What's cooking (2021/09 #01)

Junio C Hamano committed Sep 2, 2021 at 14:30 UTC 13ca902789b227b6a3cc62eee35ea16df892a365
1 file changed +151 -206
whats-cooking.txt
+151 -206
@@ -1,10 +1,10 @@
1 To: git@vger.kernel.org
2 Bcc: lwn@lwn.net
3 -Subject: What's cooking in git.git (Aug 2021, #10; Tue, 31)
3 +Subject: What's cooking in git.git (Sep 2021, #01; Thu, 2)
4 X-master-at: 6c40894d2466d4e7fddc047a05116aa9d14712ee
5 -X-next-at: f2128466b98f6f7214f448ad9ae8252811cadb24
5 +X-next-at: 3052b894382a0debd2b7c8345f86f70277b76a3e
6
7 -What's cooking in git.git (Aug 2021, #10; Tue, 31)
7 +What's cooking in git.git (Sep 2021, #01; Thu, 2)
8 --------------------------------------------------
9
10 Here are the topics that have been cooking in my tree. Commits
@@ -18,11 +18,6 @@ useful"). Do not read too much into a topic being in (or not in)
18 'seen'. The ones marked with '.' do not appear in any of the
19 integration branches, but I am still holding onto them.
20
21 -Now 'merge -sort' is the default two-head merge strategy on the
22 -'master' front. Older topics that are not in 'seen' need help from
23 -readers to be sifted into bins for 'next' and the remainder to be
24 -kicked back to the authors.
25 -
21 Copies of the source code to Git live in many repositories, and the
22 following is a list of the ones I push into or their mirrors. Some
23 repositories have only a subset of branches.
@@ -52,148 +47,173 @@ Release tarballs are available at:
47 https://www.kernel.org/pub/software/scm/git/
48
49 --------------------------------------------------
55 -[Graduated to 'master']
50 +[New Topics]
51 +
52 +* ms/customizable-ident-expansion (2021-09-01) 1 commit
53 + - keyword expansion: make "$Id$" string configurable
54
57 -* bc/t5607-avoid-broken-test-fail-prereqs (2021-08-11) 1 commit
58 - (merged to 'next' on 2021-08-24 at 66c1e510ff)
59 - + t5607: avoid using prerequisites to select algorithm
55 + Instead of "$Id$", user-specified string (like $FreeBSD$) can be
56 + used as an in-blob placeholder for keyword expansion.
57
61 - The current implementation of GIT_TEST_FAIL_PREREQS is broken in
62 - that checking for the lack of a prerequisite would not work. Avoid
63 - the use of "if ! test_have_prereq X" in a test script.
58
59 +* pb/test-use-user-env (2021-09-01) 3 commits
60 + - test-lib-functions: optionally keep HOME and TERM in 'debug'
61 + - test-lib-functions: optionally keep HOME, TERM and SHELL in 'test_pause'
62 + - test-lib-functions: use 'TEST_SHELL_PATH' in 'test_pause'
63
66 -* cb/builtin-merge-format-string-fix (2021-08-09) 1 commit
67 - (merged to 'next' on 2021-08-24 at 6e94c19f1e)
68 - + builtin/merge: avoid -Wformat-extra-args from ancient Xcode
64 + Teach test_pause and test_debug to allow using the HOME and TERM
65 + environment variables the user usually uses.
66 +
67 + Will merge to 'next'?
68
70 - Code clean-up.
69
70 +* js/retire-preserve-merges (2021-09-01) 7 commits
71 + - tests: stop testing `git rebase --preserve-merges`
72 + - remote: warn about unhandled branch.<name>.rebase values
73 + - t5520: do not use `pull.rebase=preserve`
74 + - rebase: drop the internal `rebase--interactive` command
75 + - git-svn: drop support for `--preserve-merges`
76 + - rebase: drop support for `--preserve-merges`
77 + - pull: remove support for `--rebase=preserve`
78
73 -* cb/ci-freebsd-update (2021-08-12) 1 commit
74 - (merged to 'next' on 2021-08-24 at 1ecf63af7d)
75 - + ci: update freebsd 12 cirrus job
79 + The "--preserve-merges" option of "git rebase" has been removed.
