 +- builtin/pack-redundant.c | 10 +- builtin/patch-id.c | 6 +- builtin/pull.c | 6 +- builtin/receive-pack.c | 4 +- builtin/replace.c | 2 +- builtin/rm.c | 2 +- builtin/tag.c | 2 +- builtin/unpack-objects.c | 9 +- builtin/update-ref.c | 8 +- bulk-checkin.c | 3 + bundle-uri.c | 2 + bundle.c | 2 + cache-tree.c | 7 +- checkout.c | 2 + checkout.h | 2 +- chunk-format.c | 2 + chunk-format.h | 2 +- combine-diff.c | 2 + commit-graph.c | 22 +- commit-graph.h | 2 + commit-reach.c | 2 + commit.c | 2 + common-main.c | 2 + compat/fsmonitor/fsm-ipc-darwin.c | 7 +- compat/sha1-chunked.c | 2 +- compat/win32/trace2_win32_process_info.c | 2 + config.c | 3 + connected.c | 2 + convert.c | 2 + convert.h | 2 +- csum-file.c | 9 +- csum-file.h | 2 +- delta-islands.c | 2 + diagnose.c | 2 + diff-lib.c | 7 +- diff.c | 9 +- diff.h | 2 +- diffcore-break.c | 3 + diffcore-rename.c | 7 +- diffcore.h | 2 +- dir.c | 9 +- dir.h | 2 +- entry.c | 2 + environment.c | 3 + fetch-pack.c | 2 + fmt-merge-msg.c | 2 + fsck.c | 5 +- fsmonitor-ipc.c | 2 + git.c | 2 + hash-ll.h | 310 ---------------- hash-lookup.c | 5 +- hash.h | 366 ++++++++++++++++--- help.c | 2 + hex.c | 8 +- hex.h | 28 +- http-backend.c | 2 + http-fetch.c | 8 +- http-push.c | 5 +- http-walker.c | 6 +- http.c | 2 + list-objects-filter-options.c | 2 + list-objects-filter.c | 2 + list-objects.c | 2 + log-tree.c | 2 + loose.c | 2 + loose.h | 2 + ls-refs.c | 2 + mailmap.c | 2 + match-trees.c | 6 +- merge-blobs.c | 2 + merge-ort.c | 2 + merge-ort.h | 2 +- merge-recursive.c | 3 + merge.c | 2 + midx-write.c | 2 + midx.c | 5 +- negotiator/default.c | 2 + negotiator/skipping.c | 2 + notes-cache.c | 2 + notes-merge.c | 8 +- notes-utils.c | 2 + notes.c | 14 +- object-file-convert.c | 6 +- object-file.c | 19 +- object-name.c | 2 + object.c | 2 + object.h | 2 +- oid-array.c | 2 + oidmap.h | 2 +- oidset.c | 8 +- oidset.h | 4 +- oidtree.c | 4 +- oidtree.h | 2 +- oss-fuzz/fuzz-commit-graph.c | 2 + pack-bitmap-write.c | 6 +- pack-bitmap.c | 5 +- pack-check.c | 7 +- pack-revindex.c | 2 + pack-write.c | 5 +- packfile.c | 20 +- packfile.h | 2 + parallel-checkout.c | 8 +- parse-options-cb.c | 2 + path.c | 3 + pathspec.c | 2 + pretty.c | 2 + progress.c | 2 + promisor-remote.c | 2 + protocol-caps.c | 5 +- range-diff.c | 2 + reachable.c | 2 + read-cache-ll.h | 2 +- read-cache.c | 21 +- rebase-interactive.c | 2 + ref-filter.c | 2 + reflog-walk.c | 2 + reflog.c | 2 + refs.c | 8 +- refs.h | 8 +- refs/files-backend.c | 8 +- refs/packed-backend.c | 8 +- refs/ref-cache.h | 2 +- refs/reftable-backend.c | 39 +- refspec.c | 2 + reftable/dump.c | 2 +- reftable/reftable-record.h | 2 +- reftable/system.h | 2 +- remote-curl.c | 2 + remote.c | 10 +- remote.h | 2 +- replace-object.c | 2 +- repository.c | 8 + repository.h | 9 +- rerere.c | 2 + reset.c | 2 + reset.h | 2 +- resolve-undo.c | 5 +- resolve-undo.h | 2 +- revision.c | 2 + run-command.c | 2 + scalar.c | 2 + send-pack.c | 2 + sequencer.c | 8 +- serve.c | 4 +- server-info.c | 2 + setup.c | 2 + shallow.c | 2 + split-index.c | 4 +- split-index.h | 2 +- streaming.c | 3 + submodule-config.c | 4 +- submodule.c | 8 +- t/helper/test-bitmap.c | 2 + t/helper/test-bloom.c | 2 + t/helper/test-cache-tree.c | 2 + t/helper/test-dump-cache-tree.c | 2 + t/helper/test-dump-fsmonitor.c | 2 + t/helper/test-dump-split-index.c | 2 + t/helper/test-dump-untracked-cache.c | 2 + t/helper/test-find-pack.c | 2 + t/helper/test-fsmonitor-client.c | 2 + t/helper/test-hash-speed.c | 2 +- t/helper/test-lazy-init-name-hash.c | 2 + t/helper/test-match-trees.c | 2 + t/helper/test-oid-array.c | 4 + t/helper/test-oidmap.c | 2 + t/helper/test-pack-mtimes.c | 2 + t/helper/test-partial-clone.c | 2 +- t/helper/test-proc-receive.c | 9 +- t/helper/test-reach.c | 2 + t/helper/test-read-cache.c | 2 + t/helper/test-read-graph.c | 2 + t/helper/test-read-midx.c | 2 + t/helper/test-ref-store.c | 2 + t/helper/test-repository.c | 2 + t/helper/test-revision-walking.c | 2 + t/helper/test-scrap-cache-tree.c | 2 + t/helper/test-sha1.c | 2 +- t/helper/test-sha256.c | 2 +- t/helper/test-submodule-config.c | 4 +- t/helper/test-submodule-nested-repo-config.c | 2 + t/helper/test-submodule.c | 2 + t/helper/test-trace2.c | 2 + t/helper/test-write-cache.c | 2 + t/t0064-oid-array.sh | 18 + t/t5550-http-fetch-dumb.sh | 6 + t/test-lib-functions.sh | 5 + t/unit-tests/lib-oid.h | 2 +- t/unit-tests/t-example-decorate.c | 2 + tag.c | 2 + tmp-objdir.c | 2 + transport-helper.c | 2 + transport.c | 2 + tree-diff.c | 1 + tree-walk.c | 6 +- tree-walk.h | 2 +- tree.c | 2 + unpack-trees.c | 2 + upload-pack.c | 2 + walker.c | 2 + worktree.c | 2 + wt-status.c | 6 +- xdiff-interface.c | 2 + xdiff-interface.h | 2 +- 227 files changed, 1002 insertions(+), 617 deletions(-) delete mode 100644 hash-ll.h
Range-diff against v2:
1: d2154e8c45 = 1: d2154e8c45 hash: drop (mostly) unused `is_empty_{blob,tree}_sha1()` functions
2: aa468c3d88 ! 2: c481479598 hash: require hash algorithm in `hasheq()`, `hashcmp()` and `hashclr()` @@ Commit message
As those functions are now independent of the_repository, we can move them from "hash.h" to "hash-ll.h".
+ Note that both in this and subsequent commits in this series we always
+ just pass the_repository->hash_algo as input even if it is obvious
+ that there is a repository in the context that we should be using the
+ hash from instead. This is done to be on the safe side and not introduce
+ any regressions. All callsites should eventually be amended to use a
+ repo passed via parameters, but this is outside the scope of this patch
+ series.
+
## builtin/index-pack.c ##
@@ pack-check.c: static int verify_packfile(struct repository *r,
r->hash_algo->final_fn(hash, &ctx);
pack_sig = use_pack(p, w_curs, pack_sig_ofs, NULL);
- if (!hasheq(hash, pack_sig))
-+ if (!hasheq(hash, pack_sig, r->hash_algo))
++ if (!hasheq(hash, pack_sig, the_repository->hash_algo))
err = error("%s pack checksum mismatch",
p->pack_name);
- if (!hasheq(index_base + index_size - r->hash_algo->hexsz, pack_sig))
+ if (!hasheq(index_base + index_size - r->hash_algo->hexsz, pack_sig,
-+ r->hash_algo))
++ the_repository->hash_algo))
err = error("%s pack checksum does not match its index",
p->pack_name);
unuse_pack(w_curs);
3: 403ea4485b ! 3: 226173a92b hash: require hash algorithm in oidread() and oidclr()
@@ commit-graph.c: static struct tree *load_tree_for_commit(struct repository *r,
st_mult(GRAPH_DATA_WIDTH, graph_pos - g->num_commits_in_base);
- oidread(&oid, commit_data);
-+ oidread(&oid, commit_data, r->hash_algo);
++ oidread(&oid, commit_data, the_repository->hash_algo);
set_commit_tree(c, lookup_tree(r, &oid));
return c->maybe_tree;
@@ commit-graph.c: int write_commit_graph(struct object_directory *odb,
struct object_id oid;
- oidread(&oid, g->chunk_oid_lookup + st_mult(g->hash_len, i));
+ oidread(&oid, g->chunk_oid_lookup + st_mult(g->hash_len, i),
-+ r->hash_algo);
++ the_repository->hash_algo);
oid_array_append(&ctx->oids, &oid);
}
}
@@ commit-graph.c: static int verify_one_commit_graph(struct repository *r,
- oidread(&cur_oid, g->chunk_oid_lookup + st_mult(g->hash_len, i));
+ oidread(&cur_oid, g->chunk_oid_lookup + st_mult(g->hash_len, i),
-+ r->hash_algo);
++ the_repository->hash_algo);
if (i && oidcmp(&prev_oid, &cur_oid) >= 0)
graph_report(_("commit-graph has incorrect OID order: %s then %s"),
4: fa263d6b07 = 4: 3947c0c04d global: ensure that object IDs are always padded
5: a7df209bda = 5: 91eb94bc0b hash: convert `oidcmp()` and `oideq()` to compare whole hash
6: 9058837c93 = 6: 9ae29e1912 hash: make `is_null_oid()` independent of `the_repository`
7: d26584dc8f = 7: da48d5fdea hash: require hash algorithm in `is_empty_{blob,tree}_oid()`
8: 4858cca25f = 8: 1cf25bec3e hash: require hash algorithm in `empty_tree_oid_hex()`
9: cb3694ad0e ! 9: 7e023a335f global: introduce `USE_THE_REPOSITORY_VARIABLE` macro
@@ remote.c
#include "abspath.h"
#include "config.h"
+ ## repository.c ##
+@@
+ #include "promisor-remote.h"
+ #include "refs.h"
+
++/*
++ * We do not define `USE_THE_REPOSITORY_VARIABLE` in this file because we do
++ * not want to rely on functions that implicitly use `the_repository`. This
++ * means that the `extern` declaration of `the_repository` isn't visible here,
++ * which makes sparse unhappy. We thus declare it here.
++ */
++extern struct repository *the_repository;
++
+ /* The main repository */
+ static struct repository the_repo;
+ struct repository *the_repository = &the_repo;
+
## repository.h ##
@@ repository.h: struct repository {
unsigned different_commondir:1;
10: 4492548209 = 10: 74c88ebc39 refs: avoid include cycle with "repository.h"
11: f3cbc4b9f9 = 11: fe4550ba0c hash-ll: merge with "hash.h"
12: 9178098dd7 = 12: 30babd8a67 http-fetch: don't crash when parsing packfile without a repo
13: 0b4436c32b = 13: fa41a85c7e oidset: pass hash algorithm when parsing file
14: c7abfbc489 = 14: 0b42208e2f protocol-caps: use hash algorithm from passed-in repository
15: 9ae4fdb8f1 = 15: e1f4bffea1 replace-object: use hash algorithm from passed-in repository
16: 3ceb726655 = 16: 5b8f981ea2 compat/fsmonitor: fix socket path in networked SHA256 repos
17: 74e5489bd0 = 17: ad83b17ad0 t/helper: use correct object hash in partial-clone helper
18: 470aea1fc8 = 18: a488363bcb t/helper: fix segfault in "oid-array" command without repository
19: 1f0682fc7d = 19: 5de7a01af5 t/helper: remove dependency on `the_repository` in "proc-receive"
20: 16fb86c2b2 = 20: 436ffc0570 hex: guard declarations with `USE_THE_REPOSITORY_VARIABLE`Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 1/20 hash: drop (mostly) unused `is_empty_{blob,tree}_sha1()` functions
The functions is_empty_{blob,tree}_sha1() are mostly unused, except for a single callsite in "read-cache.c". Most callsites have long since been converted to use the equivalents that accept a struct object_id instead of a string.
Adapt the remaining callsite and drop those functions.
--- hash.h | 10 ---------- read-cache.c | 2 +- 2 files changed, 1 insertion(+), 11 deletions(-)
hash.h +0 -10
@@ -84,21 +84,11 @@ static inline void oidread(struct object_id *oid, const unsigned char *hash)oidread_algop (oid, hash, the_hash_algo); } -static inline int is_empty_blob_sha1 (const unsigned char *sha1) -{ -return hasheq (sha1, the_hash_algo->empty_blob->hash); -} -static inline int is_empty_blob_oid (const struct object_id *oid) {return oideq (oid, the_hash_algo->empty_blob); } -static inline int is_empty_tree_sha1 (const unsigned char *sha1) -{ -return hasheq (sha1, the_hash_algo->empty_tree->hash); -} -static inline int is_empty_tree_oid (const struct object_id *oid) {return oideq (oid, the_hash_algo->empty_tree);
read-cache.c +1 -1
@@ -337,7 +337,7 @@ static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)/* Racily smudged entry? */ if (!ce->ce_stat_data.sd_size) { -if (!is_empty_blob_sha1 (ce->oid.hash)) +if (!is_empty_blob_oid (&ce->oid)) changed |=DATA_CHANGED ; }
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 2/20 hash: require hash algorithm in `hasheq()`, `hashcmp()` and `hashclr()`
In reply to Patrick SteinhardtMany of our hash functions have two variants, one receiving a struct git_hash_algo and one that derives it via the_repository. Adapt all of those functions to always require the hash algorithm as input and drop the variants that do not accept one.
As those functions are now independent of the_repository, we can move them from "hash.h" to "hash-ll.h".
Note that both in this and subsequent commits in this series we always just pass the_repository->hash_algo as input even if it is obvious that there is a repository in the context that we should be using the hash from instead. This is done to be on the safe side and not introduce any regressions. All callsites should eventually be amended to use a repo passed via parameters, but this is outside the scope of this patch series.
--- builtin/index-pack.c | 6 +++--- builtin/pack-redundant.c | 8 +++++--- builtin/unpack-objects.c | 3 ++- commit-graph.c | 3 ++- csum-file.c | 6 +++--- hash-ll.h | 15 +++++++++++++-- hash-lookup.c | 3 ++- hash.h | 24 ++---------------------- http-walker.c | 2 +- match-trees.c | 2 +- notes.c | 2 +- pack-bitmap-write.c | 4 ++-- pack-bitmap.c | 3 ++- pack-check.c | 5 +++-- pack-write.c | 3 ++- packfile.c | 8 ++++---- read-cache.c | 8 ++++---- 17 files changed, 52 insertions(+), 53 deletions(-)
builtin/index-pack.c +3 -3
@@ -1204,7 +1204,7 @@ static void parse_pack_objects(unsigned char *hash) the_hash_algo->init_fn (&tmp_ctx); the_hash_algo->clone_fn (&tmp_ctx, &input_ctx); the_hash_algo->final_fn (hash, &tmp_ctx); -if (!hasheq (fill (the_hash_algo->rawsz), hash)) +if (!hasheq (fill (the_hash_algo->rawsz), hash, the_repository->hash_algo))die (_ ("pack is corrupted (SHA1 mismatch)" ));use (the_hash_algo->rawsz);
@@ -1307,11 +1307,11 @@ static void conclude_pack(int fix_thin_pack, const char *curr_pack, unsigned chastop_progress_msg (&progress, msg.buf);strbuf_release (&msg);finalize_hashfile (f, tail_hash,FSYNC_COMPONENT_PACK ,0 ); -hashcpy (read_hash, pack_hash); +hashcpy (read_hash, pack_hash, the_repository->hash_algo);fixup_pack_header_footer (output_fd, pack_hash, curr_pack, nr_objects, read_hash, consumed_bytes-the_hash_algo->rawsz); -if (!hasheq (read_hash, tail_hash)) +if (!hasheq (read_hash, tail_hash, the_repository->hash_algo))die (_ ("Unexpected tail checksum for %s " "(disk corruption?)" ), curr_pack); }
builtin/pack-redundant.c +5 -3
@@ -155,7 +155,7 @@ static inline struct llist_item * llist_sorted_remove(struct llist *list, const l = (hint ==NULL ) ? list->front : hint; prev =NULL ;while (l) { -const int cmp =hashcmp (l->oid.hash, oid); +const int cmp =hashcmp (l->oid.hash, oid, the_repository->hash_algo);if (cmp >0 )/* not in list, since sorted */ return prev;if (!cmp) {/* found */
@@ -258,7 +258,8 @@ static void cmp_two_packs(struct pack_list *p1, struct pack_list *p2)while (p1_off < p1->pack->num_objects * p1_step && p2_off < p2->pack->num_objects * p2_step) { -const int cmp =hashcmp (p1_base + p1_off, p2_base + p2_off); +const int cmp =hashcmp (p1_base + p1_off, p2_base + p2_off, + the_repository->hash_algo);/* cmp ~ p1 - p2 */ if (cmp ==0 ) { p1_hint = llist_sorted_remove(p1->unique_objects,
@@ -296,7 +297,8 @@ static size_t sizeof_union(struct packed_git *p1, struct packed_git *p2)while (p1_off < p1->num_objects * p1_step && p2_off < p2->num_objects * p2_step) { -int cmp =hashcmp (p1_base + p1_off, p2_base + p2_off); +int cmp =hashcmp (p1_base + p1_off, p2_base + p2_off, + the_repository->hash_algo);/* cmp ~ p1 - p2 */ if (cmp ==0 ) { ret++;
builtin/unpack-objects.c +2 -1
@@ -674,7 +674,8 @@ int cmd_unpack_objects(int argc, const char **argv, const char *prefix UNUSED)if (fsck_finish (&fsck_options))die (_ ("fsck error in pack objects" )); } -if (!hasheq (fill (the_hash_algo->rawsz), oid.hash)) +if (!hasheq (fill (the_hash_algo->rawsz), oid.hash, + the_repository->hash_algo))die ("final sha1 did not match" );use (the_hash_algo->rawsz);
commit-graph.c +2 -1
@@ -565,7 +565,8 @@ static int add_graph_to_chain(struct commit_graph *g,if (!cur_g || !oideq (&oids[n], &cur_g->oid) || - !hasheq (oids[n].hash, g->chunk_base_graphs +st_mult (g->hash_len, n))) { + !hasheq (oids[n].hash, g->chunk_base_graphs +st_mult (g->hash_len, n), + the_repository->hash_algo)) {warning (_ ("commit-graph chain does not match" ));return 0 ; }
csum-file.c +3 -3
@@ -68,12 +68,12 @@ int finalize_hashfile(struct hashfile *f, unsigned char *result,hashflush (f);if (f->skip_hash) -hashclr (f->buffer); +hashclr (f->buffer, the_repository->hash_algo);else the_hash_algo->final_fn (f->buffer, &f->ctx);if (result) -hashcpy (result, f->buffer); +hashcpy (result, f->buffer, the_repository->hash_algo);if (flags &CSUM_HASH_IN_STREAM )flush (f, f->buffer, the_hash_algo->rawsz);if (flags &CSUM_FSYNC )
@@ -237,5 +237,5 @@ int hashfile_checksum_valid(const unsigned char *data, size_t total_len) the_hash_algo->update_fn (&ctx, data, data_len); the_hash_algo->final_fn (got, &ctx); -return hasheq (got, data + data_len); +return hasheq (got, data + data_len, the_repository->hash_algo); }
hash-ll.h +13 -2
@@ -245,7 +245,7 @@ static inline int hash_algo_by_ptr(const struct git_hash_algo *p)const struct object_id *null_oid (void ); -static inline int hashcmp_algop (const unsigned char *sha1,const unsigned char *sha2,const struct git_hash_algo *algop) +static inline int hashcmp (const unsigned char *sha1,const unsigned char *sha2,const struct git_hash_algo *algop) {/* * Teach the compiler that there are only two possibilities of hash size
@@ -256,7 +256,7 @@ static inline int hashcmp_algop(const unsigned char *sha1, const unsigned char *return memcmp (sha1, sha2,GIT_SHA1_RAWSZ ); } -static inline int hasheq_algop (const unsigned char *sha1,const unsigned char *sha2,const struct git_hash_algo *algop) +static inline int hasheq (const unsigned char *sha1,const unsigned char *sha2,const struct git_hash_algo *algop) {/* * We write this here instead of deferring to hashcmp so that the
@@ -267,6 +267,17 @@ static inline int hasheq_algop(const unsigned char *sha1, const unsigned char *sreturn !memcmp (sha1, sha2,GIT_SHA1_RAWSZ ); } +static inline void hashcpy (unsigned char *sha_dst,const unsigned char *sha_src, +const struct git_hash_algo *algop) +{ +memcpy (sha_dst, sha_src, algop->rawsz); +} + +static inline void hashclr (unsigned char *hash,const struct git_hash_algo *algop) +{ +memset (hash,0 , algop->rawsz); +} +static inline void oidcpy (struct object_id *dst,const struct object_id *src) {memcpy (dst->hash, src->hash,GIT_MAX_RAWSZ );
hash-lookup.c +2 -1
@@ -112,7 +112,8 @@ int bsearch_hash(const unsigned char *hash, const uint32_t *fanout_nbo,while (lo < hi) {unsigned mi = lo + (hi - lo) /2 ; -int cmp =hashcmp (table + mi * stride, hash); +int cmp =hashcmp (table + mi * stride, hash, + the_repository->hash_algo);if (!cmp) {if (result)
hash.h +2 -22
@@ -6,11 +6,6 @@#define the_hash_algo the_repository->hash_algo -static inline int hashcmp (const unsigned char *sha1,const unsigned char *sha2) -{ -return hashcmp_algop (sha1, sha2, the_hash_algo); -} -static inline int oidcmp (const struct object_id *oid1,const struct object_id *oid2) {const struct git_hash_algo *algop;
@@ -18,12 +13,7 @@ static inline int oidcmp(const struct object_id *oid1, const struct object_id *o algop = the_hash_algo; else algop = &hash_algos[oid1->algo]; -return hashcmp_algop (oid1->hash, oid2->hash, algop); -} - -static inline int hasheq (const unsigned char *sha1,const unsigned char *sha2) -{ -return hasheq_algop (sha1, sha2, the_hash_algo); +return hashcmp (oid1->hash, oid2->hash, algop); }static inline int oideq (const struct object_id *oid1,const struct object_id *oid2)
@@ -33,7 +23,7 @@ static inline int oideq(const struct object_id *oid1, const struct object_id *oi algop = the_hash_algo; else algop = &hash_algos[oid1->algo]; -return hasheq_algop (oid1->hash, oid2->hash, algop); +return hasheq (oid1->hash, oid2->hash, algop); }static inline int is_null_oid (const struct object_id *oid)
@@ -41,11 +31,6 @@ static inline int is_null_oid(const struct object_id *oid)return oideq (oid,null_oid ()); } -static inline void hashcpy (unsigned char *sha_dst,const unsigned char *sha_src) -{ -memcpy (sha_dst, sha_src, the_hash_algo->rawsz); -} -/* Like oidcpy() but zero-pads the unused bytes in dst's hash array. */ static inline void oidcpy_with_padding (struct object_id *dst,const struct object_id *src)
@@ -62,11 +47,6 @@ static inline void oidcpy_with_padding(struct object_id *dst, dst->algo = src->algo; } -static inline void hashclr (unsigned char *hash) -{ -memset (hash,0 , the_hash_algo->rawsz); -} -static inline void oidclr (struct object_id *oid) {memset (oid->hash,0 ,GIT_MAX_RAWSZ );
http-walker.c +1 -1
@@ -485,7 +485,7 @@ static int fetch_object(struct walker *walker, unsigned char *hash)list_for_each (pos, head) { obj_req = list_entry(pos, struct object_request, node); -if (hasheq (obj_req->oid.hash, hash)) +if (hasheq (obj_req->oid.hash, hash, the_repository->hash_algo))break ; }if (!obj_req)
match-trees.c +1 -1
@@ -237,7 +237,7 @@ static int splice_tree(const struct object_id *oid1, const char *prefix, }else { rewrite_with = oid2; } -hashcpy (rewrite_here, rewrite_with->hash); +hashcpy (rewrite_here, rewrite_with->hash, the_repository->hash_algo); status =write_object_file (buf, sz,OBJ_TREE , result);free (buf);return status;
notes.c +1 -1
@@ -149,7 +149,7 @@ static struct leaf_node *note_tree_find(struct notes_tree *t,void **p =note_tree_search (t, &tree, &n, key_sha1);if (GET_PTR_TYPE (*p) ==PTR_TYPE_NOTE ) {struct leaf_node *l = (struct leaf_node *)CLR_PTR_TYPE (*p); -if (hasheq (key_sha1, l->key_oid.hash)) +if (hasheq (key_sha1, l->key_oid.hash, the_repository->hash_algo))return l; }return NULL ;
pack-bitmap-write.c +2 -2
@@ -790,7 +790,7 @@ static void write_hash_cache(struct hashfile *f,void bitmap_writer_set_checksum (struct bitmap_writer *writer,const unsigned char *sha1) { -hashcpy (writer->pack_checksum, sha1); +hashcpy (writer->pack_checksum, sha1, the_repository->hash_algo); } void bitmap_writer_finish(struct bitmap_writer *writer,
@@ -816,7 +816,7 @@ void bitmap_writer_finish(struct bitmap_writer *writer, header.version =htons (default_version); header.options =htons (flags | options); header.entry_count =htonl (writer->selected_nr); -hashcpy (header.checksum, writer->pack_checksum); +hashcpy (header.checksum, writer->pack_checksum, the_repository->hash_algo);hashwrite (f, &header,sizeof (header) -GIT_MAX_RAWSZ + the_hash_algo->rawsz);dump_bitmap (f, writer->commits);
pack-bitmap.c +2 -1
@@ -367,7 +367,8 @@ static int open_midx_bitmap_1(struct bitmap_index *bitmap_git,if (load_bitmap_header (bitmap_git) <0 ) goto cleanup; -if (!hasheq (get_midx_checksum (bitmap_git->midx), bitmap_git->checksum)) { +if (!hasheq (get_midx_checksum (bitmap_git->midx), bitmap_git->checksum, + the_repository->hash_algo)) {error (_ ("checksum doesn't match in MIDX and bitmap" )); goto cleanup; }
pack-check.c +3 -2
@@ -78,10 +78,11 @@ static int verify_packfile(struct repository *r, }while (offset < pack_sig_ofs); r->hash_algo->final_fn (hash, &ctx); pack_sig =use_pack (p, w_curs, pack_sig_ofs,NULL ); -if (!hasheq (hash, pack_sig)) +if (!hasheq (hash, pack_sig, the_repository->hash_algo)) err =error ("%s pack checksum mismatch" , p->pack_name); -if (!hasheq (index_base + index_size - r->hash_algo->hexsz, pack_sig)) +if (!hasheq (index_base + index_size - r->hash_algo->hexsz, pack_sig, + the_repository->hash_algo)) err =error ("%s pack checksum does not match its index" , p->pack_name);unuse_pack (w_curs);
pack-write.c +2 -1
@@ -428,7 +428,8 @@ void fixup_pack_header_footer(int pack_fd,if (partial_pack_offset ==0 ) {unsigned char hash[GIT_MAX_RAWSZ ]; the_hash_algo->final_fn (hash, &old_hash_ctx); -if (!hasheq (hash, partial_pack_hash)) +if (!hasheq (hash, partial_pack_hash, + the_repository->hash_algo))die ("Unexpected checksum for %s " "(disk corruption?)" , pack_name);/*
packfile.c +4 -4
@@ -242,7 +242,7 @@ struct packed_git *parse_pack_index(unsigned char *sha1, const char *idx_path)struct packed_git *p =alloc_packed_git (alloc);memcpy (p->pack_name, path, alloc);/* includes NUL */ -hashcpy (p->hash, sha1); +hashcpy (p->hash, sha1, the_repository->hash_algo);if (check_packed_git_idx (idx_path, p)) {free (p);return NULL ;
@@ -596,7 +596,7 @@ static int open_packed_git_1(struct packed_git *p)if (read_result != hashsz)return error ("packfile %s signature is unavailable" , p->pack_name); idx_hash = ((unsigned char *)p->index_data) + p->index_size - hashsz *2 ; -if (!hasheq (hash, idx_hash)) +if (!hasheq (hash, idx_hash, the_repository->hash_algo))return error ("packfile %s does not match index" , p->pack_name);return 0 ; }
@@ -751,7 +751,7 @@ struct packed_git *add_packed_git(const char *path, size_t path_len, int local) p->mtime = st.st_mtime;if (path_len < the_hash_algo->hexsz ||get_hash_hex (path + path_len - the_hash_algo->hexsz, p->hash)) -hashclr (p->hash); +hashclr (p->hash, the_repository->hash_algo);return p; }
@@ -1971,7 +1971,7 @@ off_t find_pack_entry_one(const unsigned char *sha1,return 0 ; } -hashcpy (oid.hash, sha1); +hashcpy (oid.hash, sha1, the_repository->hash_algo);if (bsearch_pack (&oid, p, &result))return nth_packed_object_offset (p, result);return 0 ;
read-cache.c +4 -4
@@ -1735,7 +1735,7 @@ static int verify_hdr(const struct cache_header *hdr, unsigned long size) the_hash_algo->init_fn (&c); the_hash_algo->update_fn (&c, hdr, size - the_hash_algo->rawsz); the_hash_algo->final_fn (hash, &c); -if (!hasheq (hash, start)) +if (!hasheq (hash, start, the_repository->hash_algo))return error (_ ("bad index file sha1 signature" ));return 0 ; }
@@ -2641,7 +2641,7 @@ static void copy_cache_entry_to_ondisk(struct ondisk_cache_entry *ondisk, ondisk->uid =htonl (ce->ce_stat_data.sd_uid); ondisk->gid =htonl (ce->ce_stat_data.sd_gid); ondisk->size =htonl (ce->ce_stat_data.sd_size); -hashcpy (ondisk->data, ce->oid.hash); +hashcpy (ondisk->data, ce->oid.hash, the_repository->hash_algo); flags = ce->ce_flags & ~CE_NAMEMASK ; flags |= (ce_namelen (ce) >=CE_NAMEMASK ?CE_NAMEMASK :ce_namelen (ce));
@@ -2730,7 +2730,7 @@ static int verify_index_from(const struct index_state *istate, const char *path)if (n != the_hash_algo->rawsz) goto out; -if (!hasheq (istate->oid.hash, hash)) +if (!hasheq (istate->oid.hash, hash, the_repository->hash_algo)) goto out;close (fd);
@@ -3603,7 +3603,7 @@ static size_t read_eoie_extension(const char *mmap, size_t mmap_size) src_offset += extsize; } the_hash_algo->final_fn (hash, &c); -if (!hasheq (hash, (const unsigned char *)index)) +if (!hasheq (hash, (const unsigned char *)index, the_repository->hash_algo))return 0 ;/* Validate that the extension offsets returned us back to the eoie extension. */
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 3/20 hash: require hash algorithm in `oidread()` and `oidclr()`
In reply to Patrick SteinhardtBoth oidread() and oidclr() use the_repository to derive the hash function that shall be used. Require callers to pass in the hash algorithm to get rid of this implicit dependency.
