c96e9bd64c · matheusbsbs/matheus-bagliano

@@ -1,10 +1,10 @@

1 To: git@vger.kernel.org

2 Bcc: lwn@lwn.net

3 -Subject: What's cooking in git.git (Nov 2021, #03; Tue, 9)

4 -X-master-at: 6c220937e2b26d85920bf2d38ff2464a0d57fd6b

5 -X-next-at: 773e6cccef14fb4ef03dc964d7f80ceb691c12be

3 +Subject: What's cooking in git.git (Nov 2021, #04; Sun, 14)

4 +X-master-at: cd3e606211bb1cf8bc57f7d76bab98cc17a150bc

5 +X-next-at: ca35af825273b98fc8dc11527488952f5db8eb80

6

7 -What's cooking in git.git (Nov 2021, #03; Tue, 9)

7 +What's cooking in git.git (Nov 2021, #04; Sun, 14)

8 --------------------------------------------------

9

10 Here are the topics that have been cooking in my tree. Commits

@@ -18,8 +18,19 @@ 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 -Git 2.34-rc2 has been tagged with a few more regression fixes.

21 +Git 2.34 (final) has been tagged with a few more regression fixes.

22 +

23 +As we have been fairly aggressive to deliberately merge down many

24 +topics that are under-reviewed, mostly due to reviewer fatigue and

25 +limited reviewer bandwidth, it was expected that we'd see unexpected

26 +breakages after things are in 'master', and it was quite nice to see

27 +people found and squashed these problems fairly quickly.

28 +

29 Thank you all for finding, reporting and fixing them so quickly.

30 +Before starting to queue new topics for the next cycle or merging

31 +down topics in-flight, we'd wait about a week to see if people

32 +encounter any more issues, and after that we'd restart the usual

33 +cycle.

34

35 Copies of the source code to Git live in many repositories, and the

36 following is a list of the ones I push into or their mirrors. Some

@@ -50,102 +61,140 @@ Release tarballs are available at:

61 https://www.kernel.org/pub/software/scm/git/

62

63 --------------------------------------------------

53 -[New Topics]

64 +[Graduated to 'master']

65

55 -* ak/fetch-not-overwrite-any-current-branch (2021-11-08) 2 commits

56 - - receive-pack: protect current branch for bare repository worktree

57 - - fetch: protect branches checked out in all worktrees

66 +* ab/fsck-unexpected-type (2021-11-11) 2 commits

67 + (merged to 'next' on 2021-11-11 at f8c9666880)

68 + + object-file: free(*contents) only in read_loose_object() caller

69 + + object-file: fix SEGV on free() regression in v2.34.0-rc2

70

59 - "git fetch" without the "--update-head-ok" option ought to protect

60 - a checked out branch from getting updated, to prevent the working

61 - tree that checks it out to go out of sync. The code was written

62 - before the use of "git worktree" got widespread, and only checked

63 - the branch that was checked out in the current worktree, which has

64 - been updated.

71 + Regression fix.

72

66 - Reroll exists.

73

74 +* ds/no-usable-cron-on-macos (2021-11-10) 1 commit

75 + (merged to 'next' on 2021-11-10 at 18eb9c13bc)

76 + + maintenance: disable cron on macOS

77

69 -* jk/test-bitmap-fix (2021-11-05) 1 commit

70 - - test_bitmap_hashes(): handle repository without bitmaps

78 + "git maintenance run" learned to use system supplied scheduler

79 + backend, but cron on macOS turns out to be unusable for this

80 + purpose.

81

72 - Tighten code for testing pack-bitmap.

82

74 - Will merge to 'next'.

83 +* jc/fix-pull-ff-only-when-already-up-to-date (2021-10-29) 1 commit

84 + (merged to 'next' on 2021-10-29 at ad4753e668)

85 + + pull: --ff-only should make it a noop when already-up-to-date

86

87 + "git pull --ff-only" and "git pull --rebase --ff-only" should make

88 + it a no-op to attempt pulling from a remote that is behind us, but

89 + instead the command errored out by saying it was impossible to

90 + fast-forward, which may technically be true, but not a useful thing

91 + to diagnose as an error. This has been corrected.

92

77 -* jk/jump-merge-with-pathspec (2021-11-09) 1 commit

78 - - git-jump: pass "merge" arguments to ls-files

93

80 - The "merge" subcommand of "git jump" (in contrib/) silently ignored

81 - pathspec and other parameters.

94 +* jk/ssh-signing-fix (2021-11-10) 1 commit

95 + (merged to 'next' on 2021-11-10 at b6195caa04)

96 + + t/lib-gpg: avoid broken versions of ssh-keygen

97

83 - Will merge to 'next'.

98 + Reject OpenSSH 8.7 whose "ssh-keygen -Y find-principals" is

99 + unusable from running the ssh signature tests.