80
77 - Update FreeBSD CI job
81 + Will merge to 'next'?
82
83
80 -* en/merge-strategy-docs (2021-08-05) 10 commits
81 - (merged to 'next' on 2021-08-24 at ceb66b1062)
82 - + Update error message and code comment
83 - + merge-strategies.txt: add coverage of the `ort` merge strategy
84 - + git-rebase.txt: correct out-of-date and misleading text about renames
85 - + merge-strategies.txt: fix simple capitalization error
86 - + merge-strategies.txt: avoid giving special preference to patience algorithm
87 - + merge-strategies.txt: do not imply using copy detection is desired
88 - + merge-strategies.txt: update wording for the resolve strategy
89 - + Documentation: edit awkward references to `git merge-recursive`
90 - + directory-rename-detection.txt: small updates due to merge-ort optimizations
91 - + git-rebase.txt: correct antiquated claims about --rebase-merges
92 - (this branch is used by en/ort-becomes-the-default.)
84 +* cb/remote-ndebug-fix (2021-09-02) 1 commit
85 + - remote: avoid -Wunused-but-set-variable in gcc with -DNDEBUG
86
94 - Documentation updates.
87 + Build fix.
88
89 + Will merge to 'next'.
90
97 -* en/ort-becomes-the-default (2021-08-05) 2 commits
98 - (merged to 'next' on 2021-08-24 at 92f4e08d83)
99 - + Update docs for change of default merge backend
100 - + Change default merge backend from recursive to ort
101 - (this branch uses en/merge-strategy-docs.)
91
103 - Use `ort` instead of `recursive` as the default merge strategy.
92 +* mr/bisect-in-c-4 (2021-09-02) 6 commits
93 + - bisect--helper: retire `--bisect-next-check` subcommand
94 + - bisect--helper: reimplement `bisect_run` shell
95 + - bisect--helper: reimplement `bisect_visualize()`shell function in C
96 + - run-command: make `exists_in_PATH()` non-static
97 + - t6030-bisect-porcelain: add test for bisect visualize
98 + - t6030-bisect-porcelain: add tests to control bisect run exit cases
99
100 + Rewrite of "git bisect" in C continues.
101
106 -* en/pull-conflicting-options (2021-07-22) 8 commits
107 - (merged to 'next' on 2021-08-24 at e1e5b2c551)
108 - + pull: fix handling of multiple heads
109 - + pull: update docs & code for option compatibility with rebasing
110 - + pull: abort by default when fast-forwarding is not possible
111 - + pull: make --rebase and --no-rebase override pull.ff=only
112 - + pull: since --ff-only overrides, handle it first
113 - + pull: abort if --ff-only is given and fast-forwarding is impossible
114 - + t7601: add tests of interactions with multiple merge heads and config
115 - + t7601: test interaction of merge/rebase/fast-forward flags and options
102 +--------------------------------------------------
103 +[Stalled]
104
117 - "git pull" had various corner cases that were not well thought out
118 - around its --rebase backend, e.g. "git pull --ff-only" did not stop
119 - but went ahead and rebased when the history on other side is not a
120 - descendant of our history. The series tries to fix them up.
105 +* gh/gitweb-branch-sort (2021-06-10) 1 commit
106 + - gitweb: use HEAD as secondary sort key in git_get_heads_list()
107 +
108 + Tie-break branches that point at the same object in the list of
109 + branches on GitWeb to show the one pointed at by HEAD early.
110 +
111 + Waiting for reviews.