--- apply.c | 2 +- blame.c | 2 +- builtin/am.c | 8 +++---- builtin/fast-export.c | 2 +- builtin/fast-import.c | 39 +++++++++++++++++--------------- builtin/fetch-pack.c | 4 ++-- builtin/index-pack.c | 5 ++-- builtin/log.c | 2 +- builtin/merge.c | 4 ++-- builtin/notes.c | 2 +- builtin/pack-objects.c | 3 ++- builtin/pack-redundant.c | 2 +- builtin/patch-id.c | 6 ++--- builtin/pull.c | 6 ++--- builtin/receive-pack.c | 2 +- builtin/replace.c | 2 +- builtin/rm.c | 2 +- builtin/tag.c | 2 +- builtin/unpack-objects.c | 6 ++--- builtin/update-ref.c | 8 +++---- cache-tree.c | 3 ++- commit-graph.c | 17 +++++++++----- diff-lib.c | 4 ++-- diff.c | 6 ++--- dir.c | 6 ++--- hash-ll.h | 14 ++++++++++++ hash.h | 17 -------------- http-push.c | 2 +- http-walker.c | 2 +- match-trees.c | 2 +- midx.c | 3 ++- notes-merge.c | 6 ++--- notes.c | 8 +++---- object-file-convert.c | 2 +- object-file.c | 4 ++-- packfile.c | 10 ++++---- read-cache.c | 8 ++++--- refs.c | 6 ++--- refs/files-backend.c | 6 ++--- refs/packed-backend.c | 6 ++--- refs/reftable-backend.c | 37 ++++++++++++++++++------------ remote.c | 8 +++---- resolve-undo.c | 3 ++- sequencer.c | 4 ++-- split-index.c | 2 +- submodule-config.c | 2 +- t/helper/test-submodule-config.c | 2 +- tree-walk.c | 4 ++-- 48 files changed, 163 insertions(+), 140 deletions(-)
apply.c +1 -1
@@ -3680,7 +3680,7 @@ static int try_threeway(struct apply_state *state,if (status) { patch->conflicted_threeway =1 ;if (patch->is_new) -oidclr (&patch->threeway_stage[0 ]); +oidclr (&patch->threeway_stage[0 ], the_repository->hash_algo);else oidcpy (&patch->threeway_stage[0 ], &pre_oid);oidcpy (&patch->threeway_stage[1 ], &our_oid);
blame.c +1 -1
@@ -1246,7 +1246,7 @@ static int fill_blob_sha1_and_mode(struct repository *r, goto error_out;return 0 ; error_out: -oidclr (&origin->blob_oid); +oidclr (&origin->blob_oid, the_repository->hash_algo); origin->mode =S_IFINVALID ;return -1 ; }
builtin/am.c +4 -4
@@ -408,7 +408,7 @@ static void am_load(struct am_state *state)read_commit_msg (state);if (read_state_file (&sb, state,"original-commit" ,1 ) <0 ) -oidclr (&state->orig_commit); +oidclr (&state->orig_commit, the_repository->hash_algo);else if (get_oid_hex (sb.buf, &state->orig_commit) <0 )die (_ ("could not parse %s" ),am_path (state,"original-commit" ));
@@ -1121,7 +1121,7 @@ static void am_next(struct am_state *state)unlink (am_path (state,"author-script" ));unlink (am_path (state,"final-commit" )); -oidclr (&state->orig_commit); +oidclr (&state->orig_commit, the_repository->hash_algo);unlink (am_path (state,"original-commit" ));refs_delete_ref (get_main_ref_store (the_repository),NULL ,"REBASE_HEAD" ,NULL ,REF_NO_DEREF );
@@ -2151,11 +2151,11 @@ static int safe_to_abort(const struct am_state *state)if (get_oid_hex (sb.buf, &abort_safety))die (_ ("could not parse %s" ),am_path (state,"abort-safety" )); }else -oidclr (&abort_safety); +oidclr (&abort_safety, the_repository->hash_algo);strbuf_release (&sb);if (repo_get_oid (the_repository,"HEAD" , &head)) -oidclr (&head); +oidclr (&head, the_repository->hash_algo);if (oideq (&head, &abort_safety))return 1 ;
builtin/fast-export.c +1 -1
@@ -415,7 +415,7 @@ static char *generate_fake_oid(void)struct object_id oid;char *hex =xmallocz (GIT_MAX_HEXSZ ); -oidclr (&oid); +oidclr (&oid, the_repository->hash_algo);put_be32 (oid.hash + hashsz -4 , counter++);return oid_to_hex_r (hex, &oid); }
builtin/fast-import.c +21 -18
@@ -1279,8 +1279,10 @@ static void load_tree(struct tree_entry *root) e->versions[0 ].mode = e->versions[1 ].mode; e->name =to_atom (c,strlen (c)); c += e->name->str_len +1 ; -oidread (&e->versions[0 ].oid, (unsigned char *)c); -oidread (&e->versions[1 ].oid, (unsigned char *)c); +oidread (&e->versions[0 ].oid, (unsigned char *)c, + the_repository->hash_algo); +oidread (&e->versions[1 ].oid, (unsigned char *)c, + the_repository->hash_algo); c += the_hash_algo->rawsz; }free (buf);
@@ -1386,7 +1388,7 @@ static void tree_content_replace( {if (!S_ISDIR (mode))die ("Root cannot be a non-directory" ); -oidclr (&root->versions[0 ].oid); +oidclr (&root->versions[0 ].oid, the_repository->hash_algo);oidcpy (&root->versions[1 ].oid, oid);if (root->tree)release_tree_content_recursive (root->tree);
@@ -1445,7 +1447,7 @@ static int tree_content_set(if (S_ISDIR (e->versions[0 ].mode)) e->versions[0 ].mode |=NO_DELTA ; -oidclr (&root->versions[1 ].oid); +oidclr (&root->versions[1 ].oid, the_repository->hash_algo);return 1 ; }if (!S_ISDIR (e->versions[1 ].mode)) {
@@ -1455,7 +1457,7 @@ static int tree_content_set(if (!e->tree)load_tree (e);if (tree_content_set (e, slash1 +1 , oid, mode, subtree)) { -oidclr (&root->versions[1 ].oid); +oidclr (&root->versions[1 ].oid, the_repository->hash_algo);return 1 ; }return 0 ;
@@ -1467,7 +1469,7 @@ static int tree_content_set( e =new_tree_entry (); e->name =to_atom (p, n); e->versions[0 ].mode =0 ; -oidclr (&e->versions[0 ].oid); +oidclr (&e->versions[0 ].oid, the_repository->hash_algo); t->entries[t->entry_count++] = e;if (*slash1) { e->tree =new_tree_content (8 );
@@ -1478,7 +1480,7 @@ static int tree_content_set( e->versions[1 ].mode = mode;oidcpy (&e->versions[1 ].oid, oid); } -oidclr (&root->versions[1 ].oid); +oidclr (&root->versions[1 ].oid, the_repository->hash_algo);return 1 ; }
@@ -1523,7 +1525,8 @@ static int tree_content_remove(if (tree_content_remove (e, slash1 +1 , backup_leaf,0 )) {for (n =0 ; n < e->tree->entry_count; n++) {if (e->tree->entries[n]->versions[1 ].mode) { -oidclr (&root->versions[1 ].oid); +oidclr (&root->versions[1 ].oid, + the_repository->hash_algo);return 1 ; } }
@@ -1542,8 +1545,8 @@ static int tree_content_remove(release_tree_content_recursive (e->tree); e->tree =NULL ; e->versions[1 ].mode =0 ; -oidclr (&e->versions[1 ].oid); -oidclr (&root->versions[1 ].oid); +oidclr (&e->versions[1 ].oid, the_repository->hash_algo); +oidclr (&root->versions[1 ].oid, the_repository->hash_algo);return 1 ; }
@@ -1609,7 +1612,7 @@ static int update_branch(struct branch *b)return 0 ; }if (refs_read_ref (get_main_ref_store (the_repository), b->name, &old_oid)) -oidclr (&old_oid); +oidclr (&old_oid, the_repository->hash_algo);if (!force_update && !is_null_oid (&old_oid)) {struct commit *old_cmit, *new_cmit;int ret;
@@ -2550,8 +2553,8 @@ static void note_change_n(const char *p, struct branch *b, unsigned char *old_fastatic void file_change_deleteall (struct branch *b) {release_tree_content_recursive (b->branch_tree.tree); -oidclr (&b->branch_tree.versions[0 ].oid); -oidclr (&b->branch_tree.versions[1 ].oid); +oidclr (&b->branch_tree.versions[0 ].oid, the_repository->hash_algo); +oidclr (&b->branch_tree.versions[1 ].oid, the_repository->hash_algo);load_tree (&b->branch_tree); b->num_notes =0 ; }
@@ -2570,8 +2573,8 @@ static void parse_from_commit(struct branch *b, char *buf, unsigned long size)static void parse_from_existing (struct branch *b) {if (is_null_oid (&b->oid)) { -oidclr (&b->branch_tree.versions[0 ].oid); -oidclr (&b->branch_tree.versions[1 ].oid); +oidclr (&b->branch_tree.versions[0 ].oid, the_repository->hash_algo); +oidclr (&b->branch_tree.versions[1 ].oid, the_repository->hash_algo); }else {unsigned long size;char *buf;
@@ -2894,9 +2897,9 @@ static void parse_reset_branch(const char *arg) b =lookup_branch (arg);if (b) { -oidclr (&b->oid); -oidclr (&b->branch_tree.versions[0 ].oid); -oidclr (&b->branch_tree.versions[1 ].oid); +oidclr (&b->oid, the_repository->hash_algo); +oidclr (&b->branch_tree.versions[0 ].oid, the_repository->hash_algo); +oidclr (&b->branch_tree.versions[1 ].oid, the_repository->hash_algo);if (b->branch_tree.tree) {release_tree_content_recursive (b->branch_tree.tree); b->branch_tree.tree =NULL ;
builtin/fetch-pack.c +2 -2
@@ -29,11 +29,11 @@ static void add_sought_entry(struct ref ***sought, int *nr, int *alloc, ;/* <oid>, leave oid as name */ }else {/* <ref>, clear cruft from oid */ -oidclr (&oid); +oidclr (&oid, the_repository->hash_algo); } }else {/* <ref>, clear cruft from get_oid_hex */ -oidclr (&oid); +oidclr (&oid, the_repository->hash_algo); } ref =alloc_ref (name);
builtin/index-pack.c +3 -2
@@ -528,7 +528,8 @@ static void *unpack_raw_entry(struct object_entry *obj,switch (obj->type) {case OBJ_REF_DELTA : -oidread (ref_oid,fill (the_hash_algo->rawsz)); +oidread (ref_oid,fill (the_hash_algo->rawsz), + the_repository->hash_algo);use (the_hash_algo->rawsz);break ;case OBJ_OFS_DELTA :
@@ -1372,7 +1373,7 @@ static struct object_entry *append_obj_to_pack(struct hashfile *f, obj[1 ].idx.offset +=write_compressed (f, buf, size); obj[0 ].idx.crc32 =crc32_end (f);hashflush (f); -oidread (&obj->idx.oid, sha1); +oidread (&obj->idx.oid, sha1, the_repository->hash_algo);return obj; }
builtin/log.c +1 -1
@@ -1938,7 +1938,7 @@ static void print_bases(struct base_tree_info *bases, FILE *file)free (bases->patch_id); bases->nr_patch_id =0 ; bases->alloc_patch_id =0 ; -oidclr (&bases->base_commit); +oidclr (&bases->base_commit, the_repository->hash_algo); } static const char *diff_title(struct strbuf *sb,
builtin/merge.c +2 -2
@@ -494,7 +494,7 @@ static void merge_name(const char *remote, struct strbuf *msg)strbuf_branchname (&bname, remote,0 ); remote = bname.buf; -oidclr (&branch_head); +oidclr (&branch_head, the_repository->hash_algo); remote_head =get_merge_parent (remote);if (!remote_head)die (_ ("'%s' does not point to a commit" ), remote);
@@ -1690,7 +1690,7 @@ int cmd_merge(int argc, const char **argv, const char *prefix)* index and working tree polluted. */ if (save_state (&stash)) -oidclr (&stash); +oidclr (&stash, the_repository->hash_algo);for (i =0 ; i < use_strategies_nr; i++) {int ret, cnt;
builtin/notes.c +1 -1
@@ -828,7 +828,7 @@ static int merge_commit(struct notes_merge_options *o)if (partial->parents)oidcpy (&parent_oid, &partial->parents->item->object.oid);else -oidclr (&parent_oid); +oidclr (&parent_oid, the_repository->hash_algo);CALLOC_ARRAY (t,1 );init_notes (t,"NOTES_MERGE_PARTIAL" , combine_notes_overwrite,0 );
builtin/pack-objects.c +2 -1
@@ -2078,7 +2078,8 @@ static void check_object(struct object_entry *entry, uint32_t object_index)oidread (&base_ref,use_pack (p, &w_curs, entry->in_pack_offset + used, -NULL )); +NULL ), + the_repository->hash_algo); have_base =1 ; } entry->in_pack_header_size = used + the_hash_algo->rawsz;
builtin/pack-redundant.c +1 -1
@@ -100,7 +100,7 @@ static inline struct llist_item *llist_insert(struct llist *list, const unsigned char *oid) {struct llist_item *new_item =llist_item_get (); -oidread (&new_item->oid, oid); +oidread (&new_item->oid, oid, the_repository->hash_algo); new_item->next =NULL ;if (after) {
builtin/patch-id.c +3 -3
@@ -70,7 +70,7 @@ static int get_one_patchid(struct object_id *next_oid, struct object_id *result,git_hash_ctx ctx; the_hash_algo->init_fn (&ctx); -oidclr (result); +oidclr (result, the_repository->hash_algo);while (strbuf_getwholeline (line_buf, stdin,'\n' ) !=EOF ) {char *line = line_buf->buf;
@@ -166,7 +166,7 @@ static int get_one_patchid(struct object_id *next_oid, struct object_id *result, }if (!found_next) -oidclr (next_oid); +oidclr (next_oid, the_repository->hash_algo);flush_one_hunk (result, &ctx);
@@ -179,7 +179,7 @@ static void generate_id_list(int stable, int verbatim)int patchlen;struct strbuf line_buf =STRBUF_INIT ; -oidclr (&oid); +oidclr (&oid, the_repository->hash_algo);while (!feof (stdin)) { patchlen =get_one_patchid (&n, &result, &line_buf, stable, verbatim);flush_current_id (patchlen, &oid, &result);
builtin/pull.c +3 -3
@@ -1038,7 +1038,7 @@ int cmd_pull(int argc, const char **argv, const char *prefix)die_conclude_merge ();if (repo_get_oid (the_repository,"HEAD" , &orig_head)) -oidclr (&orig_head); +oidclr (&orig_head, the_repository->hash_algo);if (opt_rebase) {if (opt_autostash ==-1 )
@@ -1053,7 +1053,7 @@ int cmd_pull(int argc, const char **argv, const char *prefix) _("Please commit or stash them."), 1, 0);if (get_rebase_fork_point (&rebase_fork_point, repo, *refspecs)) -oidclr (&rebase_fork_point); +oidclr (&rebase_fork_point, the_repository->hash_algo); }if (run_fetch (repo, refspecs))
@@ -1063,7 +1063,7 @@ int cmd_pull(int argc, const char **argv, const char *prefix)return 0 ;if (repo_get_oid (the_repository,"HEAD" , &curr_head)) -oidclr (&curr_head); +oidclr (&curr_head, the_repository->hash_algo);if (!is_null_oid (&orig_head) && !is_null_oid (&curr_head) && !oideq (&orig_head, &curr_head)) {
builtin/receive-pack.c +1 -1
@@ -741,7 +741,7 @@ static void prepare_push_cert_sha1(struct child_process *proc) already_done =1 ;if (write_object_file (push_cert.buf, push_cert.len,OBJ_BLOB , &push_cert_oid)) -oidclr (&push_cert_oid); +oidclr (&push_cert_oid, the_repository->hash_algo);memset (&sigcheck,'\0' ,sizeof (sigcheck));
builtin/replace.c +1 -1
@@ -167,7 +167,7 @@ static int check_ref_valid(struct object_id *object,return error (_ ("'%s' is not a valid ref name" ), ref->buf);if (refs_read_ref (get_main_ref_store (the_repository), ref->buf, prev)) -oidclr (prev); +oidclr (prev, the_repository->hash_algo);else if (!force)return error (_ ("replace ref '%s' already exists" ), ref->buf);return 0 ;
builtin/rm.c +1 -1
@@ -377,7 +377,7 @@ int cmd_rm(int argc, const char **argv, const char *prefix)if (!force) {struct object_id oid;if (repo_get_oid (the_repository,"HEAD" , &oid)) -oidclr (&oid); +oidclr (&oid, the_repository->hash_algo);if (check_local_mod (&oid, index_only))exit (1 ); }
builtin/tag.c +1 -1
@@ -650,7 +650,7 @@ int cmd_tag(int argc, const char **argv, const char *prefix)die (_ ("'%s' is not a valid tag name." ), tag);if (refs_read_ref (get_main_ref_store (the_repository), ref.buf, &prev)) -oidclr (&prev); +oidclr (&prev, the_repository->hash_algo);else if (!force)die (_ ("tag '%s' already exists" ), tag);
builtin/unpack-objects.c +3 -3
@@ -439,7 +439,7 @@ static void unpack_delta_entry(enum object_type type, unsigned long delta_size,struct object_id base_oid;if (type ==OBJ_REF_DELTA ) { -oidread (&base_oid,fill (the_hash_algo->rawsz)); +oidread (&base_oid,fill (the_hash_algo->rawsz), the_repository->hash_algo);use (the_hash_algo->rawsz); delta_data =get_data (delta_size);if (!delta_data)
@@ -451,7 +451,7 @@ static void unpack_delta_entry(enum object_type type, unsigned long delta_size,return ;/* we are done */ else {/* cannot resolve yet --- queue it */ -oidclr (&obj_list[nr].oid); +oidclr (&obj_list[nr].oid, the_repository->hash_algo);add_delta_to_list (nr, &base_oid,0 , delta_data, delta_size);return ; }
@@ -500,7 +500,7 @@ static void unpack_delta_entry(enum object_type type, unsigned long delta_size,* The delta base object is itself a delta that * has not been resolved yet. */ -oidclr (&obj_list[nr].oid); +oidclr (&obj_list[nr].oid, the_repository->hash_algo);add_delta_to_list (nr,null_oid (), base_offset, delta_data, delta_size);return ;
builtin/update-ref.c +4 -4
@@ -122,7 +122,7 @@ static int parse_next_oid(const char **next, const char *end, goto invalid; }else {/* Without -z, an empty value means all zeros: */ -oidclr (oid); +oidclr (oid, the_repository->hash_algo); } }else {/* With -z, read the next NUL-terminated line */
@@ -142,7 +142,7 @@ static int parse_next_oid(const char **next, const char *end,/* With -z, treat an empty value as all zeros: */ warning ("%s %s: missing <new-oid>, treating as zero" , command, refname); -oidclr (oid); +oidclr (oid, the_repository->hash_algo); }else {/* * With -z, an empty non-required value means
@@ -291,7 +291,7 @@ static void parse_cmd_verify(struct ref_transaction *transaction,if (parse_next_oid (&next, end, &old_oid,"verify" , refname,PARSE_SHA1_OLD )) -oidclr (&old_oid); +oidclr (&old_oid, the_repository->hash_algo);if (*next != line_termination)die ("verify %s: extra input: %s" , refname, next);
@@ -564,7 +564,7 @@ int cmd_update_ref(int argc, const char **argv, const char *prefix)* The empty string implies that the reference * must not already exist: */ -oidclr (&oldoid); +oidclr (&oldoid, the_repository->hash_algo); else if (repo_get_oid(the_repository, oldval, &oldoid)) die("%s: not a valid old SHA1", oldval); }
cache-tree.c +2 -1
@@ -578,7 +578,8 @@ static struct cache_tree *read_one(const char **buffer, unsigned long *size_p)if (0 <= it->entry_count) {if (size < rawsz) goto free_return; -oidread (&it->oid, (const unsigned char *)buf); +oidread (&it->oid, (const unsigned char *)buf, + the_repository->hash_algo); buf += rawsz; size -= rawsz; }
commit-graph.c +11 -6
@@ -475,7 +475,8 @@ struct commit_graph *parse_commit_graph(struct repo_settings *s,FREE_AND_NULL (graph->bloom_filter_settings); } -oidread (&graph->oid, graph->data + graph->data_len - graph->hash_len); +oidread (&graph->oid, graph->data + graph->data_len - graph->hash_len, + the_repository->hash_algo);free_chunkfile (cf);return graph;
@@ -838,7 +839,8 @@ static void load_oid_from_graph(struct commit_graph *g, lex_index = pos - g->num_commits_in_base; -oidread (oid, g->chunk_oid_lookup +st_mult (g->hash_len, lex_index)); +oidread (oid, g->chunk_oid_lookup +st_mult (g->hash_len, lex_index), + the_repository->hash_algo); } static struct commit_list **insert_parent_or_die(struct repository *r,
@@ -1080,7 +1082,7 @@ static struct tree *load_tree_for_commit(struct repository *r, commit_data = g->chunk_commit_data +st_mult (GRAPH_DATA_WIDTH , graph_pos - g->num_commits_in_base); -oidread (&oid, commit_data); +oidread (&oid, commit_data, the_repository->hash_algo);set_commit_tree (c,lookup_tree (r, &oid));return c->maybe_tree;
@@ -2556,7 +2558,8 @@ int write_commit_graph(struct object_directory *odb,struct commit_graph *g = ctx->r->objects->commit_graph;for (i =0 ; i < g->num_commits; i++) {struct object_id oid; -oidread (&oid, g->chunk_oid_lookup +st_mult (g->hash_len, i)); +oidread (&oid, g->chunk_oid_lookup +st_mult (g->hash_len, i), + the_repository->hash_algo);oid_array_append (&ctx->oids, &oid); } }
@@ -2675,7 +2678,8 @@ static int verify_one_commit_graph(struct repository *r,for (i =0 ; i < g->num_commits; i++) {struct commit *graph_commit; -oidread (&cur_oid, g->chunk_oid_lookup +st_mult (g->hash_len, i)); +oidread (&cur_oid, g->chunk_oid_lookup +st_mult (g->hash_len, i), + the_repository->hash_algo);if (i &&oidcmp (&prev_oid, &cur_oid) >=0 ) graph_report(_("commit-graph has incorrect OID order: %s then %s"),
@@ -2719,7 +2723,8 @@ static int verify_one_commit_graph(struct repository *r,timestamp_t generation;display_progress (progress, ++(*seen)); -oidread (&cur_oid, g->chunk_oid_lookup +st_mult (g->hash_len, i)); +oidread (&cur_oid, g->chunk_oid_lookup +st_mult (g->hash_len, i), + the_repository->hash_algo); graph_commit =lookup_commit (r, &cur_oid); odb_commit = (struct commit *)create_object (r, &cur_oid,alloc_commit_node (r));
diff-lib.c +2 -2
@@ -160,7 +160,7 @@ void run_diff_files(struct rev_info *revs, unsigned int option) dpath->next =NULL ;memcpy (dpath->path, ce->name, path_len); dpath->path[path_len] ='\0' ; -oidclr (&dpath->oid); +oidclr (&dpath->oid, the_repository->hash_algo);memset (&(dpath->parent[0 ]),0 ,sizeof (struct combine_diff_parent )*5 );
@@ -412,7 +412,7 @@ static int show_modified(struct rev_info *revs,memcpy (p->path, new_entry->name, pathlen); p->path[pathlen] =0 ; p->mode = mode; -oidclr (&p->oid); +oidclr (&p->oid, the_repository->hash_algo);memset (p->parent,0 ,2 *sizeof (struct combine_diff_parent )); p->parent[0 ].status =DIFF_STATUS_MODIFIED ; p->parent[0 ].mode = new_entry->ce_mode;
diff.c +3 -3
@@ -4567,7 +4567,7 @@ static void diff_fill_oid_info(struct diff_filespec *one, struct index_state *isif (!one->oid_valid) {struct stat st;if (one->is_stdin) { -oidclr (&one->oid); +oidclr (&one->oid, the_repository->hash_algo);return ; }if (lstat (one->path, &st) <0 )