112 +
113 +
114 +* es/superproject-aware-submodules (2021-08-19) 5 commits
115 + - fixup! introduce submodule.superprojectGitDir record
116 + - submodule: record superproject gitdir during 'update'
117 + - submodule: record superproject gitdir during absorbgitdirs
118 + - introduce submodule.superprojectGitDir record
119 + - t7400-submodule-basic: modernize inspect() helper
120
121 + A configuration variable in a submodule points at the location of
122 + the superproject it is bound to (RFC).
123
123 -* jc/userdiff-pattern-hint (2021-08-11) 1 commit
124 - (merged to 'next' on 2021-08-24 at f29f51148f)
125 - + userdiff: comment on the builtin patterns
124 + Looking good.
125
127 - Remind developers that the userdiff patterns should be kept simple
128 - and permissive, assuming that the contents they apply are always
129 - syntactically correct.
126
127 +* en/zdiff3 (2021-06-15) 2 commits
128 + - update documentation for new zdiff3 conflictStyle
129 + - xdiff: implement a zealous diff3, or "zdiff3"
130
132 -* jk/apply-binary-hunk-parsing-fix (2021-08-10) 1 commit
133 - (merged to 'next' on 2021-08-24 at c786cd1e8b)
134 - + apply: keep buffer/size pair in sync when parsing binary hunks
131 + "Zealous diff3" style of merge conflict presentation has been added.
132
136 - "git apply" miscounted the bytes and failed to read to the end of
137 - binary hunks.
133 + Expecting a reroll.
134 + cf. <CABPp-BE7-E03+x38EK-=AE5mwwdST+d50hiiud2eY2Nsf3rM5g@mail.gmail.com>
135
136
140 -* jk/range-diff-fixes (2021-08-10) 3 commits
141 - (merged to 'next' on 2021-08-24 at eff05f579e)
142 - + range-diff: use ssize_t for parsed "len" in read_patches()
143 - + range-diff: handle unterminated lines in read_patches()
144 - + range-diff: drop useless "offset" variable from read_patches()
137 +* ao/p4-avoid-decoding (2021-04-12) 2 commits
138 + - git-p4: do not decode data from perforce by default
139 + - git-p4: avoid decoding more data from perforce
140
146 - "git range-diff" code clean-up.
141 + "git p4" in Python-2 days used to accept a lot more kinds of data
142 + from Perforce server as uninterrupted byte sequence, but after
143 + switching to Python-3, too many things are expected to be in UTF-8,
144 + which broke traditional use cases.
145
146 + Waiting for reviews.
147
149 -* js/log-protocol-version (2021-08-10) 1 commit
150 - (merged to 'next' on 2021-08-24 at aeeaee04fb)
151 - + connect, protocol: log negotiated protocol version
148
153 - Debugging aid.
149 +* tv/p4-fallback-encoding (2021-04-30) 1 commit
150 + - git-p4: git-p4.fallbackEncoding to specify non UTF-8 charset
151
152 + "git p4" learns the fallbackEncoding configuration variable to
153 + safely accept changeset descriptions that aren't written in UTF-8.
154
156 -* th/userdiff-more-java (2021-08-11) 1 commit
157 - (merged to 'next' on 2021-08-24 at a5617abd3e)
158 - + userdiff: improve java hunk header regex
155 + Waiting for reviews.
156
160 - The userdiff pattern for "java" language has been updated.
157
158 +* ab/fsck-unexpected-type (2021-07-12) 21 commits
159 + - fsck: report invalid object type-path combinations
160 + - fsck: report invalid types recorded in objects
161 + - object-store.h: move read_loose_object() below 'struct object_info'
162 + - fsck: don't hard die on invalid object types
163 + - object-file.c: return -2 on "header too long" in unpack_loose_header()
164 + - object-file.c: return -1, not "status" from unpack_loose_header()
165 + - object-file.c: guard against future bugs in loose_object_info()
166 + - object-file.c: stop dying in parse_loose_header()
167 + - object-file.c: split up ternary in parse_loose_header()
168 + - object-file.c: simplify unpack_loose_short_header()
169 + - object-file.c: add missing braces to loose_object_info()
170 + - object-file.c: make parse_loose_header_extended() public
171 + - object-file.c: don't set "typep" when returning non-zero
172 + - cache.h: move object functions to object-store.h
173 + - cat-file tests: test for current --allow-unknown-type behavior
174 + - cat-file tests: add corrupt loose object test
175 + - rev-list tests: test for behavior with invalid object types
176 + - cat-file tests: test that --allow-unknown-type isn't on by default
177 + - cat-file tests: test for missing object with -t and -s
178 + - fsck tests: add test for fsck-ing an unknown type
179 + - fsck tests: refactor one test to use a sub-repo
180
163 -* tl/traverse-non-commits-rename (2021-08-12) 1 commit
164 - (merged to 'next' on 2021-08-24 at 55cc85c20a)
165 - + list-objects.c: rename "traverse_trees_and_blobs" to "traverse_non_commits"
181 + "git fsck" has been taught to report mismatch between expected and
182 + actual types of an object better.
183
167 - Meh.
184 + Needs review.
185
186 --------------------------------------------------
170 -[New Topics]
187 +[Cooking]
188
189 * tb/add-objects-in-unpacked-packs-simplify (2021-08-29) 3 commits
173 - - builtin/pack-objects.c: remove duplicate hash lookup
174 - - builtin/pack-objects.c: simplify add_objects_in_unpacked_packs()
175 - - object-store.h: teach for_each_packed_object to ignore kept packs
190 + (merged to 'next' on 2021-09-01 at 66b0f8564d)
191 + + builtin/pack-objects.c: remove duplicate hash lookup
192 + + builtin/pack-objects.c: simplify add_objects_in_unpacked_packs()
193 + + object-store.h: teach for_each_packed_object to ignore kept packs
194
195 Code simplification with reduced memory usage.
196
179 - Will merge to 'next'.
197 + Will merge to 'master'.
198
199
200 * es/walken-tutorial-fix (2021-08-30) 1 commit
183 - - doc: fix syntax error and the format of printf
201 + (merged to 'next' on 2021-09-01 at 73719bee8a)
202 + + doc: fix syntax error and the format of printf
203
204 Typofix.
205
187 - Will merge to 'next'.
206 + Will merge to 'master'.
207
208
209 * jh/sparse-index-resize-fix (2021-08-30) 1 commit
191 - - sparse-index: copy dir_hash in ensure_full_index()
210 + (merged to 'next' on 2021-09-01 at 5ee2615670)
211 + + sparse-index: copy dir_hash in ensure_full_index()
212
213 The sparse-index support can corrupt the index structure by storing
214 a stale and/or uninitialized data, which has been corrected.
215
196 - Will merge to 'next'.
216 + Will merge to 'master'.
217
218
219 * mh/send-email-reset-in-reply-to (2021-08-30) 1 commit
@@ -208,30 +228,31 @@ Release tarballs are available at:
228
229
230 * rs/more-fspathcmp (2021-08-30) 1 commit
211 - - merge-recursive: use fspathcmp() in path_hashmap_cmp()
231 + (merged to 'next' on 2021-09-01 at 3052b89438)
232 + + merge-recursive: use fspathcmp() in path_hashmap_cmp()
233
234 Code simplification.
235
215 - Will merge to 'next'.
236 + Will merge to 'master'.
237
238
239 * sg/set-ceiling-during-tests (2021-08-30) 1 commit
219 - - test-lib: set GIT_CEILING_DIRECTORIES to protect the surrounding repository
240 + (merged to 'next' on 2021-09-01 at 5378ce3817)
241 + + test-lib: set GIT_CEILING_DIRECTORIES to protect the surrounding repository
242
243 Buggy tests could damage repositories outside the throw-away test
244 area we created. We now by default export GIT_CEILING_DIRECTORIES
245 to limit the damage from such a stray test.