@@ -4577,7 +4577,7 @@ static void diff_fill_oid_info(struct diff_filespec *one, struct index_state *is } }else -oidclr (&one->oid); +oidclr (&one->oid, the_repository->hash_algo); }static void strip_prefix (int prefix_length,const char **namep,const char **otherp)
@@ -6404,7 +6404,7 @@ static int diff_get_patch_id(struct diff_options *options, struct object_id *oid the_hash_algo->init_fn (&ctx);memset (&data,0 ,sizeof (struct patch_id_t )); data.ctx = &ctx; -oidclr (oid); +oidclr (oid, the_repository->hash_algo);for (i =0 ; i < q->nr; i++) {xpparam_t xpp;
dir.c +3 -3
@@ -1687,7 +1687,7 @@ static void prep_exclude(struct dir_struct *dir, }/* Try to read per-directory file */ -oidclr (&oid_stat.oid); +oidclr (&oid_stat.oid, the_repository->hash_algo); oid_stat.valid =0 ; if (dir->exclude_per_dir &&/*
@@ -3794,7 +3794,7 @@ static void read_oid(size_t pos, void *cb) rd->data = rd->end +1 ;return ; } -oidread (&ud->exclude_oid, rd->data); +oidread (&ud->exclude_oid, rd->data, the_repository->hash_algo); rd->data += the_hash_algo->rawsz; }
@@ -3802,7 +3802,7 @@ static void load_oid_stat(struct oid_stat *oid_stat, const unsigned char *data, const unsigned char *sha1) {stat_data_from_disk (&oid_stat->stat, data); -oidread (&oid_stat->oid, sha1); +oidread (&oid_stat->oid, sha1, the_repository->hash_algo); oid_stat->valid =1 ; }
hash-ll.h +14 -0
@@ -284,6 +284,20 @@ static inline void oidcpy(struct object_id *dst, const struct object_id *src) dst->algo = src->algo; } +static inline void oidread (struct object_id *oid,const unsigned char *hash, +const struct git_hash_algo *algop) +{ +memcpy (oid->hash, hash, algop->rawsz); + oid->algo =hash_algo_by_ptr (algop); +} + +static inline void oidclr (struct object_id *oid, +const struct git_hash_algo *algop) +{ +memset (oid->hash,0 ,GIT_MAX_RAWSZ ); + oid->algo =hash_algo_by_ptr (algop); +} +static inline struct object_id *oiddup (const struct object_id *src) {struct object_id *dst =xmalloc (sizeof (struct object_id ));
hash.h +0 -17
@@ -47,23 +47,6 @@ static inline void oidcpy_with_padding(struct object_id *dst, dst->algo = src->algo; } -static inline void oidclr (struct object_id *oid) -{ -memset (oid->hash,0 ,GIT_MAX_RAWSZ ); - oid->algo =hash_algo_by_ptr (the_hash_algo); -} - -static inline void oidread_algop (struct object_id *oid,const unsigned char *hash,const struct git_hash_algo *algop) -{ -memcpy (oid->hash, hash, algop->rawsz); - oid->algo =hash_algo_by_ptr (algop); -} - -static inline void oidread (struct object_id *oid,const unsigned char *hash) -{ -oidread_algop (oid, hash, the_hash_algo); -} -static inline int is_empty_blob_oid (const struct object_id *oid) {return oideq (oid, the_hash_algo->empty_blob);
http-push.c +1 -1
@@ -1552,7 +1552,7 @@ static void fetch_symref(const char *path, char **symref, struct object_id *oid)free (url);FREE_AND_NULL (*symref); -oidclr (oid); +oidclr (oid, the_repository->hash_algo);if (buffer.len ==0 )return ;
http-walker.c +1 -1
@@ -152,7 +152,7 @@ static void prefetch(struct walker *walker, unsigned char *sha1) newreq =xmalloc (sizeof (*newreq)); newreq->walker = walker; -oidread (&newreq->oid, sha1); +oidread (&newreq->oid, sha1, the_repository->hash_algo); newreq->repo = data->alt; newreq->state =WAITING ; newreq->req =NULL ;
match-trees.c +1 -1
@@ -229,7 +229,7 @@ static int splice_tree(const struct object_id *oid1, const char *prefix, oid_to_hex(oid1));if (*subpath) {struct object_id tree_oid; -oidread (&tree_oid, rewrite_here); +oidread (&tree_oid, rewrite_here, the_repository->hash_algo); status =splice_tree (&tree_oid, subpath, oid2, &subtree);if (status)return status;
midx.c +2 -1
@@ -304,7 +304,8 @@ struct object_id *nth_midxed_object_oid(struct object_id *oid,if (n >= m->num_objects)return NULL ; -oidread (oid, m->chunk_oid_lookup +st_mult (m->hash_len, n)); +oidread (oid, m->chunk_oid_lookup +st_mult (m->hash_len, n), + the_repository->hash_algo);return oid; }
notes-merge.c +3 -3
@@ -240,7 +240,7 @@ static void diff_tree_local(struct notes_merge_options *o,* (will be overwritten by following addition) */ if (oideq (&mp->local, &uninitialized)) -oidclr (&mp->local); +oidclr (&mp->local, the_repository->hash_algo); }else if (is_null_oid (&p->one->oid)) {/* addition */ /* * Either this is a true addition (1), or it is part
@@ -556,7 +556,7 @@ int notes_merge(struct notes_merge_options *o,assert (o->local_ref && o->remote_ref);assert (!strcmp (o->local_ref, local_tree->ref)); -oidclr (result_oid); +oidclr (result_oid, the_repository->hash_algo);trace_printf ("notes_merge(o->local_ref = %s, o->remote_ref = %s)\n" , o->local_ref, o->remote_ref);
@@ -579,7 +579,7 @@ int notes_merge(struct notes_merge_options *o,* unborn ref, perform the merge using an empty notes tree. */ if (!check_refname_format (o->remote_ref,0 )) { -oidclr (&remote_oid); +oidclr (&remote_oid, the_repository->hash_algo); remote =NULL ; }else { die("Failed to resolve remote notes ref '%s'",
notes.c +4 -4
@@ -353,7 +353,7 @@ static void add_non_note(struct notes_tree *t, char *path, n->next =NULL ; n->path = path; n->mode = mode; -oidread (&n->oid, sha1); +oidread (&n->oid, sha1, the_repository->hash_algo); t->prev_non_note = n;if (!t->first_non_note) {
@@ -1036,7 +1036,7 @@ void init_notes(struct notes_tree *t, const char *notes_ref,die ("Failed to read notes tree referenced by %s (%s)" , notes_ref,oid_to_hex (&object_oid)); -oidclr (&root_tree.key_oid); +oidclr (&root_tree.key_oid, the_repository->hash_algo);oidcpy (&root_tree.val_oid, &oid);load_subtree (t, &root_tree, t->root,0 ); }
@@ -1146,8 +1146,8 @@ int remove_note(struct notes_tree *t, const unsigned char *object_sha1)if (!t) t = &default_notes_tree;assert (t->initialized); -oidread (&l.key_oid, object_sha1); -oidclr (&l.val_oid); +oidread (&l.key_oid, object_sha1, the_repository->hash_algo); +oidclr (&l.val_oid, the_repository->hash_algo);note_tree_remove (t, t->root,0 , &l);if (is_null_oid (&l.val_oid))/* no note was removed */ return 1 ;
object-file-convert.c +1 -1
@@ -56,7 +56,7 @@ static int decode_tree_entry_raw(struct object_id *oid, const char **path,return -1 ; *len =strlen (*path) +1 ; -oidread_algop (oid, (const unsigned char *)*path + *len, algo); +oidread (oid, (const unsigned char *)*path + *len, algo);return 0 ; }
object-file.c +2 -2
@@ -1446,7 +1446,7 @@ static int loose_object_info(struct repository *r,int allow_unknown = flags &OBJECT_INFO_ALLOW_UNKNOWN_TYPE ;if (oi->delta_base_oid) -oidclr (oi->delta_base_oid); +oidclr (oi->delta_base_oid, the_repository->hash_algo);/* * If we don't care about type or size, then we don't
@@ -1580,7 +1580,7 @@ static int do_oid_object_info_extended(struct repository *r,if (oi->disk_sizep) *(oi->disk_sizep) =0 ;if (oi->delta_base_oid) -oidclr (oi->delta_base_oid); +oidclr (oi->delta_base_oid, the_repository->hash_algo);if (oi->type_name)strbuf_addstr (oi->type_name,type_name (co->type));if (oi->contentp)
packfile.c +6 -4
@@ -1251,7 +1251,7 @@ static int get_delta_base_oid(struct packed_git *p, {if (type ==OBJ_REF_DELTA ) {unsigned char *base =use_pack (p, w_curs, curpos,NULL ); -oidread (oid, base); +oidread (oid, base, the_repository->hash_algo);return 0 ; }else if (type ==OBJ_OFS_DELTA ) {uint32_t base_pos;
@@ -1593,7 +1593,7 @@ int packed_object_info(struct repository *r, struct packed_git *p, goto out; } }else -oidclr (oi->delta_base_oid); +oidclr (oi->delta_base_oid, the_repository->hash_algo); } oi->whence = in_delta_base_cache(p, obj_offset) ? OI_DBCACHED :
@@ -1917,10 +1917,12 @@ int nth_packed_object_id(struct object_id *oid,return -1 ; index +=4 *256 ;if (p->index_version ==1 ) { -oidread (oid, index +st_add (st_mult (hashsz +4 , n),4 )); +oidread (oid, index +st_add (st_mult (hashsz +4 , n),4 ), + the_repository->hash_algo); }else { index +=8 ; -oidread (oid, index +st_mult (hashsz, n)); +oidread (oid, index +st_mult (hashsz, n), + the_repository->hash_algo); }return 0 ; }
read-cache.c +5 -3
@@ -1728,7 +1728,7 @@ static int verify_hdr(const struct cache_header *hdr, unsigned long size) end = (unsigned char *)hdr + size; start = end - the_hash_algo->rawsz; -oidread (&oid, start); +oidread (&oid, start, the_repository->hash_algo);if (oideq (&oid,null_oid ()))return 0 ;
@@ -1876,7 +1876,8 @@ static struct cache_entry *create_from_disk(struct mem_pool *ce_mem_pool, ce->ce_flags = flags & ~CE_NAMEMASK ; ce->ce_namelen = len; ce->index =0 ; -oidread (&ce->oid, (const unsigned char *)ondisk + offsetof(struct ondisk_cache_entry , data)); +oidread (&ce->oid, (const unsigned char *)ondisk + offsetof(struct ondisk_cache_entry , data), + the_repository->hash_algo);if (expand_name_field) {if (copy_len)
@@ -2249,7 +2250,8 @@ int do_read_index(struct index_state *istate, const char *path, int must_exist)if (verify_hdr (hdr, mmap_size) <0 ) goto unmap; -oidread (&istate->oid, (const unsigned char *)hdr + mmap_size - the_hash_algo->rawsz); +oidread (&istate->oid, (const unsigned char *)hdr + mmap_size - the_hash_algo->rawsz, + the_repository->hash_algo); istate->version =ntohl (hdr->hdr_version); istate->cache_nr =ntohl (hdr->hdr_entries); istate->cache_alloc =alloc_nr (istate->cache_nr);
refs.c +3 -3
@@ -1822,7 +1822,7 @@ const char *refs_resolve_ref_unsafe(struct ref_store *refs, failure_errno != ENOTDIR)return NULL ; -oidclr (oid); +oidclr (oid, the_repository->hash_algo);if (*flags &REF_BAD_NAME ) *flags |=REF_ISBROKEN ;return refname;
@@ -1832,7 +1832,7 @@ const char *refs_resolve_ref_unsafe(struct ref_store *refs,if (!(read_flags &REF_ISSYMREF )) {if (*flags &REF_BAD_NAME ) { -oidclr (oid); +oidclr (oid, the_repository->hash_algo); *flags |=REF_ISBROKEN ; }return refname;
@@ -1840,7 +1840,7 @@ const char *refs_resolve_ref_unsafe(struct ref_store *refs, refname = sb_refname.buf;if (resolve_flags &RESOLVE_REF_NO_RECURSE ) { -oidclr (oid); +oidclr (oid, the_repository->hash_algo);return refname; }if (check_refname_format (refname,REFNAME_ALLOW_ONELEVEL )) {
refs/files-backend.c +3 -3
@@ -246,7 +246,7 @@ static void loose_fill_ref_dir_regular_file(struct files_ref_store *refs,if (!refs_resolve_ref_unsafe (&refs->base, refname,RESOLVE_REF_READING , &oid, &flag)) { -oidclr (&oid); +oidclr (&oid, the_repository->hash_algo); flag |=REF_ISBROKEN ; }else if (is_null_oid (&oid)) {/*
@@ -263,7 +263,7 @@ static void loose_fill_ref_dir_regular_file(struct files_ref_store *refs,if (check_refname_format (refname,REFNAME_ALLOW_ONELEVEL )) {if (!refname_is_safe (refname))die ("loose refname is dangerous: %s" , refname); -oidclr (&oid); +oidclr (&oid, the_repository->hash_algo); flag |=REF_BAD_NAME |REF_ISBROKEN ; }add_entry_to_dir (dir,create_ref_entry (refname, &oid, flag));
@@ -1150,7 +1150,7 @@ static struct ref_lock *lock_ref_oid_basic(struct files_ref_store *refs,if (!refs_resolve_ref_unsafe (&refs->base, lock->ref_name,0 , &lock->old_oid,NULL )) -oidclr (&lock->old_oid); +oidclr (&lock->old_oid, the_repository->hash_algo); goto out; error_return:
refs/packed-backend.c +3 -3
@@ -894,7 +894,7 @@ static int next_record(struct packed_ref_iterator *iter)if (!refname_is_safe (iter->base.refname))die ("packed refname is dangerous: %s" , iter->base.refname); -oidclr (&iter->oid); +oidclr (&iter->oid, the_repository->hash_algo); iter->base.flags |=REF_BAD_NAME |REF_ISBROKEN ; } if (iter->snapshot->peeled == PEELED_FULLY ||
@@ -919,13 +919,13 @@ static int next_record(struct packed_ref_iterator *iter)* we suppress it if the reference is broken: */ if ((iter->base.flags &REF_ISBROKEN )) { -oidclr (&iter->peeled); +oidclr (&iter->peeled, the_repository->hash_algo); iter->base.flags &= ~REF_KNOWS_PEELED ; }else { iter->base.flags |=REF_KNOWS_PEELED ; } }else { -oidclr (&iter->peeled); +oidclr (&iter->peeled, the_repository->hash_algo); }return ITER_OK ;
refs/reftable-backend.c +23 -14
@@ -217,7 +217,8 @@ static int read_ref_without_reload(struct reftable_stack *stack,strbuf_addstr (referent, ref.value.symref); *type |=REF_ISSYMREF ; }else if (reftable_ref_record_val1 (&ref)) { -oidread (oid,reftable_ref_record_val1 (&ref)); +oidread (oid,reftable_ref_record_val1 (&ref), + the_repository->hash_algo); }else {/* We got a tombstone, which should not happen. */ BUG ("unhandled reference value type %d" , ref.value_type);
@@ -483,15 +484,17 @@ static int reftable_ref_iterator_advance(struct ref_iterator *ref_iterator)switch (iter->ref.value_type) {case REFTABLE_REF_VAL1 : -oidread (&iter->oid, iter->ref.value.val1); +oidread (&iter->oid, iter->ref.value.val1, + the_repository->hash_algo);break ;case REFTABLE_REF_VAL2 : -oidread (&iter->oid, iter->ref.value.val2.value); +oidread (&iter->oid, iter->ref.value.val2.value, + the_repository->hash_algo);break ;case REFTABLE_REF_SYMREF :if (!refs_resolve_ref_unsafe (&iter->refs->base, iter->ref.refname,RESOLVE_REF_READING , &iter->oid, &flags)) -oidclr (&iter->oid); +oidclr (&iter->oid, the_repository->hash_algo);break ;default :BUG ("unhandled reference value type %d" , iter->ref.value_type);
@@ -503,7 +506,7 @@ static int reftable_ref_iterator_advance(struct ref_iterator *ref_iterator)if (check_refname_format (iter->ref.refname,REFNAME_ALLOW_ONELEVEL )) {if (!refname_is_safe (iter->ref.refname))die (_ ("refname is dangerous: %s" ), iter->ref.refname); -oidclr (&iter->oid); +oidclr (&iter->oid, the_repository->hash_algo); flags |=REF_BAD_NAME |REF_ISBROKEN ; }
@@ -545,7 +548,8 @@ static int reftable_ref_iterator_peel(struct ref_iterator *ref_iterator, (struct reftable_ref_iterator *)ref_iterator;if (iter->ref.value_type ==REFTABLE_REF_VAL2 ) { -oidread (peeled, iter->ref.value.val2.target_value); +oidread (peeled, iter->ref.value.val2.target_value, + the_repository->hash_algo);return 0 ; }
@@ -1776,8 +1780,8 @@ static int yield_log_record(struct reftable_log_record *log,struct object_id old_oid, new_oid;const char *full_committer; -oidread (&old_oid, log->value.update.old_hash); -oidread (&new_oid, log->value.update.new_hash); +oidread (&old_oid, log->value.update.old_hash, the_repository->hash_algo); +oidread (&new_oid, log->value.update.new_hash, the_repository->hash_algo);/* * When both the old object ID and the new object ID are null
@@ -2178,7 +2182,8 @@ static int reftable_be_reflog_expire(struct ref_store *ref_store,if (ret <0 ) goto done;if (reftable_ref_record_val1 (&ref_record)) -oidread (&oid,reftable_ref_record_val1 (&ref_record)); +oidread (&oid,reftable_ref_record_val1 (&ref_record), + the_repository->hash_algo);prepare_fn (refname, &oid, policy_cb_data);while (1 ) {
@@ -2193,8 +2198,10 @@ static int reftable_be_reflog_expire(struct ref_store *ref_store,break ; } -oidread (&old_oid, log.value.update.old_hash); -oidread (&new_oid, log.value.update.new_hash); +oidread (&old_oid, log.value.update.old_hash, + the_repository->hash_algo); +oidread (&new_oid, log.value.update.new_hash, + the_repository->hash_algo);/* * Skip over the reflog existence marker. We will add it back
@@ -2225,8 +2232,10 @@ static int reftable_be_reflog_expire(struct ref_store *ref_store,struct object_id old_oid, new_oid; *dest = logs[i]; -oidread (&old_oid, logs[i].value.update.old_hash); -oidread (&new_oid, logs[i].value.update.new_hash); +oidread (&old_oid, logs[i].value.update.old_hash, + the_repository->hash_algo); +oidread (&new_oid, logs[i].value.update.new_hash, + the_repository->hash_algo); if (should_prune_fn(&old_oid, &new_oid, logs[i].value.update.email, (timestamp_t)logs[i].value.update.time,
@@ -2243,7 +2252,7 @@ static int reftable_be_reflog_expire(struct ref_store *ref_store,if (flags &EXPIRE_REFLOGS_UPDATE_REF && last_hash &&reftable_ref_record_val1 (&ref_record)) -oidread (&arg.update_oid, last_hash); +oidread (&arg.update_oid, last_hash, the_repository->hash_algo); arg.refs = refs; arg.records = rewritten;
remote.c +4 -4
@@ -1164,7 +1164,7 @@ static void tail_link_ref(struct ref *ref, struct ref ***tail)static struct ref *alloc_delete_ref (void ) {struct ref *ref =alloc_ref ("(delete)" ); -oidclr (&ref->new_oid); +oidclr (&ref->new_oid, the_repository->hash_algo);return ref; }
@@ -2531,7 +2531,7 @@ static int parse_push_cas_option(struct push_cas_option *cas, const char *arg, iif (!*colon) entry->use_tracking =1 ;else if (!colon[1 ]) -oidclr (&entry->expect); +oidclr (&entry->expect, the_repository->hash_algo);else if (repo_get_oid (the_repository, colon +1 , &entry->expect))return error (_ ("cannot parse expected object name '%s'" ), colon +1 );
@@ -2733,7 +2733,7 @@ static void apply_cas(struct push_cas_option *cas,else if (remote_tracking (remote, ref->name, &ref->old_oid_expect, &ref->tracking_ref)) -oidclr (&ref->old_oid_expect); +oidclr (&ref->old_oid_expect, the_repository->hash_algo);else ref->check_reachable = cas->use_force_if_includes;return ;
@@ -2747,7 +2747,7 @@ static void apply_cas(struct push_cas_option *cas,if (remote_tracking (remote, ref->name, &ref->old_oid_expect, &ref->tracking_ref)) -oidclr (&ref->old_oid_expect); +oidclr (&ref->old_oid_expect, the_repository->hash_algo);else ref->check_reachable = cas->use_force_if_includes; }
resolve-undo.c +2 -1
@@ -93,7 +93,8 @@ struct string_list *resolve_undo_read(const char *data, unsigned long size)continue ;if (size < rawsz) goto error; -oidread (&ui->oid[i], (const unsigned char *)data); +oidread (&ui->oid[i], (const unsigned char *)data, + the_repository->hash_algo); size -= rawsz; data += rawsz; }
sequencer.c +2 -2
@@ -3334,12 +3334,12 @@ static int rollback_is_safe(void)strbuf_release (&sb); }else if (errno ==ENOENT ) -oidclr (&expected_head); +oidclr (&expected_head, the_repository->hash_algo);else die_errno (_ ("could not read '%s'" ),git_path_abort_safety_file ());if (repo_get_oid (the_repository,"HEAD" , &actual_head)) -oidclr (&actual_head); +oidclr (&actual_head, the_repository->hash_algo);return oideq (&actual_head, &expected_head); }
split-index.c +1 -1
@@ -29,7 +29,7 @@ int read_link_extension(struct index_state *istate,if (sz < the_hash_algo->rawsz)return error ("corrupt link extension (too short)" ); si =init_split_index (istate); -oidread (&si->base_oid, data); +oidread (&si->base_oid, data, the_repository->hash_algo); data += the_hash_algo->rawsz; sz -= the_hash_algo->rawsz;if (!sz)
submodule-config.c +1 -1
@@ -682,7 +682,7 @@ static int gitmodule_oid_from_commit(const struct object_id *treeish_name,int ret =0 ;if (is_null_oid (treeish_name)) { -oidclr (gitmodules_oid); +oidclr (gitmodules_oid, the_repository->hash_algo);return 1 ; }
t/helper/test-submodule-config.c +1 -1
@@ -44,7 +44,7 @@ int cmd__submodule_config(int argc, const char **argv) path_or_name = arg[1 ];if (commit[0 ] =='\0' ) -oidclr (&commit_oid); +oidclr (&commit_oid, the_repository->hash_algo);else if (repo_get_oid (the_repository, commit, &commit_oid) <0 )die_usage (argc, argv,"Commit not found." );
tree-walk.c +2 -2
@@ -38,8 +38,8 @@ static int decode_tree_entry(struct tree_desc *desc, const char *buf, unsigned l desc->entry.path = path; desc->entry.mode = (desc->flags &TREE_DESC_RAW_MODES ) ? mode :canon_mode (mode); desc->entry.pathlen = len -1 ; -oidread_algop (&desc->entry.oid, (const unsigned char *)path + len, - desc->algo); +oidread (&desc->entry.oid, (const unsigned char *)path + len, + desc->algo);return 0 ; }
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 4/20 global: ensure that object IDs are always padded
In reply to Patrick SteinhardtThe oidcmp() and oideq() functions only compare the prefix length as specified by the given hash algorithm. This mandates that the object IDs have a valid hash algorithm set, or otherwise we wouldn't be able to figure out that prefix. As we do not have a hash algorithm in many cases, for example when handling null object IDs, this assumption cannot always be fulfilled. We thus have a fallback in place that instead uses the_repository to derive the hash function. This implicit dependency is hidden away from callers and can be quite surprising, especially in contexts where there may be no repository.
In theory, we can adapt those functions to always memcmp(3P) the whole length of their hash arrays. But there exist a couple of sites where we populate struct object_ids such that only the prefix of its hash that is actually used by the hash algorithm is populated. The remaining bytes are left uninitialized. The fact that those bytes are uninitialized also leads to warnings under Valgrind in some places where we copy those bytes.
Refactor callsites where we populate object IDs to always initialize all bytes. This also allows us to get rid of oidcpy_with_padding(), for one because the input is now fully initialized, and because oidcpy() will now always copy the whole hash array.
--- hash-ll.h | 2 ++ hash.h | 16 ---------------- hex.c | 6 +++++- http-push.c | 1 + notes.c | 2 ++ object-file.c | 2 ++ oidtree.c | 4 ++-- parallel-checkout.c | 8 +------- 8 files changed, 15 insertions(+), 26 deletions(-)
hash-ll.h +2 -0
@@ -288,6 +288,8 @@ static inline void oidread(struct object_id *oid, const unsigned char *hash, const struct git_hash_algo *algop) {memcpy (oid->hash, hash, algop->rawsz); +if (algop->rawsz <GIT_MAX_RAWSZ ) +memset (oid->hash + algop->rawsz,0 ,GIT_MAX_RAWSZ - algop->rawsz); oid->algo =hash_algo_by_ptr (algop); }
hash.h +0 -16
@@ -31,22 +31,6 @@ static inline int is_null_oid(const struct object_id *oid)return oideq (oid,null_oid ()); } -/* Like oidcpy() but zero-pads the unused bytes in dst's hash array. */ -static inline void oidcpy_with_padding (struct object_id *dst, -const struct object_id *src) -{ -size_t hashsz; - -if (!src->algo) - hashsz = the_hash_algo->rawsz; -else - hashsz = hash_algos[src->algo].rawsz; - -memcpy (dst->hash, src->hash, hashsz); -memset (dst->hash + hashsz,0 ,GIT_MAX_RAWSZ - hashsz); - dst->algo = src->algo; -} -static inline int is_empty_blob_oid (const struct object_id *oid) {return oideq (oid, the_hash_algo->empty_blob);
hex.c +5 -1
@@ -25,8 +25,12 @@ int get_oid_hex_algop(const char *hex, struct object_id *oid, const struct git_hash_algo *algop) {int ret =get_hash_hex_algop (hex, oid->hash, algop); -if (!ret) +if (!ret) {oid_set_algo (oid, algop); +if (algop->rawsz !=GIT_MAX_RAWSZ ) +memset (oid->hash + algop->rawsz,0 , +GIT_MAX_RAWSZ - algop->rawsz); + }return ret; }
http-push.c +1 -0
@@ -1016,6 +1016,7 @@ static void remote_ls(const char *path, int flags,/* extract hex from sharded "xx/x{38}" filename */ static int get_oid_hex_from_objpath (const char *path,struct object_id *oid) { +memset (oid->hash,0 ,GIT_MAX_RAWSZ ); oid->algo =hash_algo_by_ptr (the_hash_algo);if (strlen (path) != the_hash_algo->hexsz +1 )
notes.c +2 -0
@@ -427,6 +427,8 @@ static void load_subtree(struct notes_tree *t, struct leaf_node *subtree, hashsz - prefix_len)) goto handle_non_note;/* entry.path is not a SHA1 */ +memset (object_oid.hash + hashsz,0 ,GIT_MAX_RAWSZ - hashsz); + type =PTR_TYPE_NOTE ; }else if (path_len ==2 ) {/* This is potentially an internal node */
object-file.c +2 -0
@@ -2743,6 +2743,8 @@ int for_each_file_in_obj_subdir(unsigned int subdir_nr, !hex_to_bytes(oid.hash + 1, de->d_name, the_hash_algo->rawsz - 1)) {oid_set_algo (&oid, the_hash_algo); +memset (oid.hash + the_hash_algo->rawsz,0 , +GIT_MAX_RAWSZ - the_hash_algo->rawsz);if (obj_cb) { r =obj_cb (&oid, path->buf, data);if (r)
oidtree.c +2 -2
@@ -42,7 +42,7 @@ void oidtree_insert(struct oidtree *ot, const struct object_id *oid)* Clear the padding and copy the result in separate steps to * respect the 4-byte alignment needed by struct object_id. */ -oidcpy_with_padding (&k, oid); +oidcpy (&k, oid);memcpy (on->k, &k,sizeof (k));/*
@@ -60,7 +60,7 @@ int oidtree_contains(struct oidtree *ot, const struct object_id *oid)struct object_id k;size_t klen =sizeof (k); -oidcpy_with_padding (&k, oid); +oidcpy (&k, oid);if (oid->algo ==GIT_HASH_UNKNOWN ) klen -=sizeof (oid->algo);
parallel-checkout.c +1 -7
@@ -429,13 +429,7 @@ static void send_one_item(int fd, struct parallel_checkout_item *pc_item) fixed_portion->ident = pc_item->ca.ident; fixed_portion->name_len = name_len; fixed_portion->working_tree_encoding_len = working_tree_encoding_len; -/* -* We pad the unused bytes in the hash array because, otherwise, -* Valgrind would complain about passing uninitialized bytes to a -* write() syscall. The warning doesn't represent any real risk here, -* but it could hinder the detection of actual errors. -*/ -oidcpy_with_padding (&fixed_portion->oid, &pc_item->ce->oid); +oidcpy (&fixed_portion->oid, &pc_item->ce->oid); variant = data +sizeof (*fixed_portion);if (working_tree_encoding_len) {
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 5/20 hash: convert `oidcmp()` and `oideq()` to compare whole hash
In reply to Patrick SteinhardtWith the preceding commit, the hash array of object IDs is now fully zero-padded even when the hash algorithm's output is smaller than the array length. With that, we can now adapt both oidcmp() and oideq() to unconditionally memcmp(3P) the whole array instead of depending on the hash size.
While it may feel inefficient to compare unused bytes for e.g. SHA-1, in practice the compiler should now be able to produce code that is better optimized both because we have no branch anymore, but also because the size to compare is now known at compile time. Goldbolt spits out the following assembly on an x86_64 platform with GCC 14.1 for the old and new implementations of oidcmp():
oidcmp_old:
movsx rax, DWORD PTR [rdi+32]
test eax, eax
jne .L2
mov rax, QWORD PTR the_repository[rip]
cmp QWORD PTR [rax+16], 32
je .L6
.L4:
mov edx, 20
jmp memcmp
.L2:
lea rdx, [rax+rax*2]
lea rax, [rax+rdx*4]
lea rax, hash_algos[0+rax*8]
cmp QWORD PTR [rax+16], 32
jne .L4
.L6:
mov edx, 32
jmp memcmp
oidcmp_new:
mov edx, 32
jmp memcmp
The new implementation gets ridi of all the branches and effectively only ends setting up edx for memcmp() and then calling it.
And for oideq():
oideq_old:
movsx rcx, DWORD PTR [rdi+32]
mov rax, rdi
mov rdx, rsi
test ecx, ecx
jne .L2
mov rcx, QWORD PTR the_repository[rip]
cmp QWORD PTR [rcx+16], 32
mov rcx, QWORD PTR [rax]
je .L12
.L4:
mov rsi, QWORD PTR [rax+8]
xor rcx, QWORD PTR [rdx]
xor rsi, QWORD PTR [rdx+8]
or rcx, rsi
je .L13
.L8:
mov eax, 1
test eax, eax
sete al
movzx eax, al
ret
.L2:
lea rsi, [rcx+rcx*2]
lea rcx, [rcx+rsi*4]
lea rcx, hash_algos[0+rcx*8]
cmp QWORD PTR [rcx+16], 32
mov rcx, QWORD PTR [rax]
jne .L4
.L12:
mov rsi, QWORD PTR [rax+8]
xor rcx, QWORD PTR [rdx]
xor rsi, QWORD PTR [rdx+8]
or rcx, rsi
jne .L8
mov rcx, QWORD PTR [rax+16]
mov rax, QWORD PTR [rax+24]
xor rcx, QWORD PTR [rdx+16]
xor rax, QWORD PTR [rdx+24]
or rcx, rax
jne .L8
xor eax, eax
.L14:
test eax, eax
sete al
movzx eax, al
ret
.L13:
mov edi, DWORD PTR [rdx+16]
cmp DWORD PTR [rax+16], edi
jne .L8
xor eax, eax
jmp .L14
oideq_new:
mov rax, QWORD PTR [rdi]
mov rdx, QWORD PTR [rdi+8]
xor rax, QWORD PTR [rsi]
xor rdx, QWORD PTR [rsi+8]
or rax, rdx
je .L5
.L2:
mov eax, 1
xor eax, 1
ret
.L5:
mov rax, QWORD PTR [rdi+16]
mov rdx, QWORD PTR [rdi+24]
xor rax, QWORD PTR [rsi+16]
xor rdx, QWORD PTR [rsi+24]
or rax, rdx
jne .L2
xor eax, eax
xor eax, 1
ret
Interestingly, the compiler decides to split the comparisons into two so that it first compares the lower half of the object ID for equality and then the upper half. If the first check shows a difference, then we wouldn't even end up comparing the second half.
In both cases, the new generated code is significantly shorter and has way less branches. While I didn't benchmark the change, I'd be surprised if the new code was slower.
--- hash-ll.h | 10 ++++++++++ hash.h | 20 -------------------- 2 files changed, 10 insertions(+), 20 deletions(-)
hash-ll.h +10 -0
@@ -278,6 +278,16 @@ static inline void hashclr(unsigned char *hash, const struct git_hash_algo *algomemset (hash,0 , algop->rawsz); } +static inline int oidcmp (const struct object_id *oid1,const struct object_id *oid2) +{ +return memcmp (oid1->hash, oid2->hash,GIT_MAX_RAWSZ ); +} + +static inline int oideq (const struct object_id *oid1,const struct object_id *oid2) +{ +return !memcmp (oid1->hash, oid2->hash,GIT_MAX_RAWSZ ); +} +static inline void oidcpy (struct object_id *dst,const struct object_id *src) {memcpy (dst->hash, src->hash,GIT_MAX_RAWSZ );
hash.h +0 -20
@@ -6,26 +6,6 @@#define the_hash_algo the_repository->hash_algo -static inline int oidcmp (const struct object_id *oid1,const struct object_id *oid2) -{ -const struct git_hash_algo *algop; -if (!oid1->algo) - algop = the_hash_algo; -else - algop = &hash_algos[oid1->algo]; -return hashcmp (oid1->hash, oid2->hash, algop); -} - -static inline int oideq (const struct object_id *oid1,const struct object_id *oid2) -{ -const struct git_hash_algo *algop; -if (!oid1->algo) - algop = the_hash_algo; -else - algop = &hash_algos[oid1->algo]; -return hasheq (oid1->hash, oid2->hash, algop); -} -static inline int is_null_oid (const struct object_id *oid) {return oideq (oid,null_oid ());
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
Re: [PATCH v3 05/20] hash: convert `oidcmp()` and `oideq()` to compare whole hash
Quoted 33 lines · Patrick Steinhardt
With the preceding commit, the hash array of object IDs is now fully zero-padded even when the hash algorithm's output is smaller than the array length. With that, we can now adapt both
oidcmp()andoideq()to unconditionally memcmp(3P) the whole array instead of depending on the hash size.While it may feel inefficient to compare unused bytes for e.g. SHA-1, in practice the compiler should now be able to produce code that is better optimized both because we have no branch anymore, but also because the size to compare is now known at compile time. Goldbolt spits out the following assembly on an
x86_64platform with GCC 14.1 for the old and new implementations ofoidcmp():
oidcmp_old:
movsx rax, DWORD PTR [rdi+32]
test eax, eax
jne .L2
mov rax, QWORD PTR the_repository[rip]
cmp QWORD PTR [rax+16], 32
je .L6
.L4:
mov edx, 20
jmp memcmp
.L2:
lea rdx, [rax+rax*2]
lea rax, [rax+rdx*4]
lea rax, hash_algos[0+rax*8]
cmp QWORD PTR [rax+16], 32
jne .L4
.L6:
mov edx, 32
jmp memcmp
Quoted 5 lines
oidcmp_new:
mov edx, 32
jmp memcmpThe new implementation gets ridi of all the branches and effectively
s/ridi/rid
Quoted 80 lines
only ends setting up
edxformemcmp()and then calling it.And for
oideq():
oideq_old:
movsx rcx, DWORD PTR [rdi+32]
mov rax, rdi
mov rdx, rsi
test ecx, ecx
jne .L2
mov rcx, QWORD PTR the_repository[rip]
cmp QWORD PTR [rcx+16], 32
mov rcx, QWORD PTR [rax]
je .L12
.L4:
mov rsi, QWORD PTR [rax+8]
xor rcx, QWORD PTR [rdx]
xor rsi, QWORD PTR [rdx+8]
or rcx, rsi
je .L13
.L8:
mov eax, 1
test eax, eax
sete al
movzx eax, al
ret
.L2:
lea rsi, [rcx+rcx*2]
lea rcx, [rcx+rsi*4]
lea rcx, hash_algos[0+rcx*8]
cmp QWORD PTR [rcx+16], 32
mov rcx, QWORD PTR [rax]
jne .L4
.L12:
mov rsi, QWORD PTR [rax+8]
xor rcx, QWORD PTR [rdx]
xor rsi, QWORD PTR [rdx+8]
or rcx, rsi
jne .L8
mov rcx, QWORD PTR [rax+16]
mov rax, QWORD PTR [rax+24]
xor rcx, QWORD PTR [rdx+16]
xor rax, QWORD PTR [rdx+24]
or rcx, rax
jne .L8
xor eax, eax
.L14:
test eax, eax
sete al
movzx eax, al
ret
.L13:
mov edi, DWORD PTR [rdx+16]
cmp DWORD PTR [rax+16], edi
jne .L8
xor eax, eax
jmp .L14
oideq_new:
mov rax, QWORD PTR [rdi]
mov rdx, QWORD PTR [rdi+8]
xor rax, QWORD PTR [rsi]
xor rdx, QWORD PTR [rsi+8]
or rax, rdx
je .L5
.L2:
mov eax, 1
xor eax, 1
ret
.L5:
mov rax, QWORD PTR [rdi+16]
mov rdx, QWORD PTR [rdi+24]
xor rax, QWORD PTR [rsi+16]
xor rdx, QWORD PTR [rsi+24]
or rax, rdx
jne .L2
xor eax, eax
xor eax, 1
ret
Quoted 9 lines
Interestingly, the compiler decides to split the comparisons into two so that it first compares the lower half of the object ID for equality and then the upper half. If the first check shows a difference, then we wouldn't even end up comparing the second half.
In both cases, the new generated code is significantly shorter and has way less branches. While I didn't benchmark the change, I'd be surprised if the new code was slower.
This was nice to read, thanks for adding the ASM here.
Quoted 60 lines
Signed-off-by Patrick Steinhardt <[email protected]>--- hash-ll.h | 10 ++++++++++ hash.h | 20 -------------------- 2 files changed, 10 insertions(+), 20 deletions(-)hash-ll.h +10
-0@@ -278,6 +278,16 @@ static inline void hashclr(unsigned char *hash, const struct git_hash_algo *algomemset (hash,0 , algop->rawsz); } +static inline int oidcmp (const struct object_id *oid1,const struct object_id *oid2) +{ +return memcmp (oid1->hash, oid2->hash,GIT_MAX_RAWSZ ); +} + +static inline int oideq (const struct object_id *oid1,const struct object_id *oid2) +{ +return !memcmp (oid1->hash, oid2->hash,GIT_MAX_RAWSZ ); +} +static inline void oidcpy (struct object_id *dst,const struct object_id *src) {memcpy (dst->hash, src->hash,GIT_MAX_RAWSZ );hash.h +0
-21@@ -6,26 +6,6 @@#define the_hash_algo the_repository->hash_algo -static inline int oidcmp (const struct object_id *oid1,const struct object_id *oid2) -{ -const struct git_hash_algo *algop; -if (!oid1->algo) - algop = the_hash_algo; -else - algop = &hash_algos[oid1->algo]; -return hashcmp (oid1->hash, oid2->hash, algop); -} - -static inline int oideq (const struct object_id *oid1,const struct object_id *oid2) -{ -const struct git_hash_algo *algop; -if (!oid1->algo) - algop = the_hash_algo; -else - algop = &hash_algos[oid1->algo]; -return hasheq (oid1->hash, oid2->hash, algop); -} -static inline int is_null_oid (const struct object_id *oid) {return oideq (oid,null_oid ()); --2.45.2.457.g8d94cfb545.dirty
PATCH v3 6/20 hash: make `is_null_oid()` independent of `the_repository`
In reply to Patrick SteinhardtThe function is_null_oid() uses oideq(oid, null_oid()) to check whether a given object ID is the all-zero object ID. null_oid() implicitly relies on the_repository though to return the correct null object ID.
Get rid of this dependency by always comparing the complete hash array for being all-zeroes. This is possible due to the refactoring of object IDs so that their hash arrays are always fully initialized.
--- hash-ll.h | 6 ++++++ hash.h | 5 ----- 2 files changed, 6 insertions(+), 5 deletions(-)
hash-ll.h +6 -0
@@ -341,6 +341,12 @@ static inline unsigned int oidhash(const struct object_id *oid)return hash; } +static inline int is_null_oid (const struct object_id *oid) +{ +static const unsigned char null_hash[GIT_MAX_RAWSZ ]; +return !memcmp (oid->hash, null_hash,GIT_MAX_RAWSZ ); +} +const char *empty_tree_oid_hex (void );const char *empty_blob_oid_hex (void );
hash.h +0 -5
@@ -6,11 +6,6 @@#define the_hash_algo the_repository->hash_algo -static inline int is_null_oid (const struct object_id *oid) -{ -return oideq (oid,null_oid ()); -} -static inline int is_empty_blob_oid (const struct object_id *oid) {return oideq (oid, the_hash_algo->empty_blob);
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 7/20 hash: require hash algorithm in `is_empty_{blob,tree}_oid()`
In reply to Patrick SteinhardtBoth functions is_empty_{blob,tree}_oid() use the_repository to derive the hash function that shall be used. Require callers to pass in the hash algorithm to get rid of this implicit dependency.
--- builtin/fast-import.c | 4 +++- cache-tree.c | 2 +- diffcore-rename.c | 4 ++-- hash-ll.h | 12 ++++++++++++ hash.h | 10 ---------- read-cache.c | 2 +- 6 files changed, 19 insertions(+), 15 deletions(-)
builtin/fast-import.c +3 -1
@@ -2361,7 +2361,9 @@ static void file_change_m(const char *p, struct branch *b)parse_path_eol (&path, p,"path" );/* Git does not track empty, non-toplevel directories. */ -if (S_ISDIR (mode) &&is_empty_tree_oid (&oid) && *path.buf) { +if (S_ISDIR (mode) && +is_empty_tree_oid (&oid, the_repository->hash_algo) && + *path.buf) {tree_content_remove (&b->branch_tree, path.buf,NULL ,0 );return ; }
cache-tree.c +1 -1
@@ -422,7 +422,7 @@ static int update_one(struct cache_tree *it,/* * "sub" can be an empty tree if all subentries are i-t-a. */ -if (contains_ita &&is_empty_tree_oid (oid)) +if (contains_ita &&is_empty_tree_oid (oid, the_repository->hash_algo))continue ;strbuf_grow (&buffer, entlen +100 );
diffcore-rename.c +2 -2
@@ -1422,7 +1422,7 @@ void diffcore_rename_extended(struct diff_options *options, strcmp(options->single_follow, p->two->path)) continue; /* not interested */ else if (!options->flags.rename_empty && - is_empty_blob_oid(&p->two->oid)) + is_empty_blob_oid(&p->two->oid, the_repository->hash_algo))continue ; else if (add_rename_dst(p) < 0) { warning("skipping rename detection, detected"
@@ -1432,7 +1432,7 @@ void diffcore_rename_extended(struct diff_options *options, } }else if (!options->flags.rename_empty && -is_empty_blob_oid (&p->one->oid)) +is_empty_blob_oid (&p->one->oid, the_repository->hash_algo))continue ;else if (!DIFF_PAIR_UNMERGED (p) && !DIFF_FILE_VALID (p->two)) {/*
hash-ll.h +12 -0
@@ -350,4 +350,16 @@ static inline int is_null_oid(const struct object_id *oid)const char *empty_tree_oid_hex (void );const char *empty_blob_oid_hex (void ); +static inline int is_empty_blob_oid (const struct object_id *oid, +const struct git_hash_algo *algop) +{ +return oideq (oid, algop->empty_blob); +} + +static inline int is_empty_tree_oid (const struct object_id *oid, +const struct git_hash_algo *algop) +{ +return oideq (oid, algop->empty_tree); +} +#endif
hash.h +0 -10
@@ -6,14 +6,4 @@#define the_hash_algo the_repository->hash_algo -static inline int is_empty_blob_oid (const struct object_id *oid) -{ -return oideq (oid, the_hash_algo->empty_blob); -} - -static inline int is_empty_tree_oid (const struct object_id *oid) -{ -return oideq (oid, the_hash_algo->empty_tree); -} -#endif
read-cache.c +1 -1
@@ -337,7 +337,7 @@ static int ce_match_stat_basic(const struct cache_entry *ce, struct stat *st)/* Racily smudged entry? */ if (!ce->ce_stat_data.sd_size) { -if (!is_empty_blob_oid (&ce->oid)) +if (!is_empty_blob_oid (&ce->oid, the_repository->hash_algo)) changed |=DATA_CHANGED ; }
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 8/20 hash: require hash algorithm in `empty_tree_oid_hex()`
In reply to Patrick SteinhardtThe empty_tree_oid_hex() function use the_repository to derive the hash function that shall be used. Require callers to pass in the hash algorithm to get rid of this implicit dependency.
While at it, remove the unused empty_blob_oid_hex() function.