246
225 - Will merge to 'next'.
247 + Will merge to 'master'.
248
249
228 -* ab/gc-log-rephrase (2021-08-31) 1 commit
250 +* ab/gc-log-rephrase (2021-09-02) 1 commit
251 - gc: remove trailing dot from "gc.log" line
230 - (this branch uses uk/userdiff-php-enum.)
252
253 A pathname in an advice message has been made cut-and-paste ready.
254
234 - Will merge to 'next'?
255 + Will merge to 'next'.
256
257
258 * ab/mailmap-leakfix (2021-08-31) 1 commit
@@ -288,98 +309,11 @@ Release tarballs are available at:
309
310 * uk/userdiff-php-enum (2021-08-31) 1 commit
311 - userdiff: support enum keyword in PHP hunk header
291 - (this branch is used by ab/gc-log-rephrase.)
312
313 Update the userdiff pattern for PHP.
314
315 Will merge to 'next'.
316
297 ---------------------------------------------------
298 -[Stalled]
299 -
300 -* gh/gitweb-branch-sort (2021-06-10) 1 commit
301 - - gitweb: use HEAD as secondary sort key in git_get_heads_list()
302 -
303 - Tie-break branches that point at the same object in the list of
304 - branches on GitWeb to show the one pointed at by HEAD early.
305 -
306 - Waiting for reviews.
307 -
308 -
309 -* es/superproject-aware-submodules (2021-08-19) 5 commits
310 - - fixup! introduce submodule.superprojectGitDir record
311 - - submodule: record superproject gitdir during 'update'
312 - - submodule: record superproject gitdir during absorbgitdirs
313 - - introduce submodule.superprojectGitDir record
314 - - t7400-submodule-basic: modernize inspect() helper
315 -
316 - A configuration variable in a submodule points at the location of
317 - the superproject it is bound to (RFC).
318 -
319 - Looking good.
320 -
321 -
322 -* en/zdiff3 (2021-06-15) 2 commits
323 - - update documentation for new zdiff3 conflictStyle
324 - - xdiff: implement a zealous diff3, or "zdiff3"
325 -
326 - "Zealous diff3" style of merge conflict presentation has been added.
327 -
328 - Expecting a reroll.
329 - cf. <CABPp-BE7-E03+x38EK-=AE5mwwdST+d50hiiud2eY2Nsf3rM5g@mail.gmail.com>
330 -
331 -
332 -* ao/p4-avoid-decoding (2021-04-12) 2 commits
333 - - git-p4: do not decode data from perforce by default
334 - - git-p4: avoid decoding more data from perforce
335 -
336 - "git p4" in Python-2 days used to accept a lot more kinds of data
337 - from Perforce server as uninterrupted byte sequence, but after
338 - switching to Python-3, too many things are expected to be in UTF-8,
339 - which broke traditional use cases.
340 -
341 - Waiting for reviews.
342 -
343 -
344 -* tv/p4-fallback-encoding (2021-04-30) 1 commit
345 - - git-p4: git-p4.fallbackEncoding to specify non UTF-8 charset
346 -
347 - "git p4" learns the fallbackEncoding configuration variable to
348 - safely accept changeset descriptions that aren't written in UTF-8.
349 -
350 - Waiting for reviews.