--- add-interactive.c | 2 +- add-patch.c | 2 +- builtin/merge.c | 3 ++- builtin/receive-pack.c | 2 +- hash-ll.h | 3 +-- object-file.c | 10 ++-------- sequencer.c | 2 +- submodule.c | 6 +++--- wt-status.c | 4 ++-- 9 files changed, 14 insertions(+), 20 deletions(-)
add-interactive.c +1 -1
@@ -557,7 +557,7 @@ static int get_modified_files(struct repository *r, s.skip_unseen = filter && i; opt.def = is_initial ? -empty_tree_oid_hex () :oid_to_hex (&head_oid); +empty_tree_oid_hex (the_repository->hash_algo) :oid_to_hex (&head_oid);repo_init_revisions (r, &rev,NULL );setup_revisions (0 ,NULL , &rev, &opt);
add-patch.c +1 -1
@@ -420,7 +420,7 @@ static int parse_diff(struct add_p_state *s, const struct pathspec *ps)/* could be on an unborn branch */ !strcmp ("HEAD" , s->revision) &&repo_get_oid (the_repository,"HEAD" , &oid) ? - empty_tree_oid_hex() : s->revision); + empty_tree_oid_hex(the_repository->hash_algo) : s->revision); } color_arg_index = args.nr;/* Use `--no-color` explicitly, just in case `diff.color = always`. */
builtin/merge.c +2 -1
@@ -330,7 +330,8 @@ static void read_empty(const struct object_id *oid) {struct child_process cmd =CHILD_PROCESS_INIT ; -strvec_pushl (&cmd.args,"read-tree" ,"-m" ,"-u" ,empty_tree_oid_hex (), +strvec_pushl (&cmd.args,"read-tree" ,"-m" ,"-u" , +empty_tree_oid_hex (the_repository->hash_algo),oid_to_hex (oid),NULL ); cmd.git_cmd =1 ;
builtin/receive-pack.c +1 -1
@@ -1371,7 +1371,7 @@ static const char *push_to_deploy(unsigned char *sha1,strvec_pushl (&child.args,"diff-index" ,"--quiet" ,"--cached" ,"--ignore-submodules" ,/* diff-index with either HEAD or an empty tree */ -head_has_history () ?"HEAD" :empty_tree_oid_hex (), +head_has_history () ?"HEAD" :empty_tree_oid_hex (the_repository->hash_algo),"--" ,NULL );strvec_pushv (&child.env, env->v); child.no_stdin =1 ;
hash-ll.h +1 -2
@@ -347,8 +347,7 @@ static inline int is_null_oid(const struct object_id *oid)return !memcmp (oid->hash, null_hash,GIT_MAX_RAWSZ ); } -const char *empty_tree_oid_hex (void ); -const char *empty_blob_oid_hex (void ); +const char *empty_tree_oid_hex (const struct git_hash_algo *algop);static inline int is_empty_blob_oid (const struct object_id *oid,const struct git_hash_algo *algop)
object-file.c +2 -8
@@ -227,16 +227,10 @@ const struct object_id *null_oid(void)return the_hash_algo->null_oid; } -const char *empty_tree_oid_hex (void ) +const char *empty_tree_oid_hex (const struct git_hash_algo *algop) {static char buf[GIT_MAX_HEXSZ +1 ]; -return oid_to_hex_r (buf, the_hash_algo->empty_tree); -} - -const char *empty_blob_oid_hex (void ) -{ -static char buf[GIT_MAX_HEXSZ +1 ]; -return oid_to_hex_r (buf, the_hash_algo->empty_blob); +return oid_to_hex_r (buf, algop->empty_tree); }int hash_algo_by_name (const char *name)
sequencer.c +1 -1
@@ -2263,7 +2263,7 @@ static int do_pick_commit(struct repository *r, unborn =1 ; }else if (unborn)oidcpy (&head, the_hash_algo->empty_tree); -if (index_differs_from (r, unborn ?empty_tree_oid_hex () :"HEAD" , +if (index_differs_from (r, unborn ?empty_tree_oid_hex (the_repository->hash_algo) :"HEAD" ,NULL ,0 ))return error_dirty_index (r, opts); }
submodule.c +3 -3
@@ -2119,7 +2119,7 @@ static void submodule_reset_index(const char *path, const char *super_prefix)strvec_pushf (&cp.args,"--super-prefix=%s%s/" , (super_prefix ? super_prefix :"" ), path); -strvec_push (&cp.args,empty_tree_oid_hex ()); +strvec_push (&cp.args,empty_tree_oid_hex (the_repository->hash_algo));if (run_command (&cp))die (_ ("could not reset submodule index" ));
@@ -2229,9 +2229,9 @@ int submodule_move_head(const char *path, const char *super_prefix,strvec_push (&cp.args,"-m" );if (!(flags &SUBMODULE_MOVE_HEAD_FORCE )) -strvec_push (&cp.args, old_head ? old_head :empty_tree_oid_hex ()); +strvec_push (&cp.args, old_head ? old_head :empty_tree_oid_hex (the_repository->hash_algo)); -strvec_push (&cp.args, new_head ? new_head :empty_tree_oid_hex ()); +strvec_push (&cp.args, new_head ? new_head :empty_tree_oid_hex (the_repository->hash_algo));if (run_command (&cp)) { ret =error (_ ("Submodule '%s' could not be updated." ), path);
wt-status.c +2 -2
@@ -641,7 +641,7 @@ static void wt_status_collect_changes_index(struct wt_status *s)repo_init_revisions (s->repo, &rev,NULL );memset (&opt,0 ,sizeof (opt)); - opt.def = s->is_initial ?empty_tree_oid_hex () : s->reference; + opt.def = s->is_initial ?empty_tree_oid_hex (the_repository->hash_algo) : s->reference;setup_revisions (0 ,NULL , &rev, &opt); rev.diffopt.flags.override_submodule_config =1 ;
@@ -1136,7 +1136,7 @@ static void wt_longstatus_print_verbose(struct wt_status *s) rev.diffopt.ita_invisible_in_index =1 ;memset (&opt,0 ,sizeof (opt)); - opt.def = s->is_initial ?empty_tree_oid_hex () : s->reference; + opt.def = s->is_initial ?empty_tree_oid_hex (the_repository->hash_algo) : s->reference;setup_revisions (0 ,NULL , &rev, &opt); rev.diffopt.output_format |=DIFF_FORMAT_PATCH ;
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 9/20 global: introduce `USE_THE_REPOSITORY_VARIABLE` macro
In reply to Patrick SteinhardtUse of the the_repository variable is deprecated nowadays, and we slowly but steadily convert the codebase to not use it anymore. Instead, callers should be passing down the repository to work on via parameters.
It is hard though to prove that a given code unit does not use this variable anymore. The most trivial case, merely demonstrating that there is no direct use of the_repository, is already a bit of a pain during code reviews as the reviewer needs to manually verify claims made by the patch author. The bigger problem though is that we have many interfaces that implicitly rely on the_repository.
Introduce a new USE_THE_REPOSITORY_VARIABLE macro that allows code units to opt into usage of the_repository. The intent of this macro is to demonstrate that a certain code unit does not use this variable anymore, and to keep it from new dependencies on it in future changes, be it explicit or implicit
For now, the macro only guards the_repository itself as well as the_hash_algo. There are many more known interfaces where we have an implicit dependency on the_repository, but those are not guarded at the current point in time. Over time though, we should start to add guards as required (or even better, just remove them).
Define the macro as required in our code units. As expected, most of our code still relies on the global variable. Nearly all of our builtins rely on the variable as there is no way yet to pass the_repository to their entry point. For now, declare the macro in "biultin.h" to keep the required changes at least a little bit more contained.
--- add-interactive.c | 2 ++ add-patch.c | 2 ++ apply.c | 2 ++ archive-tar.c | 3 +++ archive-zip.c | 3 +++ archive.c | 2 ++ attr.c | 2 ++ bisect.c | 2 ++ blame.c | 2 ++ branch.c | 2 ++ builtin.h | 8 ++++++++ builtin/blame.c | 2 +- builtin/log.c | 2 +- bulk-checkin.c | 3 +++ bundle-uri.c | 2 ++ bundle.c | 2 ++ cache-tree.c | 2 ++ checkout.c | 2 ++ chunk-format.c | 2 ++ combine-diff.c | 2 ++ commit-graph.c | 2 ++ commit-reach.c | 2 ++ commit.c | 2 ++ common-main.c | 2 ++ compat/win32/trace2_win32_process_info.c | 2 ++ config.c | 3 +++ connected.c | 2 ++ convert.c | 2 ++ csum-file.c | 3 +++ delta-islands.c | 2 ++ diagnose.c | 2 ++ diff-lib.c | 3 +++ diff.c | 3 +++ diffcore-break.c | 3 +++ diffcore-rename.c | 3 +++ dir.c | 3 +++ entry.c | 2 ++ environment.c | 3 +++ fetch-pack.c | 2 ++ fmt-merge-msg.c | 2 ++ fsck.c | 2 ++ fsmonitor-ipc.c | 2 ++ git.c | 2 ++ hash-lookup.c | 2 ++ hash.h | 4 +++- help.c | 2 ++ hex.c | 2 ++ http-backend.c | 2 ++ http-push.c | 2 ++ http-walker.c | 2 ++ http.c | 2 ++ list-objects-filter-options.c | 2 ++ list-objects-filter.c | 2 ++ list-objects.c | 2 ++ log-tree.c | 2 ++ loose.c | 2 ++ ls-refs.c | 2 ++ mailmap.c | 2 ++ match-trees.c | 2 ++ merge-blobs.c | 2 ++ merge-ort.c | 2 ++ merge-recursive.c | 3 +++ merge.c | 2 ++ midx-write.c | 2 ++ midx.c | 2 ++ negotiator/default.c | 2 ++ negotiator/skipping.c | 2 ++ notes-cache.c | 2 ++ notes-merge.c | 2 ++ notes-utils.c | 2 ++ notes.c | 2 ++ object-file-convert.c | 2 ++ object-file.c | 3 +++ object-name.c | 2 ++ object.c | 2 ++ oid-array.c | 2 ++ oss-fuzz/fuzz-commit-graph.c | 2 ++ pack-bitmap-write.c | 2 ++ pack-bitmap.c | 2 ++ pack-check.c | 2 ++ pack-revindex.c | 2 ++ pack-write.c | 2 ++ packfile.c | 2 ++ parse-options-cb.c | 2 ++ path.c | 3 +++ pathspec.c | 2 ++ pretty.c | 2 ++ progress.c | 2 ++ promisor-remote.c | 2 ++ range-diff.c | 2 ++ reachable.c | 2 ++ read-cache.c | 3 +++ rebase-interactive.c | 2 ++ ref-filter.c | 2 ++ reflog-walk.c | 2 ++ reflog.c | 2 ++ refs.c | 2 ++ refs/files-backend.c | 2 ++ refs/packed-backend.c | 2 ++ refs/reftable-backend.c | 2 ++ refspec.c | 2 ++ remote-curl.c | 2 ++ remote.c | 2 ++ repository.c | 8 ++++++++ repository.h | 2 ++ rerere.c | 2 ++ reset.c | 2 ++ resolve-undo.c | 2 ++ revision.c | 2 ++ run-command.c | 2 ++ scalar.c | 2 ++ send-pack.c | 2 ++ sequencer.c | 2 ++ serve.c | 2 ++ server-info.c | 2 ++ setup.c | 2 ++ shallow.c | 2 ++ split-index.c | 2 ++ streaming.c | 3 +++ submodule-config.c | 2 ++ submodule.c | 2 ++ t/helper/test-bitmap.c | 2 ++ t/helper/test-bloom.c | 2 ++ t/helper/test-cache-tree.c | 2 ++ t/helper/test-dump-cache-tree.c | 2 ++ t/helper/test-dump-fsmonitor.c | 2 ++ t/helper/test-dump-split-index.c | 2 ++ t/helper/test-dump-untracked-cache.c | 2 ++ t/helper/test-find-pack.c | 2 ++ t/helper/test-fsmonitor-client.c | 2 ++ t/helper/test-lazy-init-name-hash.c | 2 ++ t/helper/test-match-trees.c | 2 ++ t/helper/test-oidmap.c | 2 ++ t/helper/test-pack-mtimes.c | 2 ++ t/helper/test-reach.c | 2 ++ t/helper/test-read-cache.c | 2 ++ t/helper/test-read-graph.c | 2 ++ t/helper/test-read-midx.c | 2 ++ t/helper/test-ref-store.c | 2 ++ t/helper/test-repository.c | 2 ++ t/helper/test-revision-walking.c | 2 ++ t/helper/test-scrap-cache-tree.c | 2 ++ t/helper/test-submodule-config.c | 2 ++ t/helper/test-submodule-nested-repo-config.c | 2 ++ t/helper/test-submodule.c | 2 ++ t/helper/test-trace2.c | 2 ++ t/helper/test-write-cache.c | 2 ++ t/unit-tests/t-example-decorate.c | 2 ++ tag.c | 2 ++ tmp-objdir.c | 2 ++ transport-helper.c | 2 ++ transport.c | 2 ++ tree-walk.c | 2 ++ tree.c | 2 ++ unpack-trees.c | 2 ++ upload-pack.c | 2 ++ walker.c | 2 ++ worktree.c | 2 ++ wt-status.c | 2 ++ xdiff-interface.c | 2 ++ 160 files changed, 347 insertions(+), 3 deletions(-)
add-interactive.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "add-interactive.h" #include "color.h"
add-patch.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "add-interactive.h" #include "advice.h"
apply.c +2 -0
@@ -7,6 +7,8 @@* */ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "base85.h"
archive-tar.c +3 -0
@@ -1,6 +1,9 @@/* * Copyright (c) 2005, 2006 Rene Scharfe */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "gettext.h"
archive-zip.c +3 -0
@@ -1,6 +1,9 @@/* * Copyright (c) 2006 Rene Scharfe */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "archive.h"
archive.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "config.h"
attr.c +2 -0
@@ -6,6 +6,8 @@* an insanely large number of attributes. */ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "environment.h"
bisect.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "commit.h"
blame.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "refs.h" #include "object-store-ll.h"
branch.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "advice.h" #include "config.h"
builtin.h +8 -0
@@ -1,6 +1,14 @@#ifndef BUILTIN_H #define BUILTIN_H +/* +* TODO: Almost all of our builtins access `the_repository` by necessity +* because they do not get passed a pointer to it. We should adapt the function +* signature of those main functions to accept a `struct repository *` and then +* remove the macro here. +*/ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" /*
builtin/blame.c +1 -1
@@ -5,7 +5,7 @@* See COPYING for licensing conditions */ -#include "git-compat-util.h" +#include "builtin.h" #include "config.h" #include "color.h" #include "builtin.h"
builtin/log.c +1 -1
@@ -4,7 +4,7 @@* (C) Copyright 2006 Linus Torvalds * 2006 Junio Hamano */ -#include "git-compat-util.h" +#include "builtin.h" #include "abspath.h" #include "config.h" #include "environment.h"
bulk-checkin.c +3 -0
@@ -1,6 +1,9 @@/* * Copyright (c) 2011, Google Inc. */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "bulk-checkin.h" #include "environment.h"
bundle-uri.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "bundle-uri.h" #include "bundle.h"
bundle.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "lockfile.h" #include "bundle.h"
cache-tree.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "hex.h"
checkout.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "object-name.h" #include "remote.h"
chunk-format.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "chunk-format.h" #include "csum-file.h"
combine-diff.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "object-store-ll.h" #include "commit.h"
commit-graph.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "csum-file.h"
commit-reach.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "commit.h" #include "commit-graph.h"
commit.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "tag.h" #include "commit.h"
common-main.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "exec-cmd.h" #include "gettext.h"
compat/win32/trace2_win32_process_info.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "../../git-compat-util.h" #include "../../json-writer.h" #include "../../repository.h"
config.c +3 -0
@@ -5,6 +5,9 @@* Copyright (C) Johannes Schindelin, 2005 * */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "advice.h"
connected.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "hex.h"
convert.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "advice.h" #include "config.h"
csum-file.c +3 -0
@@ -7,6 +7,9 @@* files. Useful when you write a file that you want to be * able to verify hasn't been messed with afterwards. */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "progress.h" #include "csum-file.h"
delta-islands.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "object.h" #include "commit.h"
diagnose.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "diagnose.h" #include "compat/disk.h"
diff-lib.c +3 -0
@@ -1,6 +1,9 @@/* * Copyright (C) 2005 Junio C Hamano */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "commit.h" #include "diff.h"
diff.c +3 -0
@@ -1,6 +1,9 @@/* * Copyright (C) 2005 Junio C Hamano */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "base85.h"
diffcore-break.c +3 -0
@@ -1,6 +1,9 @@/* * Copyright (C) 2005 Junio C Hamano */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "diffcore.h" #include "hash.h"
diffcore-rename.c +3 -0
@@ -2,6 +2,9 @@* * Copyright (C) 2005 Junio C Hamano */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "diff.h" #include "diffcore.h"
dir.c +3 -0
@@ -5,6 +5,9 @@* Copyright (C) Linus Torvalds, 2005-2006 * Junio Hamano, 2005-2006 */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "config.h"
entry.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "object-store-ll.h" #include "dir.h"
environment.c +3 -0
@@ -7,6 +7,9 @@* even if you might want to know where the git directory etc * are. */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "branch.h"
fetch-pack.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "repository.h" #include "config.h"
fmt-merge-msg.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "environment.h"
fsck.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "date.h" #include "dir.h"
fsmonitor-ipc.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "simple-ipc.h"
git.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "builtin.h" #include "config.h" #include "environment.h"
hash-lookup.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "hash.h" #include "hash-lookup.h"
hash.h +3 -1
@@ -4,6 +4,8 @@#include "hash-ll.h" #include "repository.h" -#define the_hash_algo the_repository->hash_algo +#ifdef USE_THE_REPOSITORY_VARIABLE +# define the_hash_algo the_repository->hash_algo +#endif #endif
help.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "builtin.h"
hex.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "hash.h" #include "hex.h"
http-backend.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "environment.h"
http-push.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "hex.h"
http-walker.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "repository.h" #include "hex.h"
http.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "git-curl-compat.h" #include "hex.h"
list-objects-filter-options.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "gettext.h"
list-objects-filter.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "dir.h" #include "gettext.h"
list-objects.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "tag.h" #include "commit.h"
log-tree.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "commit-reach.h" #include "config.h"
loose.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "hash.h" #include "path.h"
ls-refs.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "gettext.h"
mailmap.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "string-list.h"
match-trees.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "hex.h" #include "match-trees.h"
merge-blobs.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "merge-ll.h" #include "blob.h"
merge-ort.c +2 -0
@@ -14,6 +14,8 @@* "cale", "peedy", or "ins" instead of "ort"?) */ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "merge-ort.h"
merge-recursive.c +3 -0
@@ -3,6 +3,9 @@* Fredrik Kuivinen. * The thieves were Alex Riesen and Johannes Schindelin, in June/July 2006 */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "merge-recursive.h"
merge.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "hash.h"
midx-write.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "config.h"
midx.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "dir.h"
negotiator/default.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "default.h" #include "../commit.h"
negotiator/skipping.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "skipping.h" #include "../commit.h"
notes-cache.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "notes-cache.h" #include "object-store-ll.h"
notes-merge.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "advice.h" #include "commit.h"
notes-utils.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "commit.h"
notes.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "environment.h"
object-file-convert.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "strbuf.h"
object-file.c +3 -0
@@ -6,6 +6,9 @@* This handles basic git object files - packing, unpacking, * creation etc. */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "config.h"
object-name.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "object-name.h" #include "advice.h"
object.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "hex.h"
oid-array.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "oid-array.h" #include "hash-lookup.h"
oss-fuzz/fuzz-commit-graph.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "commit-graph.h" #include "repository.h"
pack-bitmap-write.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "gettext.h"
pack-bitmap.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "commit.h" #include "gettext.h"
pack-check.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "hex.h"
pack-revindex.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "pack-revindex.h"
pack-write.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "gettext.h"
packfile.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "gettext.h"
parse-options-cb.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "parse-options.h" #include "branch.h"
path.c +3 -0
@@ -1,6 +1,9 @@/* * Utilities for paths and pathnames */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "environment.h"
pathspec.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "parse.h"
pretty.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "commit.h"
progress.c +2 -0
@@ -9,6 +9,8 @@*/ #define GIT_TEST_PROGRESS_ONLY +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "pager.h" #include "progress.h"
promisor-remote.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "hex.h"
range-diff.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "gettext.h"
reachable.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "hex.h"
read-cache.c +3 -0
@@ -3,6 +3,9 @@* * Copyright (C) Linus Torvalds, 2005 */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "bulk-checkin.h" #include "config.h"
rebase-interactive.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "commit.h" #include "editor.h"
ref-filter.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "gettext.h"
reflog-walk.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "commit.h" #include "refs.h"
reflog.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "object-store-ll.h"
refs.c +2 -0
@@ -2,6 +2,8 @@* The backend-independent part of the reference module. */ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "advice.h" #include "config.h"
refs/files-backend.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "../git-compat-util.h" #include "../copy.h" #include "../environment.h"
refs/packed-backend.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "../git-compat-util.h" #include "../config.h" #include "../dir.h"
refs/reftable-backend.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "../git-compat-util.h" #include "../abspath.h" #include "../chdir-notify.h"
refspec.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "hash.h"
remote-curl.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "git-curl-compat.h" #include "config.h"
remote.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "config.h"
repository.c +8 -0
@@ -16,6 +16,14 @@#include "promisor-remote.h" #include "refs.h" +/* +* We do not define `USE_THE_REPOSITORY_VARIABLE` in this file because we do +* not want to rely on functions that implicitly use `the_repository`. This +* means that the `extern` declaration of `the_repository` isn't visible here, +* which makes sparse unhappy. We thus declare it here. +*/ +extern struct repository *the_repository; +/* The main repository */ static struct repository the_repo;struct repository *the_repository = &the_repo;
repository.h +2 -0
@@ -197,7 +197,9 @@ struct repository { unsigned different_commondir:1; }; +#ifdef USE_THE_REPOSITORY_VARIABLE extern struct repository *the_repository; +#endif /* * Define a custom repository layout. Any field can be NULL, which
rerere.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "config.h"
reset.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "cache-tree.h" #include "gettext.h"
resolve-undo.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "dir.h" #include "hash.h"
revision.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "environment.h"
run-command.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "run-command.h" #include "environment.h"
scalar.c +2 -0
@@ -2,6 +2,8 @@* The Scalar command-line interface. */ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "gettext.h"
send-pack.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "commit.h"
sequencer.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "advice.h"
serve.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "repository.h" #include "config.h"
server-info.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "dir.h" #include "environment.h"
setup.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "copy.h"
shallow.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "hex.h" #include "repository.h"
split-index.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "hash.h"
streaming.c +3 -0
@@ -1,6 +1,9 @@/* * Copyright (c) 2011, Google Inc. */ + +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "convert.h" #include "environment.h"
submodule-config.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "dir.h" #include "environment.h"
submodule.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "repository.h"
t/helper/test-bitmap.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "git-compat-util.h" #include "pack-bitmap.h"
t/helper/test-bloom.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "bloom.h" #include "hex.h"
t/helper/test-cache-tree.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "gettext.h" #include "hex.h"
t/helper/test-dump-cache-tree.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "hash.h" #include "hex.h"
t/helper/test-dump-fsmonitor.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "read-cache-ll.h" #include "repository.h"
t/helper/test-dump-split-index.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "hex.h" #include "read-cache-ll.h"
t/helper/test-dump-untracked-cache.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "dir.h" #include "hex.h"
t/helper/test-find-pack.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "object-name.h" #include "object-store.h"
t/helper/test-fsmonitor-client.c +2 -0
@@ -3,6 +3,8 @@* a `git fsmonitor--daemon` daemon. */ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "parse-options.h" #include "fsmonitor-ipc.h"
t/helper/test-lazy-init-name-hash.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "environment.h" #include "name-hash.h"
t/helper/test-match-trees.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "hex.h" #include "match-trees.h"
t/helper/test-oidmap.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "hex.h" #include "object-name.h"
t/helper/test-pack-mtimes.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "hex.h" #include "strbuf.h"
t/helper/test-reach.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "commit.h" #include "commit-reach.h"
t/helper/test-read-cache.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "config.h" #include "read-cache-ll.h"
t/helper/test-read-graph.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "commit-graph.h" #include "repository.h"
t/helper/test-read-midx.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "hex.h" #include "midx.h"
t/helper/test-ref-store.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "hex.h" #include "refs.h"
t/helper/test-repository.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "commit-graph.h" #include "commit.h"
t/helper/test-revision-walking.c +2 -0
@@ -8,6 +8,8 @@* published by the Free Software Foundation. */ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "commit.h" #include "diff.h"
t/helper/test-scrap-cache-tree.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "lockfile.h" #include "read-cache-ll.h"
t/helper/test-submodule-config.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "config.h" #include "hash.h"
t/helper/test-submodule-nested-repo-config.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "repository.h" #include "setup.h"
t/helper/test-submodule.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "test-tool-utils.h" #include "parse-options.h"
t/helper/test-trace2.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "strvec.h" #include "run-command.h"
t/helper/test-write-cache.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "lockfile.h" #include "read-cache-ll.h"
t/unit-tests/t-example-decorate.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-lib.h" #include "object.h" #include "decorate.h"
tag.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "environment.h" #include "tag.h"
tmp-objdir.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "tmp-objdir.h" #include "abspath.h"
transport-helper.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "transport.h" #include "quote.h"
transport.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "advice.h" #include "config.h"
tree-walk.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "tree-walk.h" #include "dir.h"
tree.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "hex.h" #include "tree.h"
unpack-trees.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "advice.h" #include "strvec.h"
upload-pack.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "environment.h"
walker.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "hex.h"
worktree.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "abspath.h" #include "environment.h"
wt-status.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "advice.h" #include "wt-status.h"
xdiff-interface.c +2 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "gettext.h" #include "config.h"
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
Re: [PATCH v3 09/20] global: introduce `USE_THE_REPOSITORY_VARIABLE` macro
Quoted 23 lines · Patrick Steinhardt
Use of the
the_repositoryvariable is deprecated nowadays, and we slowly but steadily convert the codebase to not use it anymore. Instead, callers should be passing down the repository to work on via parameters.It is hard though to prove that a given code unit does not use this variable anymore. The most trivial case, merely demonstrating that there is no direct use of
the_repository, is already a bit of a pain during code reviews as the reviewer needs to manually verify claims made by the patch author. The bigger problem though is that we have many interfaces that implicitly rely onthe_repository.Introduce a new
USE_THE_REPOSITORY_VARIABLEmacro that allows code units to opt into usage ofthe_repository. The intent of this macro is to demonstrate that a certain code unit does not use this variable anymore, and to keep it from new dependencies on it in future changes, be it explicit or implicitFor now, the macro only guards
the_repositoryitself as well asthe_hash_algo. There are many more known interfaces where we have an implicit dependency onthe_repository, but those are not guarded at the current point in time. Over time though, we should start to add guards as required (or even better, just remove them).
Quoted 4 lines
Define the macro as required in our code units. As expected, most of our code still relies on the global variable. Nearly all of our builtins rely on the variable as there is no way yet to pass
the_repositoryto their entry point. For now, declare the macro in "biultin.h" to keep the
s/biultin.h/builtin.h
[snip]
hash.h line 8
Quoted 11 lines
hash.h +2
-1@@ -4,6 +4,8 @@#include "hash-ll.h" #include "repository.h" -#define the_hash_algo the_repository->hash_algo +#ifdef USE_THE_REPOSITORY_VARIABLE +# define the_hash_algo the_repository->hash_algo
s/# define/#define/
Re: [PATCH v3 09/20] global: introduce `USE_THE_REPOSITORY_VARIABLE` macro
Quoted 9 lines · Karthik Nayak
Quoted 8 lines · Patrick Steinhardt
hash.h +0
-0@@ -4,6 +4,8 @@#include "hash-ll.h" #include "repository.h"
Quoted 5 lines
hash.h line 8
Quoted 3 lines
hash.h +2
-1-#define the_hash_algo the_repository->hash_algo +#ifdef USE_THE_REPOSITORY_VARIABLE +# define the_hash_algo the_repository->hash_algos/# define/#define/
This is in fact intentional. We aren't strictly following this in our codebase, but when nesting preprocessor macros into ifdefs then we often indent the inner macros with spaces.
Patrick
Re: [PATCH v3 09/20] global: introduce `USE_THE_REPOSITORY_VARIABLE` macro
Quoted 16 lines · Patrick Steinhardt
Quoted 9 lines · Karthik Nayak
Quoted 8 lines · Patrick Steinhardt
hash.h +0
-0@@ -4,6 +4,8 @@#include "hash-ll.h" #include "repository.h" Quoted 5 lines
hash.h line 8
Quoted 3 lines
hash.h +2
-1-#define the_hash_algo the_repository->hash_algo +#ifdef USE_THE_REPOSITORY_VARIABLE +# define the_hash_algo the_repository->hash_algos/# define/#define/
Quoted 5 lines
This is in fact intentional. We aren't strictly following this in our codebase, but when nesting preprocessor macros into ifdefs then we often indent the inner macros with spaces.
Patrick
That's something I didn't know. Thanks.
Re: [PATCH v3 09/20] global: introduce `USE_THE_REPOSITORY_VARIABLE` macro
Quoted 9 lines · Karthik Nayak
Quoted 7 lines
Quoted 1 line
s/# define/#define/
This is in fact intentional. We aren't strictly following this in our codebase, but when nesting preprocessor macros into ifdefs then we often indent the inner macros with spaces.
Patrick
That's something I didn't know. Thanks.
Unlike borrowed sources in compat/, in our codebase, such indentation is minority. IOW "often indent" -> "sometimes indent".
A quick look at an early part of git-compat-util.h would show that even within a single file we are not consistent at all.
#if __STDC_VERSION__ - 0 < 199901L
#error "Required C99 support is in a test phase. Please see git-compat-util.h for more details." #endif
#ifdef USE_MSVC_CRTDBG
#include <stdlib.h>
#include <crtdbg.h>
#endif
#define _FILE_OFFSET_BITS 64
#if defined(__GNUC__) && defined(__GNUC_MINOR__)
# define GIT_GNUC_PREREQ(maj, min) \ ((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min)) #else #define GIT_GNUC_PREREQ(maj, min) 0 #endif
#ifndef FLEX_ARRAY
#if defined(__SUNPRO_C) && (__SUNPRO_C <= 0x580)
#elif defined(__GNUC__)
# if (__GNUC__ >= 3) # define FLEX_ARRAY /* empty */ # else # define FLEX_ARRAY 0 /* older GNU extension */ # endif #elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) # define FLEX_ARRAY /* empty */ #endif ...
We may want to eventually fix this, but we need to decide what the desirable layout is. I am not sure if the indented version is easier to read and maintain, but one thing that is sure is that a mixed mess is harder than either. In the above excerpt, you cannot tell if I quoted everything related to FLEX_ARRAY (in other words, if "#ifndef FLEX_ARRAY" is already closed in the excerpt) without carefully looking.
Re: [PATCH v3 09/20] global: introduce `USE_THE_REPOSITORY_VARIABLE` macro
Quoted 7 lines · Junio C Hamano
We may want to eventually fix this, but we need to decide what the desirable layout is. I am not sure if the indented version is easier to read and maintain, but one thing that is sure is that a mixed mess is harder than either. In the above excerpt, you cannot tell if I quoted everything related to
FLEX_ARRAY(in other words, if "#ifndefFLEX_ARRAY" is already closed in the excerpt) without carefully looking.
Here is how a patch to reindent (which is full of patch noise and not enough value by itself) looks like.
----- >8 ---------- >8 ---------- >8 ---------- >8 -----
Subject: [PATCH] git-compat-util.h: reindent CPP directives
Karthik noticed that Patrick used a not-so-consistently-used convention to indent the CPP directives inside #if/#else/#endif by one extra space, which may look like a typo to an eye unfamiliar with the convention. Unlike compat/ with borrowed sources, we do not use the convention very much, and even when it is used, we are very inconsistent.
Taking git-compat-util.h as an example, here is how a "normalized" version of that file would look like. It makes it immediately obvious that an unindented CPP directive (like #define and #include) is in effect for any and all architectures, so there might be a readability value in such a change, even though the one-time patch noise is a bit annoying.
---
use strict;
use warnings;
my $level = 0;
my %levelchange = (
if => [0, 1], ifdef => [0, 1], ifndef => [0, 1], else => [-1, 0], elif => [-1, 0], endif => [-1, -1],
);
my $file_name = undef;
while (<>) {
if (!defined $file_name || $ARGV ne $file_name) {
$. = 0;
$file_name = $ARGV;
# Special case for the header files
if ($file_name =~ /\.h$/) {
# cause early part of a header file
# #ifndef FILENAME_H
# #define FILENAME_H
# and the corresponding #endif at the end
# not to be indented.
$level = -1;
} else {
$level = 0;
}
}
# Not a CPP directive? copy it out.
if (!/^\s*#\s*(\S+)(.*)/) {
print;
next;
}
#
my ($what, $rest) = ($1, $2);
my $thislevel = $level;
if (exists $levelchange{$what}) {
$thislevel = $level + $levelchange{$what}[0];
$level = $level + $levelchange{$what}[1];
}
if ($thislevel < 0) {
$thislevel = 0;
}
print "#", " " x $thislevel, "$what$rest\n";
}
---
git-compat-util.h | 528 +++++++++++++++++++++++-----------------------
1 file changed, 264 insertions(+), 264 deletions(-)git-compat-util.h +264 -264
@@ -11,7 +11,7 @@* directive, but please report the details of your system to * [email protected]. */ -#error "Required C99 support is in a test phase. Please see git-compat-util.h for more details." +# error "Required C99 support is in a test phase. Please see git-compat-util.h for more details."#endif #ifdef USE_MSVC_CRTDBG
@@ -19,8 +19,8 @@* For these to work they must appear very early in each * file -- before most of the standard header files. */ -#include <stdlib.h> -#include <crtdbg.h> +# include <stdlib.h> +# include <crtdbg.h> #endif struct strbuf ;
@@ -41,7 +41,7 @@ struct strbuf;# define GIT_GNUC_PREREQ (maj, min) \ ((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min))#else -#define GIT_GNUC_PREREQ (maj, min) 0 +# define GIT_GNUC_PREREQ (maj, min) 0#endif
@@ -57,23 +57,23 @@ struct strbuf;* here, we can fall back to use the "safer but a bit wasteful" one * later. */ -#if defined(__SUNPRO_C) && (__SUNPRO_C <=0x580 ) -#elif defined(__GNUC__) -# if (__GNUC__ >=3 ) +# if defined(__SUNPRO_C) && (__SUNPRO_C <=0x580 ) +# elif defined(__GNUC__) +# if (__GNUC__ >=3 ) +# define FLEX_ARRAY /* empty */ +# else +# define FLEX_ARRAY 0/* older GNU extension */ +# endif +# elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >=199901L )# define FLEX_ARRAY /* empty */ -# else -# define FLEX_ARRAY 0/* older GNU extension */ # endif -#elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >=199901L ) -# define FLEX_ARRAY /* empty */ -#endif /* * Otherwise, default to safer but a bit wasteful traditional style */ -#ifndef FLEX_ARRAY -# define FLEX_ARRAY 1 -#endif +# ifndef FLEX_ARRAY +# define FLEX_ARRAY 1 +# endif #endif
@@ -153,9 +153,9 @@ struct strbuf; (a) > maximum_unsigned_value_of_type(a) >> (shift))#ifdef __GNUC__ -#define TYPEOF (x) (__typeof__(x)) +# define TYPEOF (x) (__typeof__(x))#else -#define TYPEOF (x) +# define TYPEOF (x)#endif #define MSB (x, bits) ((x) & TYPEOF(x)(~0ULL << (bitsizeof(x) - (bits))))
@@ -167,7 +167,7 @@ struct strbuf;#define decimal_length (x) ((int)(sizeof(x) * 2.56 + 0.5) + 1)#ifdef __MINGW64__ -#define _POSIX_C_SOURCE 1 +# define _POSIX_C_SOURCE 1#elif defined(__sun__)/* * On Solaris, when _XOPEN_EXTENDED is set, its header file
@@ -177,16 +177,16 @@ struct strbuf;* non XPG6 programs must be compiled with a pre-c99 compiler. */ # if __STDC_VERSION__ -0 >=199901L -# define _XOPEN_SOURCE 600 +# define _XOPEN_SOURCE 600# else -# define _XOPEN_SOURCE 500 +# define _XOPEN_SOURCE 500# endif #elif !defined(__APPLE__) && !defined(__FreeBSD__) && !defined(__USLC__) && \ !defined(_M_UNIX) && !defined(__sgi) && !defined(__DragonFly__) && \ !defined(__TANDEM) && !defined(__QNX__) && !defined(__MirBSD__) && \ !defined(__CYGWIN__) -#define _XOPEN_SOURCE 600/* glibc2 and AIX 5.3L need 500, OpenBSD needs 600 for S_ISLNK() */ -#define _XOPEN_SOURCE_EXTENDED 1/* AIX 5.3L needs this */ +# define _XOPEN_SOURCE 600/* glibc2 and AIX 5.3L need 500, OpenBSD needs 600 for S_ISLNK() */ +# define _XOPEN_SOURCE_EXTENDED 1/* AIX 5.3L needs this */ #endif #define _ALL_SOURCE 1#define _GNU_SOURCE 1
@@ -196,38 +196,38 @@ struct strbuf;#define _SGI_SOURCE 1#if GIT_GNUC_PREREQ (4 ,5 ) -#define UNUSED __attribute__((unused)) \ +# define UNUSED __attribute__((unused)) \ __attribute__((deprecated ("parameter declared as UNUSED")))#elif defined(__GNUC__) -#define UNUSED __attribute__((unused)) \ +# define UNUSED __attribute__((unused)) \ __attribute__((deprecated))#else -#define UNUSED +# define UNUSED #endif #if defined(WIN32 ) && !defined(__CYGWIN__)/* Both MinGW and MSVC */ # if !defined(_WIN32_WINNT)# define _WIN32_WINNT 0x0600# endif -#define WIN32_LEAN_AND_MEAN /* stops windows.h including winsock.h */ -#include <winsock2.h> -#ifndef NO_UNIX_SOCKETS -#include <afunix.h> -#endif -#include <windows.h> -#define GIT_WINDOWS_NATIVE +# define WIN32_LEAN_AND_MEAN /* stops windows.h including winsock.h */ +# include <winsock2.h> +# ifndef NO_UNIX_SOCKETS +# include <afunix.h> +# endif +# include <windows.h> +# define GIT_WINDOWS_NATIVE #endif #if defined(NO_UNIX_SOCKETS ) || !defined(GIT_WINDOWS_NATIVE )static inline int _have_unix_sockets (void ) { -#if defined(NO_UNIX_SOCKETS ) +# if defined(NO_UNIX_SOCKETS )return 0 ; -#else +# else return 1 ; -#endif +# endif } -#define have_unix_sockets _have_unix_sockets +# define have_unix_sockets _have_unix_sockets#endif #include <unistd.h>
@@ -240,13 +240,13 @@ static inline int _have_unix_sockets(void)#include <stdbool.h> #include <string.h> #ifdef HAVE_STRINGS_H -#include <strings.h> /* for strcasecmp() */ +# include <strings.h> /* for strcasecmp() */ #endif #include <errno.h> #include <limits.h> #include <locale.h> #ifdef NEEDS_SYS_PARAM_H -#include <sys/param.h> +# include <sys/param.h> #endif #include <sys/types.h> #include <dirent.h>
@@ -258,15 +258,15 @@ static inline int _have_unix_sockets(void)#include <utime.h> #include <syslog.h> #if !defined(NO_POLL_H ) -#include <poll.h> +# include <poll.h> #elif !defined(NO_SYS_POLL_H ) -#include <sys/poll.h> +# include <sys/poll.h> #else /* Pull the compat stuff */ -#include <poll.h> +# include <poll.h> #endif #ifdef HAVE_BSD_SYSCTL -#include <sys/sysctl.h> +# include <sys/sysctl.h> #endif /* Used by compat/win32/path-utils.h, and more */
@@ -276,44 +276,44 @@ static inline int is_xplatform_dir_sep(int c) }#if defined(__CYGWIN__) -#include "compat/win32/path-utils.h" +# include "compat/win32/path-utils.h" #endif #if defined(__MINGW32__)/* pull in Windows compatibility stuff */ -#include "compat/win32/path-utils.h" -#include "compat/mingw.h" +# include "compat/win32/path-utils.h" +# include "compat/mingw.h" #elif defined(_MSC_VER) -#include "compat/win32/path-utils.h" -#include "compat/msvc.h" -#else -#include <sys/utsname.h> -#include <sys/wait.h> -#include <sys/resource.h> -#include <sys/socket.h> -#include <sys/ioctl.h> -#include <sys/statvfs.h> -#include <termios.h> -#ifndef NO_SYS_SELECT_H -#include <sys/select.h> -#endif -#include <netinet/in.h> -#include <netinet/tcp.h> -#include <arpa/inet.h> -#include <netdb.h> -#include <pwd.h> -#include <sys/un.h> -#ifndef NO_INTTYPES_H -#include <inttypes.h> +# include "compat/win32/path-utils.h" +# include "compat/msvc.h" #else -#include <stdint.h> -#endif -#ifdef HAVE_ARC4RANDOM_LIBBSD -#include <bsd/stdlib.h> -#endif -#ifdef HAVE_GETRANDOM -#include <sys/random.h> -#endif -#ifdef NO_INTPTR_T +# include <sys/utsname.h> +# include <sys/wait.h> +# include <sys/resource.h> +# include <sys/socket.h> +# include <sys/ioctl.h> +# include <sys/statvfs.h> +# include <termios.h> +# ifndef NO_SYS_SELECT_H +# include <sys/select.h> +# endif +# include <netinet/in.h> +# include <netinet/tcp.h> +# include <arpa/inet.h> +# include <netdb.h> +# include <pwd.h> +# include <sys/un.h> +# ifndef NO_INTTYPES_H +# include <inttypes.h> +# else +# include <stdint.h> +# endif +# ifdef HAVE_ARC4RANDOM_LIBBSD +# include <bsd/stdlib.h> +# endif +# ifdef HAVE_GETRANDOM +# include <sys/random.h> +# endif +# ifdef NO_INTPTR_T /* * On I16LP32, ILP32 and LP64 "long" is the safe bet, however * on LLP86, IL33LLP64 and P64 it needs to be "long long",
@@ -323,15 +323,15 @@ static inline int is_xplatform_dir_sep(int c)*/ typedef long intptr_t ;typedef unsigned long uintptr_t ; -#endif -#undef _ALL_SOURCE/* AIX 5.3L defines a struct list with _ALL_SOURCE. */ -#include <grp.h> -#define _ALL_SOURCE 1 +# endif +# undef _ALL_SOURCE/* AIX 5.3L defines a struct list with _ALL_SOURCE. */ +# include <grp.h> +# define _ALL_SOURCE 1#endif /* used on Mac OS X */ #ifdef PRECOMPOSE_UNICODE -#include "compat/precompose_utf8.h" +# include "compat/precompose_utf8.h" #else static inline const char *precompose_argv_prefix(int argc UNUSED, const char **argv UNUSED,
@@ -344,16 +344,16 @@ static inline const char *precompose_string_if_needed(const char *in) return in; } -#define probe_utf8_pathname_composition () +# define probe_utf8_pathname_composition ()#endif #ifdef MKDIR_WO_TRAILING_SLASH -#define mkdir (a,b) compat_mkdir_wo_trailing_slash((a),(b)) +# define mkdir (a,b) compat_mkdir_wo_trailing_slash((a),(b))int compat_mkdir_wo_trailing_slash (const char *,mode_t );#endif #ifdef time -#undef time +# undef time#endif static inline time_t git_time (time_t *tloc) {
@@ -384,34 +384,34 @@ static inline int git_setitimer(int which UNUSED, struct itimerval *newvalue UNUSED) {return 0 ;/* pretend success */ } -#undef setitimer -#define setitimer (which,value,ovalue) git_setitimer(which,value,ovalue) +# undef setitimer +# define setitimer (which,value,ovalue) git_setitimer(which,value,ovalue)#endif #ifndef NO_LIBGEN_H -#include <libgen.h> +# include <libgen.h> #else -#define basename gitbasename +# define basename gitbasenamechar *gitbasename (char *); -#define dirname gitdirname +# define dirname gitdirnamechar *gitdirname (char *);#endif #ifndef NO_ICONV -#include <iconv.h> +# include <iconv.h> #endif #ifndef NO_OPENSSL -#ifdef __APPLE__ -#undef __AVAILABILITY_MACROS_USES_AVAILABILITY -#define __AVAILABILITY_MACROS_USES_AVAILABILITY 0 -#include <AvailabilityMacros.h> -#undef DEPRECATED_ATTRIBUTE -#define DEPRECATED_ATTRIBUTE -#undef __AVAILABILITY_MACROS_USES_AVAILABILITY -#endif -#include <openssl/ssl.h> -#include <openssl/err.h> +# ifdef __APPLE__ +# undef __AVAILABILITY_MACROS_USES_AVAILABILITY +# define __AVAILABILITY_MACROS_USES_AVAILABILITY 0 +# include <AvailabilityMacros.h> +# undef DEPRECATED_ATTRIBUTE +# define DEPRECATED_ATTRIBUTE +# undef __AVAILABILITY_MACROS_USES_AVAILABILITY +# endif +# include <openssl/ssl.h> +# include <openssl/err.h> #endif #ifdef HAVE_SYSINFO
@@ -422,22 +422,22 @@ char *gitdirname(char *);* not on some systems (e.g. z/OS). */ #ifndef NI_MAXHOST -#define NI_MAXHOST 1025 +# define NI_MAXHOST 1025#endif #ifndef NI_MAXSERV -#define NI_MAXSERV 32 +# define NI_MAXSERV 32#endif /* On most systems <limits.h> would have given us this, but * not on some systems (e.g. GNU/Hurd). */ #ifndef PATH_MAX -#define PATH_MAX 4096 +# define PATH_MAX 4096#endif #ifndef NAME_MAX -#define NAME_MAX 255 +# define NAME_MAX 255#endif typedef uintmax_t timestamp_t ;
@@ -447,14 +447,14 @@ typedef uintmax_t timestamp_t;#define TIME_MIN 0#ifndef PATH_SEP -#define PATH_SEP ':' +# define PATH_SEP ':'#endif #ifdef HAVE_PATHS_H -#include <paths.h> +# include <paths.h> #endif #ifndef _PATH_DEFPATH -#define _PATH_DEFPATH "/usr/local/bin:/usr/bin:/bin" +# define _PATH_DEFPATH "/usr/local/bin:/usr/bin:/bin"#endif #ifndef platform_core_config
@@ -466,12 +466,12 @@ static inline int noop_core_config(const char *var UNUSED, {return 0 ; } -#define platform_core_config noop_core_config +# define platform_core_config noop_core_config#endif int lstat_cache_aware_rmdir (const char *path);#if !defined(__MINGW32__) && !defined(_MSC_VER) -#define rmdir lstat_cache_aware_rmdir +# define rmdir lstat_cache_aware_rmdir#endif #ifndef has_dos_drive_prefix
@@ -479,7 +479,7 @@ static inline int git_has_dos_drive_prefix(const char *path UNUSED) {return 0 ; } -#define has_dos_drive_prefix git_has_dos_drive_prefix +# define has_dos_drive_prefix git_has_dos_drive_prefix#endif #ifndef skip_dos_drive_prefix
@@ -487,7 +487,7 @@ static inline int git_skip_dos_drive_prefix(char **path UNUSED) {return 0 ; } -#define skip_dos_drive_prefix git_skip_dos_drive_prefix +# define skip_dos_drive_prefix git_skip_dos_drive_prefix#endif static inline int git_is_dir_sep (int c)
@@ -495,7 +495,7 @@ static inline int git_is_dir_sep(int c)return c =='/' ; }#ifndef is_dir_sep -#define is_dir_sep git_is_dir_sep +# define is_dir_sep git_is_dir_sep#endif #ifndef offset_1st_component
@@ -503,20 +503,20 @@ static inline int git_offset_1st_component(const char *path) {return is_dir_sep (path[0 ]); } -#define offset_1st_component git_offset_1st_component +# define offset_1st_component git_offset_1st_component#endif #ifndef is_valid_path -#define is_valid_path (path) 1 +# define is_valid_path (path) 1#endif #ifndef is_path_owned_by_current_user -#ifdef __TANDEM -#define ROOT_UID 65535 -#else -#define ROOT_UID 0 -#endif +# ifdef __TANDEM +# define ROOT_UID 65535 +# else +# define ROOT_UID 0 +# endif /* * Do not use this function when
@@ -576,7 +576,7 @@ static inline int is_path_owned_by_current_uid(const char *path, return st.st_uid == euid; } -#define is_path_owned_by_current_user is_path_owned_by_current_uid +# define is_path_owned_by_current_user is_path_owned_by_current_uid#endif #ifndef find_last_dir_sep
@@ -584,7 +584,7 @@ static inline char *git_find_last_dir_sep(const char *path) {return strrchr (path,'/' ); } -#define find_last_dir_sep git_find_last_dir_sep +# define find_last_dir_sep git_find_last_dir_sep#endif #ifndef has_dir_sep