351 -
352 -
353 -* ab/fsck-unexpected-type (2021-07-12) 21 commits
354 - - fsck: report invalid object type-path combinations
355 - - fsck: report invalid types recorded in objects
356 - - object-store.h: move read_loose_object() below 'struct object_info'
357 - - fsck: don't hard die on invalid object types
358 - - object-file.c: return -2 on "header too long" in unpack_loose_header()
359 - - object-file.c: return -1, not "status" from unpack_loose_header()
360 - - object-file.c: guard against future bugs in loose_object_info()
361 - - object-file.c: stop dying in parse_loose_header()
362 - - object-file.c: split up ternary in parse_loose_header()
363 - - object-file.c: simplify unpack_loose_short_header()
364 - - object-file.c: add missing braces to loose_object_info()
365 - - object-file.c: make parse_loose_header_extended() public
366 - - object-file.c: don't set "typep" when returning non-zero
367 - - cache.h: move object functions to object-store.h
368 - - cat-file tests: test for current --allow-unknown-type behavior
369 - - cat-file tests: add corrupt loose object test
370 - - rev-list tests: test for behavior with invalid object types
371 - - cat-file tests: test that --allow-unknown-type isn't on by default
372 - - cat-file tests: test for missing object with -t and -s
373 - - fsck tests: add test for fsck-ing an unknown type
374 - - fsck tests: refactor one test to use a sub-repo
375 -
376 - "git fsck" has been taught to report mismatch between expected and
377 - actual types of an object better.
378 -
379 - Needs review.
380 -
381 ---------------------------------------------------
382 -[Cooking]
317
318 * ab/commit-graph-usage (2021-08-30) 7 commits
319 - commit-graph: show "unexpected subcommand" error
@@ -495,7 +429,7 @@ Release tarballs are available at:
429 Will merge to 'master'.
430
431
498 -* jv/pkt-line-batch (2021-08-31) 2 commits
432 +* jv/pkt-line-batch (2021-09-01) 2 commits
433 - upload-pack: use stdio in send_ref callbacks
434 - pkt-line: add stdio packet write functions
435
@@ -534,12 +468,13 @@ Release tarballs are available at:
468
469
470 * ps/fetch-omit-formatting-under-quiet (2021-08-30) 1 commit
537 - - fetch: skip formatting updated refs with `--quiet`
471 + (merged to 'next' on 2021-09-01 at 2440a8a2aa)
472 + + fetch: skip formatting updated refs with `--quiet`
473
474 "git fetch --quiet" optimization to avoid useless computation of
475 info that will never be displayed.
476
542 - Will merge to 'next'.
477 + Will merge to 'master'.
478
479
480 * ps/ls-refs-strbuf-optim (2021-08-25) 1 commit
@@ -627,7 +562,7 @@ Release tarballs are available at:
562 possible backend.
563
564
630 -* ps/fetch-optim (2021-08-24) 8 commits
565 +* ps/fetch-optim (2021-09-01) 7 commits
566 - fetch: avoid second connectivity check if we already have all objects
567 - fetch: merge fetching and consuming refs
568 - fetch: refactor fetch refs to be more extendable
@@ -635,8 +570,9 @@ Release tarballs are available at:
570 - connected: refactor iterator to return next object ID directly
571 - fetch: avoid unpacking headers in object existence check
572 - fetch: speed up lookup of want refs via commit-graph
638 - - Merge branch 'ps/connectivity-optim' into ps/fetch-optim
639 - (this branch uses ps/connectivity-optim.)
573 +
574 + Optimize code that handles large number of refs in the "git fetch"
575 + code path.
576
577
578 * sg/make-fix-ar-invocation (2021-08-19) 1 commit
@@ -721,13 +657,14 @@ Release tarballs are available at:
657
658
659 * zh/cherry-pick-advice (2021-08-23) 1 commit
724 - - cherry-pick: use better advice message
660 + (merged to 'next' on 2021-09-01 at 6ff491cc09)
661 + + cherry-pick: use better advice message
662
663 The advice message that "git cherry-pick" gives when it asks
664 conflicted replay of a commit to be resolved by the end user has
665 been updated.
666
730 - Will merge to 'next'.
667 + Will merge to 'master'.