@@ -592,53 +592,53 @@ static inline int git_has_dir_sep(const char *path) {return !!strchr (path,'/' ); } -#define has_dir_sep (path) git_has_dir_sep(path) +# define has_dir_sep (path) git_has_dir_sep(path)#endif #ifndef query_user_email -#define query_user_email () NULL +# define query_user_email () NULL#endif #ifdef __TANDEM -#include <floss.h(floss_execl,floss_execlp,floss_execv,floss_execvp)> -#include <floss.h(floss_getpwuid)> -#ifndef NSIG +# include <floss.h(floss_execl,floss_execlp,floss_execv,floss_execvp)> +# include <floss.h(floss_getpwuid)> +# ifndef NSIG /* * NonStop NSE and NSX do not provide NSIG. SIGGUARDIAN(99) is the highest * known, by detective work using kill -l as a list is all signals * instead of signal.h where it should be. */ -# define NSIG 100 -#endif +# define NSIG 100 +# endif #endif #if defined(__HP_cc) && (__HP_cc >=61000 ) -#define NORETURN __attribute__((noreturn)) -#define NORETURN_PTR +# define NORETURN __attribute__((noreturn)) +# define NORETURN_PTR #elif defined(__GNUC__) && !defined(NO_NORETURN ) -#define NORETURN __attribute__((__noreturn__)) -#define NORETURN_PTR __attribute__((__noreturn__)) +# define NORETURN __attribute__((__noreturn__)) +# define NORETURN_PTR __attribute__((__noreturn__))#elif defined(_MSC_VER) -#define NORETURN __declspec(noreturn) -#define NORETURN_PTR +# define NORETURN __declspec(noreturn) +# define NORETURN_PTR #else -#define NORETURN -#define NORETURN_PTR -#ifndef __GNUC__ -#ifndef __attribute__ -#define __attribute__ (x) -#endif -#endif +# define NORETURN +# define NORETURN_PTR +# ifndef __GNUC__ +# ifndef __attribute__ +# define __attribute__ (x) +# endif +# endif #endif /* The sentinel attribute is valid from gcc version 4.0 */ #if defined(__GNUC__) && (__GNUC__ >=4 ) -#define LAST_ARG_MUST_BE_NULL __attribute__((sentinel)) +# define LAST_ARG_MUST_BE_NULL __attribute__((sentinel))/* warn_unused_result exists as of gcc 3.4.0, but be lazy and check 4.0 */ -#define RESULT_MUST_BE_USED __attribute__ ((warn_unused_result)) +# define RESULT_MUST_BE_USED __attribute__ ((warn_unused_result))#else -#define LAST_ARG_MUST_BE_NULL -#define RESULT_MUST_BE_USED +# define LAST_ARG_MUST_BE_NULL +# define RESULT_MUST_BE_USED #endif #define MAYBE_UNUSED __attribute__((__unused__))
@@ -660,17 +660,17 @@ void warning(const char *err, ...) __attribute__((format (printf, 1, 2)));void warning_errno (const char *err, ...) __attribute__((format (printf,1 ,2 )));#ifndef NO_OPENSSL -#ifdef APPLE_COMMON_CRYPTO -#include "compat/apple-common-crypto.h" -#else -#include <openssl/evp.h> -#include <openssl/hmac.h> -#endif /* APPLE_COMMON_CRYPTO */ -#include <openssl/x509v3.h> +# ifdef APPLE_COMMON_CRYPTO +# include "compat/apple-common-crypto.h" +# else +# include <openssl/evp.h> +# include <openssl/hmac.h> +# endif /* APPLE_COMMON_CRYPTO */ +# include <openssl/x509v3.h> #endif /* NO_OPENSSL */ #ifdef HAVE_OPENSSL_CSPRNG -#include <openssl/rand.h> +# include <openssl/rand.h> #endif /*
@@ -683,8 +683,8 @@ static inline int const_error(void) {return -1 ; } -#define error (...) (error(__VA_ARGS__), const_error()) -#define error_errno (...) (error_errno(__VA_ARGS__), const_error()) +# define error (...) (error(__VA_ARGS__), const_error()) +# define error_errno (...) (error_errno(__VA_ARGS__), const_error())#endif typedef void (*report_fn )(const char *,va_list params);
@@ -782,32 +782,32 @@ static inline bool strip_suffix(const char *str, const char *suffix,#if defined(NO_MMAP ) || defined(USE_WIN32_MMAP ) -#ifndef PROT_READ -#define PROT_READ 1 -#define PROT_WRITE 2 -#define MAP_PRIVATE 1 -#endif +# ifndef PROT_READ +# define PROT_READ 1 +# define PROT_WRITE 2 +# define MAP_PRIVATE 1 +# endif -#define mmap git_mmap -#define munmap git_munmap +# define mmap git_mmap +# define munmap git_munmapvoid *git_mmap (void *start,size_t length,int prot,int flags,int fd,off_t offset);int git_munmap (void *start,size_t length);#else /* NO_MMAP || USE_WIN32_MMAP */ -#include <sys/mman.h> +# include <sys/mman.h> #endif /* NO_MMAP || USE_WIN32_MMAP */ #ifdef NO_MMAP /* This value must be multiple of (pagesize * 2) */ -#define DEFAULT_PACKED_GIT_WINDOW_SIZE (1 * 1024 * 1024) +# define DEFAULT_PACKED_GIT_WINDOW_SIZE (1 * 1024 * 1024)#else /* NO_MMAP */ /* This value must be multiple of (pagesize * 2) */ -#define DEFAULT_PACKED_GIT_WINDOW_SIZE \ +# define DEFAULT_PACKED_GIT_WINDOW_SIZE \ (sizeof(void*) >= 8 \ ? 1 * 1024 * 1024 * 1024 \ : 32 * 1024 * 1024)
@@ -815,46 +815,46 @@ int git_munmap(void *start, size_t length);#endif /* NO_MMAP */ #ifndef MAP_FAILED -#define MAP_FAILED ((void *)-1) +# define MAP_FAILED ((void *)-1)#endif #ifdef NO_ST_BLOCKS_IN_STRUCT_STAT -#define on_disk_bytes (st) ((st).st_size) +# define on_disk_bytes (st) ((st).st_size)#else -#define on_disk_bytes (st) ((st).st_blocks * 512) +# define on_disk_bytes (st) ((st).st_blocks * 512)#endif #ifdef NEEDS_MODE_TRANSLATION -#undef S_IFMT -#undef S_IFREG -#undef S_IFDIR -#undef S_IFLNK -#undef S_IFBLK -#undef S_IFCHR -#undef S_IFIFO -#undef S_IFSOCK -#define S_IFMT 0170000 -#define S_IFREG 0100000 -#define S_IFDIR 0040000 -#define S_IFLNK 0120000 -#define S_IFBLK 0060000 -#define S_IFCHR 0020000 -#define S_IFIFO 0010000 -#define S_IFSOCK 0140000 -#ifdef stat -#undef stat -#endif -#define stat (path, buf) git_stat(path, buf) +# undef S_IFMT +# undef S_IFREG +# undef S_IFDIR +# undef S_IFLNK +# undef S_IFBLK +# undef S_IFCHR +# undef S_IFIFO +# undef S_IFSOCK +# define S_IFMT 0170000 +# define S_IFREG 0100000 +# define S_IFDIR 0040000 +# define S_IFLNK 0120000 +# define S_IFBLK 0060000 +# define S_IFCHR 0020000 +# define S_IFIFO 0010000 +# define S_IFSOCK 0140000 +# ifdef stat +# undef stat +# endif +# define stat (path, buf) git_stat(path, buf)int git_stat (const char *,struct stat *); -#ifdef fstat -#undef fstat -#endif -#define fstat (fd, buf) git_fstat(fd, buf) +# ifdef fstat +# undef fstat +# endif +# define fstat (fd, buf) git_fstat(fd, buf)int git_fstat (int ,struct stat *); -#ifdef lstat -#undef lstat -#endif -#define lstat (path, buf) git_lstat(path, buf) +# ifdef lstat +# undef lstat +# endif +# define lstat (path, buf) git_lstat(path, buf)int git_lstat (const char *,struct stat *);#endif
@@ -862,67 +862,67 @@ int git_lstat(const char *, struct stat *); ((1024L * 1024L) * (size_t)(sizeof(void*) >= 8 ? (32 * 1024L * 1024L) : 256))#ifdef NO_PREAD -#define pread git_pread +# define pread git_preadssize_t git_pread (int fd,void *buf,size_t count,off_t offset);#endif #ifdef NO_SETENV -#define setenv gitsetenv +# define setenv gitsetenvint gitsetenv (const char *,const char *,int );#endif #ifdef NO_MKDTEMP -#define mkdtemp gitmkdtemp +# define mkdtemp gitmkdtempchar *gitmkdtemp (char *);#endif #ifdef NO_UNSETENV -#define unsetenv gitunsetenv +# define unsetenv gitunsetenvint gitunsetenv (const char *);#endif #ifdef NO_STRCASESTR -#define strcasestr gitstrcasestr +# define strcasestr gitstrcasestrchar *gitstrcasestr (const char *haystack,const char *needle);#endif #ifdef NO_STRLCPY -#define strlcpy gitstrlcpy +# define strlcpy gitstrlcpysize_t gitstrlcpy (char *,const char *,size_t );#endif #ifdef NO_STRTOUMAX -#define strtoumax gitstrtoumax +# define strtoumax gitstrtoumaxuintmax_t gitstrtoumax (const char *,char **,int ); -#define strtoimax gitstrtoimax +# define strtoimax gitstrtoimaxintmax_t gitstrtoimax (const char *,char **,int );#endif #ifdef NO_HSTRERROR -#define hstrerror githstrerror +# define hstrerror githstrerrorconst char *githstrerror (int herror);#endif #ifdef NO_MEMMEM -#define memmem gitmemmem +# define memmem gitmemmemvoid *gitmemmem (const void *haystack,size_t haystacklen,const void *needle,size_t needlelen);#endif #ifdef OVERRIDE_STRDUP -#ifdef strdup -#undef strdup -#endif -#define strdup gitstrdup +# ifdef strdup +# undef strdup +# endif +# define strdup gitstrdupchar *gitstrdup (const char *s);#endif #ifdef NO_GETPAGESIZE -#define getpagesize () sysconf(_SC_PAGESIZE) +# define getpagesize () sysconf(_SC_PAGESIZE)#endif #ifndef O_CLOEXEC -#define O_CLOEXEC 0 +# define O_CLOEXEC 0#endif #ifdef FREAD_READS_DIRECTORIES
@@ -936,34 +936,34 @@ FILE *git_fopen(const char*, const char*);#endif #ifdef SNPRINTF_RETURNS_BOGUS -#ifdef snprintf -#undef snprintf -#endif -#define snprintf git_snprintf +# ifdef snprintf +# undef snprintf +# endif +# define snprintf git_snprintfint git_snprintf (char *str,size_t maxsize,const char *format, ...); -#ifdef vsnprintf -#undef vsnprintf -#endif -#define vsnprintf git_vsnprintf +# ifdef vsnprintf +# undef vsnprintf +# endif +# define vsnprintf git_vsnprintfint git_vsnprintf (char *str,size_t maxsize,const char *format,va_list ap);#endif #ifdef OPEN_RETURNS_EINTR -#undef open -#define open git_open_with_retry +# undef open +# define open git_open_with_retryint git_open_with_retry (const char *path,int flag, ...);#endif #ifdef __GLIBC_PREREQ -#if __GLIBC_PREREQ (2 ,1 ) -#define HAVE_STRCHRNUL -#endif +# if __GLIBC_PREREQ (2 ,1 ) +# define HAVE_STRCHRNUL +# endif #endif #ifndef HAVE_STRCHRNUL -#define strchrnul gitstrchrnul +# define strchrnul gitstrchrnulstatic inline char *gitstrchrnul (const char *s,int c) {while (*s && *s != c)
@@ -981,7 +981,7 @@ const char *inet_ntop(int af, const void *src, char *dst, size_t size);#endif #ifdef NO_PTHREADS -#define atexit git_atexit +# define atexit git_atexitint git_atexit (void (*handler)(void ));#endif
@@ -1252,7 +1252,7 @@ static inline size_t xsize_t(off_t len) }#ifndef HOST_NAME_MAX -#define HOST_NAME_MAX 256 +# define HOST_NAME_MAX 256#endif #include "sane-ctype.h"
@@ -1325,7 +1325,7 @@ static inline int strtol_i(char const *s, int base, int *result)void git_stable_qsort (void *base,size_t nmemb,size_t size,int (*compar)(const void *,const void *));#ifdef INTERNAL_QSORT -#define qsort git_stable_qsort +# define qsort git_stable_qsort#endif #define QSORT (base, n, compar) sane_qsort((base), (n), sizeof(*(base)), compar)
@@ -1342,7 +1342,7 @@ static inline void sane_qsort(void *base, size_t nmemb, size_t size,#ifndef HAVE_ISO_QSORT_S int git_qsort_s (void *base,size_t nmemb,size_t size,int (*compar)(const void *,const void *,void *),void *ctx); -#define qsort_s git_qsort_s +# define qsort_s git_qsort_s#endif #define QSORT_S (base, n, compar, ctx) do { \
@@ -1351,7 +1351,7 @@ int git_qsort_s(void *base, size_t nmemb, size_t size, } while (0)#ifndef REG_STARTEND -#error "Git requires REG_STARTEND support. Compile with NO_REGEX=NeedsStartEnd" +# error "Git requires REG_STARTEND support. Compile with NO_REGEX=NeedsStartEnd"#endif static inline int regexec_buf(const regex_t *preg, const char *buf, size_t size,
@@ -1365,7 +1365,7 @@ static inline int regexec_buf(const regex_t *preg, const char *buf, size_t size,#ifdef USE_ENHANCED_BASIC_REGULAR_EXPRESSIONS int git_regcomp (regex_t *preg,const char *pattern,int cflags); -#define regcomp git_regcomp +# define regcomp git_regcomp#endif #ifndef DIR_HAS_BSD_GROUP_SEMANTICS
@@ -1375,23 +1375,23 @@ int git_regcomp(regex_t *preg, const char *pattern, int cflags);#endif #ifdef NO_NSEC -#undef USE_NSEC -#define ST_CTIME_NSEC (st) 0 -#define ST_MTIME_NSEC (st) 0 +# undef USE_NSEC +# define ST_CTIME_NSEC (st) 0 +# define ST_MTIME_NSEC (st) 0#else -#ifdef USE_ST_TIMESPEC -#define ST_CTIME_NSEC (st) ((unsigned int)((st).st_ctimespec.tv_nsec)) -#define ST_MTIME_NSEC (st) ((unsigned int)((st).st_mtimespec.tv_nsec)) -#else -#define ST_CTIME_NSEC (st) ((unsigned int)((st).st_ctim.tv_nsec)) -#define ST_MTIME_NSEC (st) ((unsigned int)((st).st_mtim.tv_nsec)) -#endif +# ifdef USE_ST_TIMESPEC +# define ST_CTIME_NSEC (st) ((unsigned int)((st).st_ctimespec.tv_nsec)) +# define ST_MTIME_NSEC (st) ((unsigned int)((st).st_mtimespec.tv_nsec)) +# else +# define ST_CTIME_NSEC (st) ((unsigned int)((st).st_ctim.tv_nsec)) +# define ST_MTIME_NSEC (st) ((unsigned int)((st).st_mtim.tv_nsec)) +# endif #endif #ifdef UNRELIABLE_FSTAT -#define fstat_is_reliable () 0 +# define fstat_is_reliable () 0#else -#define fstat_is_reliable () 1 +# define fstat_is_reliable () 1#endif #ifndef va_copy
@@ -1400,11 +1400,11 @@ int git_regcomp(regex_t *preg, const char *pattern, int cflags);* pointer into the stack frame, a simple assignment will work on * many systems. But let's try to be more portable. */ -#ifdef __va_copy -#define va_copy (dst, src) __va_copy(dst, src) -#else -#define va_copy (dst, src) ((dst) = (src)) -#endif +# ifdef __va_copy +# define va_copy (dst, src) __va_copy(dst, src) +# else +# define va_copy (dst, src) ((dst) = (src)) +# endif #endif /* usage.c: only to be used for testing BUG() implementation (see test-tool) */
@@ -1425,11 +1425,11 @@ void bug_fl(const char *file, int line, const char *fmt, ...); } while (0)#ifndef FSYNC_METHOD_DEFAULT -#ifdef __APPLE__ -#define FSYNC_METHOD_DEFAULT FSYNC_METHOD_WRITEOUT_ONLY -#else -#define FSYNC_METHOD_DEFAULT FSYNC_METHOD_FSYNC -#endif +# ifdef __APPLE__ +# define FSYNC_METHOD_DEFAULT FSYNC_METHOD_WRITEOUT_ONLY +# else +# define FSYNC_METHOD_DEFAULT FSYNC_METHOD_FSYNC +# endif #endif #ifndef SHELL_PATH
@@ -1445,12 +1445,12 @@ static inline void git_funlockfile(FILE *fh UNUSED) { ;/* nothing */ } -#undef flockfile -#undef funlockfile -#undef getc_unlocked -#define flockfile (fh) git_flockfile(fh) -#define funlockfile (fh) git_funlockfile(fh) -#define getc_unlocked (fh) getc(fh) +# undef flockfile +# undef funlockfile +# undef getc_unlocked +# define flockfile (fh) git_flockfile(fh) +# define funlockfile (fh) git_funlockfile(fh) +# define getc_unlocked (fh) getc(fh)#endif #ifdef FILENO_IS_A_MACRO
@@ -1465,7 +1465,7 @@ int git_fileno(FILE *stream);int git_access (const char *path,int mode);# ifndef COMPAT_CODE_ACCESS # ifdef access -# undef access +# undef access# endif # define access (path, mode) git_access(path, mode)# endif
@@ -1510,9 +1510,9 @@ int common_exit(const char *file, int line, int code);*/ #ifdef SUPPRESS_ANNOTATED_LEAKS void unleak_memory (const void *ptr,size_t len); -#define UNLEAK (var) unleak_memory(&(var), sizeof(var)) +# define UNLEAK (var) unleak_memory(&(var), sizeof(var))#else -#define UNLEAK (var) do {} while (0) +# define UNLEAK (var) do {} while (0)#endif #define z_const
@@ -1565,9 +1565,9 @@ static inline void *container_of_or_null_offset(void *ptr, size_t offset)* everywhere. */ #if defined(__GNUC__)/* clang sets this, too */ -#define OFFSETOF_VAR (ptr, member) offsetof(__typeof__(*ptr), member) +# define OFFSETOF_VAR (ptr, member) offsetof(__typeof__(*ptr), member)#else /* !__GNUC__ */ -#define OFFSETOF_VAR (ptr, member) \ +# define OFFSETOF_VAR (ptr, member) \ ((uintptr_t)&(ptr)->member - (uintptr_t)(ptr))#endif /* !__GNUC__ */
Signature · 1 line
2.45.2-709-g1ffe7c8bd7
Re: [PATCH v3 09/20] global: introduce `USE_THE_REPOSITORY_VARIABLE` macro
In reply to Junio C HamanoQuoted 50 lines · Junio C Hamano
Quoted 9 lines · Karthik Nayak
Quoted 7 lines
Quoted 1 line
s/# define/#define/
This is in fact intentional. We aren't strictly following this in our codebase, but when nesting preprocessor macros into ifdefs then we often indent the inner macros with spaces.
Patrick
That's something I didn't know. Thanks.
Unlike borrowed sources in compat/, in our codebase, such indentation is minority. IOW "often indent" -> "sometimes indent".
A quick look at an early part of
git-compat-util.hwould show that even within a single file we are not consistent at all.#if
__STDC_VERSION__- 0 < 199901L
#error "Required C99 support is in a test phase. Please seegit-compat-util.hfor more details." #endif#ifdef
USE_MSVC_CRTDBG
#include <stdlib.h>
#include <crtdbg.h>
#endif#define
_FILE_OFFSET_BITS64#if
defined(__GNUC__)&&defined(__GNUC_MINOR__)# define GIT_GNUC_PREREQ(maj, min) \ ((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min)) #else #define GIT_GNUC_PREREQ(maj, min) 0 #endif#ifndef
FLEX_ARRAY
#ifdefined(__SUNPRO_C)&& (__SUNPRO_C<=0x580)
#elifdefined(__GNUC__)# if (__GNUC__ >= 3) # define FLEX_ARRAY /* empty */ # else # define FLEX_ARRAY 0 /* older GNU extension */ # endif #elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) # define FLEX_ARRAY /* empty */ #endif ...
Quoted 7 lines
We may want to eventually fix this, but we need to decide what the desirable layout is. I am not sure if the indented version is easier to read and maintain, but one thing that is sure is that a mixed mess is harder than either. In the above excerpt, you cannot tell if I quoted everything related to
FLEX_ARRAY(in other words, if "#ifndefFLEX_ARRAY" is already closed in the excerpt) without carefully looking.
Perhaps putting the options next to each other is good way to think of about the _desired layout_.
#ifndef FLEX_ARRAY
#if defined(__SUNPRO_C) && (__SUNPRO_C <= 0x580)
#elif defined(__GNUC__)
# if (__GNUC__ >= 3) # define FLEX_ARRAY /* empty */ # else # define FLEX_ARRAY 0 /* older GNU extension */ # endif #elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) # define FLEX_ARRAY /* empty */ #endif
#ifndef FLEX_ARRAY
#if defined(__SUNPRO_C) && (__SUNPRO_C <= 0x580)
#elif defined(__GNUC__)
#if (__GNUC__ >= 3)
#define FLEX_ARRAY /* empty */
#else
#define FLEX_ARRAY 0 /* older GNU extension */ #endif
#elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)
#define FLEX_ARRAY /* empty */
#endif
Somehow I feel the latter makes it a bit easier to grasp blocks, but overall I think its more important that we pick one and perhaps add it to 'Documentation/CodingGuidelines'.
Also, I did look at your response patch, but will leave it for someone with perl experience to review.
PATCH v3 10/20 refs: avoid include cycle with "repository.h"
In reply to Patrick SteinhardtThere is an include cycle between "refs.h" and "repository.h" via "commit.h", "object.h" and "hash.h". This has the effect that several definitions of structs and enums will not be visible once we merge "hash-ll.h" back into "hash.h" in the next commit.
The only reason that "repository.h" includes "refs.h" is the definition of enum ref_storage_format. Move it into "repository.h" and have "refs.h" include "repository.h" instead to fix the cycle.
--- refs.h | 8 +------- repository.h | 7 ++++++- 2 files changed, 7 insertions(+), 8 deletions(-)
refs.h +1 -7
@@ -2,21 +2,15 @@#define REFS_H #include "commit.h" +#include "repository.h" struct object_id ;struct ref_store ; -struct repository ;struct strbuf ;struct string_list ;struct string_list_item ;struct worktree ; -enum ref_storage_format { -REF_STORAGE_FORMAT_UNKNOWN , -REF_STORAGE_FORMAT_FILES , -REF_STORAGE_FORMAT_REFTABLE , -}; -enum ref_storage_format ref_storage_format_by_name (const char *name);const char *ref_storage_format_to_name (enum ref_storage_format ref_storage_format);
repository.h +6 -1
@@ -1,7 +1,6 @@#ifndef REPOSITORY_H #define REPOSITORY_H -#include "refs.h" #include "strmap.h" struct config_set ;
@@ -27,6 +26,12 @@ enum fetch_negotiation_setting { FETCH_NEGOTIATION_NOOP, }; +enum ref_storage_format { +REF_STORAGE_FORMAT_UNKNOWN , +REF_STORAGE_FORMAT_FILES , +REF_STORAGE_FORMAT_REFTABLE , +}; +struct repo_settings {int initialized;
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 11/20 hash-ll: merge with "hash.h"
In reply to Patrick SteinhardtThe "hash-ll.h" header was introduced via d1cbe1e6d8 (hash-ll.h: split out of hash.h to remove dependency on repository.h, 2023-04-22) to make explicit the split between hash-related functions that rely on the global the_repository, and those that don't. This split is no longer necessary now that we we have removed the reliance on the_repository.
Merge "hash-ll.h" back into "hash.h". This causes some code units to not include "repository.h" anymore, which requires us to add some forward declarations.