668
669
670 * es/config-based-hooks (2021-08-19) 7 commits
@@ -771,15 +708,18 @@ Release tarballs are available at:
708 Will merge to 'master'.
709
710
774 -* ka/want-ref-in-namespace (2021-08-13) 3 commits
775 - - docs: clarify the interaction of transfer.hideRefs and namespaces
776 - - upload-pack.c: treat want-ref relative to namespace
777 - - t5730: introduce fetch command helper
711 +* ka/want-ref-in-namespace (2021-09-01) 3 commits
712 + (merged to 'next' on 2021-09-01 at a1075a92a9)
713 + + docs: clarify the interaction of transfer.hideRefs and namespaces
714 + + upload-pack.c: treat want-ref relative to namespace
715 + + t5730: introduce fetch command helper
716
717 "git upload-pack" which runs on the other side of "git fetch"
718 forgot to take the ref namespaces into account when handling
719 want-ref requests.
720
721 + Will merge to 'master'.
722 +
723
724 * cb/ci-build-pedantic (2021-08-11) 1 commit
725 - ci: run a pedantic build as part of the GitHub workflow
@@ -808,7 +748,8 @@ Release tarballs are available at:
748
749
750 * js/advise-when-skipping-cherry-picked (2021-08-30) 1 commit
811 - - sequencer: advise if skipping cherry-picked commit
751 + (merged to 'next' on 2021-09-01 at bc883fce56)
752 + + sequencer: advise if skipping cherry-picked commit
753
754 "git rebase" by default skips changes that are equivalent to
755 commits that are already in the history the branch is rebased onto;
@@ -816,7 +757,7 @@ Release tarballs are available at:
757 skipped commits, and also teach them how to tell "rebase" to keep
758 duplicated changes.
759
819 - Will merge to 'next'.
760 + Will merge to 'master'.
761
762
763 * jt/grep-wo-submodule-odb-as-alternate (2021-08-13) 8 commits
@@ -841,7 +782,6 @@ Release tarballs are available at:
782 + revision: stop retrieving reference twice
783 + connected: do not sort input revisions
784 + revision: separate walk and unsorted flags
844 - (this branch is used by ps/fetch-optim.)
785
786 The revision traversal API has been optimized by taking advantage
787 of the commit-graph, when available, to determine if a commit is
@@ -905,6 +845,9 @@ Release tarballs are available at:
845
846 Further tweaks on progress API.
847
848 + On hold.
849 + cf. <20210901050406.GB76263@szeder.dev>
850 +
851
852 * ab/progress-users-adjust-counters (2021-08-25) 2 commits
853 - entry: show finer-grained counter in "Filtering content" progress line
@@ -974,7 +917,7 @@ Release tarballs are available at:
917 Will merge to 'next'?
918
919
977 -* tb/multi-pack-bitmaps (2021-08-24) 25 commits
920 +* tb/multi-pack-bitmaps (2021-09-01) 27 commits
921 - p5326: perf tests for MIDX bitmaps
922 - p5310: extract full and partial bitmap tests
923 - midx: respect 'GIT_TEST_MULTI_PACK_INDEX_WRITE_BITMAP'
@@ -996,6 +939,8 @@ Release tarballs are available at:
939 - midx: infer preferred pack when not given one
940 - midx: reject empty `--preferred-pack`'s
941 - midx: clear auxiliary .rev after replacing the MIDX
942 + - midx: fix `*.rev` cleanups with `--object-dir`
943 + - midx: disallow running outside of a repository
944 - Documentation: describe MIDX-based bitmaps
945 - pack-bitmap-write.c: free existing bitmaps
946 - pack-bitmap-write.c: gracefully fail to write non-closed bitmaps
@@ -1005,7 +950,7 @@ Release tarballs are available at:
950 pack, but now we've learned to generate bitmaps for history that
951 span across multiple packfiles.
952
1008 - Comments?
953 + Will merge to 'next'?
954
955
956 * ab/config-based-hooks-base (2021-08-03) 36 commits