--- apply.h | 2 +- bloom.c | 1 + checkout.h | 2 +- chunk-format.h | 2 +- commit-graph.h | 2 + compat/sha1-chunked.c | 2 +- convert.h | 2 +- csum-file.h | 2 +- diff.h | 2 +- diffcore.h | 2 +- dir.h | 2 +- hash-ll.h | 364 ------------------------------------- hash.h | 362 +++++++++++++++++++++++++++++++++++- hex.h | 2 +- loose.h | 2 + merge-ort.h | 2 +- object-file-convert.c | 2 +- object.h | 2 +- oidmap.h | 2 +- oidtree.h | 2 +- packfile.h | 2 + protocol-caps.c | 2 +- read-cache-ll.h | 2 +- refs/ref-cache.h | 2 +- reftable/dump.c | 2 +- reftable/reftable-record.h | 2 +- reftable/system.h | 2 +- remote.h | 2 +- reset.h | 2 +- resolve-undo.h | 2 +- serve.c | 2 +- split-index.h | 2 +- t/helper/test-hash-speed.c | 2 +- t/helper/test-sha1.c | 2 +- t/helper/test-sha256.c | 2 +- t/unit-tests/lib-oid.h | 2 +- tree-diff.c | 1 + tree-walk.h | 2 +- xdiff-interface.h | 2 +- 39 files changed, 400 insertions(+), 398 deletions(-) delete mode 100644 hash-ll.h
apply.h +1 -1
@@ -1,7 +1,7 @@#ifndef APPLY_H #define APPLY_H -#include "hash-ll.h" +#include "hash.h" #include "lockfile.h" #include "string-list.h" #include "strmap.h"
bloom.c +1 -0
@@ -6,6 +6,7 @@#include "commit-graph.h" #include "commit.h" #include "commit-slab.h" +#include "repository.h" define_commit_slab(bloom_filter_slab, struct bloom_filter);
checkout.h +1 -1
@@ -1,7 +1,7 @@#ifndef CHECKOUT_H #define CHECKOUT_H -#include "hash-ll.h" +#include "hash.h" /* * Check if the branch name uniquely matches a branch name on a remote
chunk-format.h +1 -1
@@ -1,7 +1,7 @@#ifndef CHUNK_FORMAT_H #define CHUNK_FORMAT_H -#include "hash-ll.h" +#include "hash.h" struct hashfile ;struct chunkfile ;
commit-graph.h +2 -0
@@ -122,6 +122,8 @@ struct commit_graph *load_commit_graph_chain_fd_st(struct repository *r,struct commit_graph *read_commit_graph_one (struct repository *r,struct object_directory *odb); +struct repo_settings ; +/* * Callers should initialize the repo_settings with prepare_repo_settings() * prior to calling parse_commit_graph().
compat/sha1-chunked.c +1 -1
@@ -1,5 +1,5 @@#include "git-compat-util.h" -#include "hash-ll.h" +#include "hash.h" int git_SHA1_Update_Chunked (platform_SHA_CTX *c,const void *data,size_t len) {
convert.h +1 -1
@@ -4,7 +4,7 @@#ifndef CONVERT_H #define CONVERT_H -#include "hash-ll.h" +#include "hash.h" #include "string-list.h" struct index_state ;
csum-file.h +1 -1
@@ -1,7 +1,7 @@#ifndef CSUM_FILE_H #define CSUM_FILE_H -#include "hash-ll.h" +#include "hash.h" #include "write-or-die.h" struct progress ;
diff.h +1 -1
@@ -4,7 +4,7 @@#ifndef DIFF_H #define DIFF_H -#include "hash-ll.h" +#include "hash.h" #include "pathspec.h" #include "strbuf.h"
diffcore.h +1 -1
@@ -4,7 +4,7 @@#ifndef DIFFCORE_H #define DIFFCORE_H -#include "hash-ll.h" +#include "hash.h" struct diff_options ;struct mem_pool ;
dir.h +1 -1
@@ -1,7 +1,7 @@#ifndef DIR_H #define DIR_H -#include "hash-ll.h" +#include "hash.h" #include "hashmap.h" #include "pathspec.h" #include "statinfo.h"
hash-ll.h deleted +0 -364
364 lines not on this page · raw message
hash.h +360 -2
371 lines not on this page · raw message
hex.h +1 -1
@@ -1,7 +1,7 @@#ifndef HEX_H #define HEX_H -#include "hash-ll.h" +#include "hash.h" #include "hex-ll.h" /*
loose.h +2 -0
@@ -3,6 +3,8 @@#include "khash.h" +struct repository ; +struct loose_object_map {kh_oid_map_t *to_compat;kh_oid_map_t *to_storage;
merge-ort.h +1 -1
@@ -2,7 +2,7 @@#define MERGE_ORT_H #include "merge-recursive.h" -#include "hash-ll.h" +#include "hash.h" struct commit ;struct tree ;
object-file-convert.c +1 -1
@@ -5,7 +5,7 @@#include "strbuf.h" #include "hex.h" #include "repository.h" -#include "hash-ll.h" +#include "hash.h" #include "hash.h" #include "object.h" #include "loose.h"
object.h +1 -1
@@ -1,7 +1,7 @@#ifndef OBJECT_H #define OBJECT_H -#include "hash-ll.h" +#include "hash.h" struct buffer_slab ;struct repository ;
oidmap.h +1 -1
@@ -1,7 +1,7 @@#ifndef OIDMAP_H #define OIDMAP_H -#include "hash-ll.h" +#include "hash.h" #include "hashmap.h" /*
oidtree.h +1 -1
@@ -2,7 +2,7 @@#define OIDTREE_H #include "cbtree.h" -#include "hash-ll.h" +#include "hash.h" #include "mem-pool.h" struct oidtree {
packfile.h +2 -0
@@ -101,6 +101,8 @@ int close_pack_fd(struct packed_git *p);uint32_t get_pack_fanout (struct packed_git *p,uint32_t value); +struct raw_object_store ; +unsigned char *use_pack (struct packed_git *,struct pack_window **,off_t ,unsigned long *);void close_pack_windows (struct packed_git *);void close_pack (struct packed_git *);
protocol-caps.c +1 -1
@@ -3,7 +3,7 @@#include "gettext.h" #include "hex.h" #include "pkt-line.h" -#include "hash-ll.h" +#include "hash.h" #include "hex.h" #include "object.h" #include "object-store-ll.h"
read-cache-ll.h +1 -1
@@ -1,7 +1,7 @@#ifndef READ_CACHE_LL_H #define READ_CACHE_LL_H -#include "hash-ll.h" +#include "hash.h" #include "hashmap.h" #include "statinfo.h"
refs/ref-cache.h +1 -1
@@ -1,7 +1,7 @@#ifndef REFS_REF_CACHE_H #define REFS_REF_CACHE_H -#include "hash-ll.h" +#include "hash.h" struct ref_dir ;struct ref_store ;
reftable/dump.c +1 -1
@@ -7,7 +7,7 @@ license that can be found in the LICENSE file or at*/ #include "git-compat-util.h" -#include "hash-ll.h" +#include "hash.h" #include "reftable-blocksource.h" #include "reftable-error.h"
reftable/reftable-record.h +1 -1
@@ -9,7 +9,7 @@ license that can be found in the LICENSE file or at#ifndef REFTABLE_RECORD_H #define REFTABLE_RECORD_H -#include "hash-ll.h" +#include "hash.h" #include <stdint.h> /*
reftable/system.h +1 -1
@@ -15,7 +15,7 @@ license that can be found in the LICENSE file or at#include "lockfile.h" #include "strbuf.h" #include "tempfile.h" -#include "hash-ll.h" /* hash ID, sizes.*/ +#include "hash.h" /* hash ID, sizes.*/ #include "dir.h" /* remove_dir_recursively, for tests.*/ int hash_size (uint32_t id);
remote.h +1 -1
@@ -1,7 +1,7 @@#ifndef REMOTE_H #define REMOTE_H -#include "hash-ll.h" +#include "hash.h" #include "hashmap.h" #include "refspec.h"
reset.h +1 -1
@@ -1,7 +1,7 @@#ifndef RESET_H #define RESET_H -#include "hash-ll.h" +#include "hash.h" #include "repository.h" #define GIT_REFLOG_ACTION_ENVIRONMENT "GIT_REFLOG_ACTION"
resolve-undo.h +1 -1
@@ -6,7 +6,7 @@ struct index_state;struct pathspec ;struct string_list ; -#include "hash-ll.h" +#include "hash.h" struct resolve_undo_info {unsigned int mode[3 ];
serve.c +1 -1
@@ -3,7 +3,7 @@#include "git-compat-util.h" #include "repository.h" #include "config.h" -#include "hash-ll.h" +#include "hash.h" #include "pkt-line.h" #include "version.h" #include "ls-refs.h"
split-index.h +1 -1
@@ -1,7 +1,7 @@#ifndef SPLIT_INDEX_H #define SPLIT_INDEX_H -#include "hash-ll.h" +#include "hash.h" struct index_state ;struct strbuf ;
t/helper/test-hash-speed.c +1 -1
@@ -1,5 +1,5 @@#include "test-tool.h" -#include "hash-ll.h" +#include "hash.h" #define NUM_SECONDS 3
t/helper/test-sha1.c +1 -1
@@ -1,5 +1,5 @@#include "test-tool.h" -#include "hash-ll.h" +#include "hash.h" int cmd__sha1 (int ac,const char **av) {
t/helper/test-sha256.c +1 -1
@@ -1,5 +1,5 @@#include "test-tool.h" -#include "hash-ll.h" +#include "hash.h" int cmd__sha256 (int ac,const char **av) {
t/unit-tests/lib-oid.h +1 -1
@@ -1,7 +1,7 @@#ifndef LIB_OID_H #define LIB_OID_H -#include "hash-ll.h" +#include "hash.h" /* * Convert arbitrary hex string to object_id.
tree-diff.c +1 -0
@@ -8,6 +8,7 @@#include "tree.h" #include "tree-walk.h" #include "environment.h" +#include "repository.h" /* * Some mode bits are also used internally for computations.
tree-walk.h +1 -1
@@ -1,7 +1,7 @@#ifndef TREE_WALK_H #define TREE_WALK_H -#include "hash-ll.h" +#include "hash.h" struct index_state ;struct repository ;
xdiff-interface.h +1 -1
@@ -1,7 +1,7 @@#ifndef XDIFF_INTERFACE_H #define XDIFF_INTERFACE_H -#include "hash-ll.h" +#include "hash.h" #include "xdiff/xdiff.h" /*
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 12/20 http-fetch: don't crash when parsing packfile without a repo
In reply to Patrick SteinhardtThe git-http-fetch(1) command accepts a --packfile= option, which allows the user to specify that it shall fetch a specific packfile, only. The parameter here is the hash of the packfile, which is specific to the object hash used by the repository. This requirement is implicit though via our use of parse_oid_hex(), which internally uses the_repository.
The git-http-fetch(1) command allows for there to be no repository though, which only exists such that we can show usage via the "-h" option. In that case though, starting with c8aed5e8da (repository: stop setting SHA1 as the default object hash, 2024-05-07), the_repository does not have its object hash initialized anymore and thus we would crash when trying to parse the object ID outside of a repository.
Fix this issue by dying immediately when we see a "--packfile=" parameter when outside a Git repository. This is not a functional regression as we would die later on with the same error anyway.
Add a test to detect the segfault. We use the "nongit" function to do so, which we need to allow-list in test_must_fail ().
--- http-fetch.c | 8 +++++++- t/t5550-http-fetch-dumb.sh | 6 ++++++ t/test-lib-functions.sh | 5 +++++ 3 files changed, 18 insertions(+), 1 deletion(-)
http-fetch.c +7 -1
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "git-compat-util.h" #include "config.h" #include "gettext.h"
@@ -127,8 +129,12 @@ int cmd_main(int argc, const char **argv) }else if (skip_prefix (argv[arg],"--packfile=" , &p)) {const char *end; +if (nongit) +die (_ ("not a git repository" )); + packfile =1 ; -if (parse_oid_hex (p, &packfile_hash, &end) || *end) +if (parse_oid_hex_algop (p, &packfile_hash, &end, + the_repository->hash_algo) || *end)die (_ ("argument to --packfile must be a valid hash (got '%s')" ), p); }else if (skip_prefix (argv[arg],"--index-pack-arg=" , &p)) {strvec_push (&index_pack_args, p);
t/t5550-http-fetch-dumb.sh +6 -0
@@ -25,6 +25,12 @@ test_expect_success 'setup repository' ' git commit -m two ' +test_expect_success 'packfile without repository does not crash' ' + echo "fatal: not a git repository" >expect && + test_must_fail nongit git http-fetch --packfile=abc 2>err && + test_cmp expect err +' + setup_post_update_server_info_hook () { test_hook --setup -C "$1" post-update <<-\EOF && exec git update-server-info
t/test-lib-functions.sh +5 -0
@@ -1096,6 +1096,11 @@ test_must_fail_acceptable () { done fi + if test "$1" = "nongit" + then + shift + fi + case "$1" in git|__git*|scalar|test-tool|test_terminal) return 0
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 13/20 oidset: pass hash algorithm when parsing file
In reply to Patrick SteinhardtThe oidset_parse_file_carefully() function implicitly depends on the_repository when parsing object IDs. Fix this by having callers pass in the hash algorithm to use.
--- builtin/blame.c | 1 + fsck.c | 3 ++- oidset.c | 8 +++++--- oidset.h | 4 +++- 4 files changed, 11 insertions(+), 5 deletions(-)
builtin/blame.c +1 -0
@@ -852,6 +852,7 @@ static void build_ignorelist(struct blame_scoreboard *sb,oidset_clear (&sb->ignore_list); else oidset_parse_file_carefully(&sb->ignore_list, i->string, + the_repository->hash_algo, peel_to_commit_oid, sb); }for_each_string_list_item (i, ignore_rev_list) {
fsck.c +2 -1
@@ -205,7 +205,8 @@ void fsck_set_msg_types(struct fsck_options *options, const char *values)if (!strcmp (buf,"skiplist" )) {if (equal == len)die ("skiplist requires a path" ); -oidset_parse_file (&options->skiplist, buf + equal +1 ); +oidset_parse_file (&options->skiplist, buf + equal +1 , + the_repository->hash_algo); buf += len +1 ;continue ; }
oidset.c +5 -3
@@ -48,12 +48,14 @@ void oidset_clear(struct oidset *set)oidset_init (set,0 ); } -void oidset_parse_file (struct oidset *set,const char *path) +void oidset_parse_file (struct oidset *set,const char *path, +const struct git_hash_algo *algop) { -oidset_parse_file_carefully (set, path,NULL ,NULL ); +oidset_parse_file_carefully (set, path, algop,NULL ,NULL ); }void oidset_parse_file_carefully (struct oidset *set,const char *path, +const struct git_hash_algo *algop,oidset_parse_tweak_fn fn,void *cbdata) {FILE *fp;
@@ -79,7 +81,7 @@ void oidset_parse_file_carefully(struct oidset *set, const char *path,if (!sb.len)continue ; -if (parse_oid_hex (sb.buf, &oid, &p) || *p !='\0' ) +if (parse_oid_hex_algop (sb.buf, &oid, &p, algop) || *p !='\0' )die ("invalid object name: %s" , sb.buf);if (fn &&fn (&oid, cbdata))continue ;
oidset.h +3 -1
@@ -80,7 +80,8 @@ void oidset_clear(struct oidset *set);* are allowed. Leading whitespace and empty or white-space only lines are * ignored. */ -void oidset_parse_file (struct oidset *set,const char *path); +void oidset_parse_file (struct oidset *set,const char *path, +const struct git_hash_algo *algop);/* * Similar to the above, but with a callback which can (1) return non-zero to
@@ -89,6 +90,7 @@ void oidset_parse_file(struct oidset *set, const char *path);*/ typedef int (*oidset_parse_tweak_fn )(struct object_id *,void *);void oidset_parse_file_carefully (struct oidset *set,const char *path, +const struct git_hash_algo *algop,oidset_parse_tweak_fn fn,void *cbdata);struct oidset_iter {
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 14/20 protocol-caps: use hash algorithm from passed-in repository
In reply to Patrick SteinhardtIn send_info(), we pass in a repository but then use get_oid_hex() to parse passed-in object IDs, which implicitly uses the_repository. Fix this by using the hash algorithm from the passed-in repository instead.
--- protocol-caps.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
protocol-caps.c +2 -1
@@ -7,6 +7,7 @@#include "hex.h" #include "object.h" #include "object-store-ll.h" +#include "repository.h" #include "string-list.h" #include "strbuf.h"
@@ -52,7 +53,7 @@ static void send_info(struct repository *r, struct packet_writer *writer,struct object_id oid;unsigned long object_size; -if (get_oid_hex (oid_str, &oid) <0 ) { +if (get_oid_hex_algop (oid_str, &oid, r->hash_algo) <0 ) { packet_writer_error( writer, "object-info: protocol error, expected to get oid, not '%s'",
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 15/20 replace-object: use hash algorithm from passed-in repository
In reply to Patrick SteinhardtIn register_replace_ref(), we pass in a repository but then use get_oid_hex() to parse passed-in object IDs, which implicitly uses the_repository. Fix this by using the hash algorithm from the passed-in repository instead.
--- replace-object.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
replace-object.c +1 -1
@@ -20,7 +20,7 @@ static int register_replace_ref(const char *refname,const char *hash = slash ? slash +1 : refname;struct replace_object *repl_obj =xmalloc (sizeof (*repl_obj)); -if (get_oid_hex (hash, &repl_obj->original.oid)) { +if (get_oid_hex_algop (hash, &repl_obj->original.oid, r->hash_algo)) {free (repl_obj);warning (_ ("bad replace ref name: %s" ), refname);return 0 ;
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 16/20 compat/fsmonitor: fix socket path in networked SHA256 repos
In reply to Patrick SteinhardtThe IPC socket used by the fsmonitor on Darwin is usually contained in the Git repository itself. When the repository is hosted on a networked filesystem though, we instead create the socket path in the user's home directory or the socket directory. In that case, we derive the path by hashing the repository path.
But while we always use SHA1 to hash the repository path, we then end up using hash_to_hex() to append the computed hash to the socket path. This is wrong because hash_to_hex() uses the hash algorithm configured in the_repository, which may not be SHA1. The consequence is that we may append uninitialized bytes to the path when operating in a SHA256 repository.
Fix this bug by using hash_to_hex_algop() with SHA1.
--- compat/fsmonitor/fsm-ipc-darwin.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
compat/fsmonitor/fsm-ipc-darwin.c +4 -3
@@ -17,7 +17,7 @@ const char *fsmonitor_ipc__get_path(struct repository *r)git_SHA_CTX sha1ctx;char *sock_dir =NULL ;struct strbuf ipc_file =STRBUF_INIT ; -unsigned char hash[GIT_MAX_RAWSZ ]; +unsigned char hash[GIT_SHA1_RAWSZ ];if (!r)BUG ("No repository passed into fsmonitor_ipc__get_path" );
@@ -41,9 +41,10 @@ const char *fsmonitor_ipc__get_path(struct repository *r)/* Create the socket file in either socketDir or $HOME */ if (sock_dir && *sock_dir) {strbuf_addf (&ipc_file,"%s/.git-fsmonitor-%s" , - sock_dir,hash_to_hex (hash)); + sock_dir,hash_to_hex_algop (hash, &hash_algos[GIT_HASH_SHA1 ])); }else { -strbuf_addf (&ipc_file,"~/.git-fsmonitor-%s" ,hash_to_hex (hash)); +strbuf_addf (&ipc_file,"~/.git-fsmonitor-%s" , +hash_to_hex_algop (hash, &hash_algos[GIT_HASH_SHA1 ])); }free (sock_dir);
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 17/20 t/helper: use correct object hash in partial-clone helper
In reply to Patrick SteinhardtThe object_info() function of the partial-clone helper is responsible for checking the object ID of a repository other than the_repository. We use parse_oid_hex() in this function though, which means that we still depend on the_repository->hash_algo.
Fix this by using the object hash of the function-local repository.
--- t/helper/test-partial-clone.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
t/helper/test-partial-clone.c +1 -1
@@ -21,7 +21,7 @@ static void object_info(const char *gitdir, const char *oid_hex)if (repo_init (&r, gitdir,NULL ))die ("could not init repo" ); -if (parse_oid_hex (oid_hex, &oid, &p)) +if (parse_oid_hex_algop (oid_hex, &oid, &p, r.hash_algo))die ("could not parse oid" );if (oid_object_info_extended (&r, &oid, &oi,0 ))die ("could not obtain object info" );
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 18/20 t/helper: fix segfault in "oid-array" command without repository
In reply to Patrick SteinhardtThe "oid-array" test helper can supposedly work without a Git repository, but will in fact crash because the_repository->hash_algo is not initialized. This is because oid_pos(), which is used by oid_array_lookup(), depends on the_hash_algo->rawsz.
Ideally, we'd adapt oid_pos() to not depend on the_hash_algo anymore. That is a bigger untertaking though, so instead we fall back to SHA1 when there is no repository.
--- t/helper/test-oid-array.c | 4 ++++ t/t0064-oid-array.sh | 18 ++++++++++++++++++ 2 files changed, 22 insertions(+)
t/helper/test-oid-array.c +4 -0
@@ -1,3 +1,5 @@ +#define USE_THE_REPOSITORY_VARIABLE +#include "test-tool.h" #include "hex.h" #include "oid-array.h"
@@ -17,6 +19,8 @@ int cmd__oid_array(int argc UNUSED, const char **argv UNUSED)int nongit_ok;setup_git_directory_gently (&nongit_ok); +if (nongit_ok) +repo_set_hash_algo (the_repository,GIT_HASH_SHA1 );while (strbuf_getline (&line, stdin) !=EOF ) {const char *arg;
t/t0064-oid-array.sh +18 -0
@@ -15,6 +15,24 @@ echoid () { done } +test_expect_success 'without repository' ' + cat >expect <<-EOF && + 4444444444444444444444444444444444444444 + 5555555555555555555555555555555555555555 + 8888888888888888888888888888888888888888 + aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa + EOF + cat >input <<-EOF && + append 4444444444444444444444444444444444444444 + append 5555555555555555555555555555555555555555 + append 8888888888888888888888888888888888888888 + append aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa + for_each_unique + EOF + nongit test-tool oid-array <input >actual && + test_cmp expect actual +' + test_expect_success 'ordered enumeration' ' echoid "" 44 55 88 aa >expect && {
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 19/20 t/helper: remove dependency on `the_repository` in "proc-receive"
In reply to Patrick SteinhardtThe "proc-receive" test helper implicitly relies on the_repository via parse_oid_hex(). This isn't necessary though, and in fact the whole command does not depend on the_repository at all.
Stop setting up the_repository and use parse_oid_hex_any() to parse object IDs.
--- t/helper/test-proc-receive.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-)
t/helper/test-proc-receive.c +3 -6
@@ -3,8 +3,8 @@#include "hex.h" #include "parse-options.h" #include "pkt-line.h" -#include "setup.h" #include "sigchain.h" +#include "string-list.h" static const char *proc_receive_usage[] = {"test-tool proc-receive [<options>]" ,
@@ -92,9 +92,9 @@ static void proc_receive_read_commands(struct packet_reader *reader,if (die_read_commands)die ("die with the --die-read-commands option" ); -if (parse_oid_hex (reader->line, &old_oid, &p) || +if (parse_oid_hex_any (reader->line, &old_oid, &p) ==GIT_HASH_UNKNOWN || *p++ !=' ' || -parse_oid_hex (p, &new_oid, &p) || +parse_oid_hex_any (p, &new_oid, &p) ==GIT_HASH_UNKNOWN || *p++ !=' ' )die ("protocol error: expected 'old new ref', got '%s'" , reader->line);
@@ -128,7 +128,6 @@ static void proc_receive_read_push_options(struct packet_reader *reader,int cmd__proc_receive (int argc,const char **argv) { -int nongit_ok =0 ;struct packet_reader reader;struct command *commands =NULL ;struct string_list push_options =STRING_LIST_INIT_DUP ;
@@ -154,8 +153,6 @@ int cmd__proc_receive(int argc, const char **argv) OPT_END() }; -setup_git_directory_gently (&nongit_ok); - argc =parse_options (argc, argv,"test-tools" , options, proc_receive_usage,0 );if (argc >0 )usage_msg_opt ("Too many arguments." , proc_receive_usage, options);
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
PATCH v3 20/20 hex: guard declarations with `USE_THE_REPOSITORY_VARIABLE`
In reply to Patrick SteinhardtGuard declarations of functions that implicitly use the_repository with USE_THE_REPOSITORY_VARIABLE such that callers don't accidentally rely on that global variable.
--- hex.h | 26 ++++++++++++++++---------- 1 file changed, 16 insertions(+), 10 deletions(-)
hex.h +16 -10
@@ -13,10 +13,6 @@* input, so it is safe to pass this function an arbitrary * null-terminated string. */ -int get_hash_hex (const char *hex,unsigned char *hash); -int get_oid_hex (const char *hex,struct object_id *oid); - -/* Like get_oid_hex, but for an arbitrary hash algorithm. */ int get_oid_hex_algop (const char *hex,struct object_id *oid,const struct git_hash_algo *algop);/*
@@ -35,7 +31,6 @@ int get_oid_hex_algop(const char *hex, struct object_id *oid, const struct git_hchar *hash_to_hex_algop_r (char *buffer,const unsigned char *hash,const struct git_hash_algo *);char *oid_to_hex_r (char *out,const struct object_id *oid);char *hash_to_hex_algop (const unsigned char *hash,const struct git_hash_algo *);/* static buffer result! */ -char *hash_to_hex (const unsigned char *hash);/* same static buffer */ char *oid_to_hex (const struct object_id *oid);/* same static buffer */ /*
@@ -45,13 +40,9 @@ char *oid_to_hex(const struct object_id *oid); /* same static buffer */* other invalid character. end is only updated on success; otherwise, it is * unmodified. */ -int parse_oid_hex (const char *hex,struct object_id *oid,const char **end); - -/* Like parse_oid_hex, but for an arbitrary hash algorithm. */ int parse_oid_hex_algop (const char *hex,struct object_id *oid,const char **end,const struct git_hash_algo *algo); -/* * These functions work like get_oid_hex and parse_oid_hex, but they will parse * a hex value for any algorithm. The algorithm is detected based on the length
@@ -61,4 +52,19 @@ int parse_oid_hex_algop(const char *hex, struct object_id *oid, const char **endint get_oid_hex_any (const char *hex,struct object_id *oid);int parse_oid_hex_any (const char *hex,struct object_id *oid,const char **end); -#endif +#ifdef USE_THE_REPOSITORY_VARIABLE + +/* Like get_oid_hex_algop, but for `the_hash_algo`. */ +int get_hash_hex (const char *hex,unsigned char *hash); +int get_oid_hex (const char *hex,struct object_id *oid); + +/* Like parse_oid_hex_algop, but uses `the_hash_algo`. */ +int parse_oid_hex (const char *hex,struct object_id *oid,const char **end); + +/* +* Same as `hash_to_hex_algop()`, but uses `the_hash_algo`. +*/ +char *hash_to_hex (const unsigned char *hash); + +#endif /* USE_THE_REPOSITORY_VARIABLE */ +#endif /* HEX_H */
Signature · 1 line
2.45.2.457.g8d94cfb545.dirty
Hi Patrick
Quoted 11 lines · Patrick Steinhardt
Changes compared to v2:
- Note in a commit message that we aim to have a faithful conversion when introducing a
struct git_hash_algoparameter to functions. So even in case the calling context has astruct git_hash_algoavailable via a local repository, we still usethe_repositorysuch that there cannot be a change in behaviour here. Fixing those sites will be left for a future patch series such that we can avoid any kind of regressions caused by this comparatively large refactoring. I also adapted some conversions to fully follow through with this intent.
Thanks for clarifying that in the commit message of Patch 2.
Best Wishes
Phillip
Quoted 21 lines · Patrick Steinhardt
Hi,
this is the third version of my patch series that introduces the
USE_THE_REPOSITORY_VARIABLEmacro. When unset, this will cause us to hidethe_repository,the_hash_algoand several other functions that implicitly rely on those global variables from our headers. This is a first step towards fully getting rid of this global state in favor of passing it down explicitly via function parameters.Changes compared to v2:
- Note in a commit message that we aim to have a faithful conversion when introducing a
struct git_hash_algoparameter to functions. So even in case the calling context has astruct git_hash_algoavailable via a local repository, we still usethe_repositorysuch that there cannot be a change in behaviour here. Fixing those sites will be left for a future patch series such that we can avoid any kind of regressions caused by this comparatively large refactoring. I also adapted some conversions to fully follow through with this intent.
Quoted 5 lines
- Fix an issue with sparse by adding another
externdeclaration ofthe_repositoryto "repository.c".Thanks!
I forgot to reply here earlier.
I went through the patches and only found 2 small typos, overall this was an elaborate set of patches. I don't expect a reroll for that however. Thanks!