100 +

101 +

102 +* js/simple-ipc-cygwin-socket-fix (2021-11-10) 1 commit

103 + (merged to 'next' on 2021-11-10 at 00788f076f)

104 + + simple-ipc: work around issues with Cygwin's Unix socket emulation

105 +

106 + The way Cygwin emulates a unix-domain socket, on top of which the

107 + simple-ipc mechanism is implemented, can race with the program on

108 + the other side that wants to use the socket, and briefly make it

109 + appear as a regular file before lstat(2) starts reporting it as a

110 + socket. We now have a workaround on the side that connects to a

111 + unix domain socket.

112

85 ---------------------------------------------------

86 -[Graduated to 'master']

113

88 -* ab/parse-options-cleanup (2021-11-09) 1 commit

89 - (merged to 'next' on 2021-11-09 at b2b859289b)

90 - + parse-options.[ch]: revert use of "enum" for parse_options()

114 +* js/trace2-raise-format-version (2021-11-11) 1 commit

115 + (merged to 'next' on 2021-11-11 at e7144ad963)

116 + + trace2: increment event format version

117

92 - Last minute fix to the update already in 'master'.

118 + When we added a new event type to trace2 event stream, we forgot to

119 + raise the format version number, which has been corrected.

120

121

95 -* ad/ssh-signing-testfix (2021-11-05) 1 commit

96 - (merged to 'next' on 2021-11-09 at f5ff9c35ad)

97 - + t/lib-git.sh: fix ACL-related permissions failure

122 +* ps/connectivity-optim (2021-11-11) 1 commit

123 + (merged to 'next' on 2021-11-11 at 8b3dccbd68)

124 + + Revert "connected: do not sort input revisions"

125 +

126 + Regression fix.

127 +

128 +--------------------------------------------------

129 +[New Topics]

130

99 - Fix ssh-signing test to work on a platform where the default ACL is

100 - overly loose to upset OpenSSH (reported on an installation of Cygwin).

131 +* rj/receive-pack-avoid-sigpipe-during-status-reporting (2021-11-10) 1 commit

132 + - receive-pack: ignore SIGPIPE while reporting status to client

133

134 + When the "git push" command is killed while the receiving end is

135 + trying to report what happened to the ref update proposals, the

136 + latter used to die, due to SIGPIPE. The code now ignores SIGPIPE

137 + to increase our chances to run the post-receive hook after it

138 + happens.

139

103 -* ar/fix-git-pull-no-verify (2021-10-28) 1 commit

104 - (merged to 'next' on 2021-11-01 at 74677b0a76)

105 - + pull: honor --no-verify and do not call the commit-msg hook

140

107 - "git pull --no-verify" did not affect the underlying "git merge".

141 +* ab/parse-options-cleanup (2021-11-10) 1 commit

142 + - parse-options.c: use "enum parse_opt_result" for parse_nodash_opt()

143

144 + Change the type of an internal function to return an enum (instead

145 + of int) and replace -2 that was used to signal an error with -1.

146

110 -* ar/no-verify-doc (2021-10-29) 1 commit

111 - (merged to 'next' on 2021-11-01 at f6809a9ddb)

112 - + Document positive variant of commit and merge option "--no-verify"

147 +

148 +* cw/protocol-v2-doc-fix (2021-11-11) 1 commit

149 + - protocol-v2.txt: align delim-pkt spec with usage

150

151 Doc update.

152

153 + Will merge to 'next'.

154

117 -* rd/http-backend-code-simplification (2021-10-25) 1 commit

118 - (merged to 'next' on 2021-11-01 at f45b33890c)

119 - + http-backend: remove a duplicated code branch

155

121 - Code simplification.

156 +* jt/pack-header-lshift-overflow (2021-11-11) 1 commit

157 + - packfile: avoid overflowing shift during decode

158

159 + The code to decode the length of packed object size has been

160 + corrected.

161

124 -* vd/pthread-setspecific-g11-fix (2021-11-03) 1 commit

125 - (merged to 'next' on 2021-11-03 at b4fce69914)

126 - + async_die_is_recursing: work around GCC v11.x issue on Fedora

162 + Will merge to 'next'.

163

128 - One CI task based on Fedora image noticed a not-quite-kosher

129 - consturct recently, which has been corrected.

164

131 ---------------------------------------------------

132 -[Stalled]

165 +* ab/update-submitting-patches (2021-11-13) 1 commit

166 + - SubmittingPatches: fix Asciidoc syntax in "GitHub CI" section

167

134 -* mp/absorb-submodule-git-dir-upon-deinit (2021-10-07) 1 commit

135 - - submodule: absorb git dir instead of dying on deinit

168 + Doc fix.

169

137 - "git submodule deinit" for a submodule whose .git metadata

138 - directory is embedded in its working tree refused to work, until

139 - the submodule gets converted to use the "absorbed" form where the

140 - metadata directory is stored in superproject, and a gitfile at the

141 - top-level of the working tree of the submodule points at it. The

142 - command is taught to convert such submodules to the absorbed form

143 - as needed.

170 + Will merge to 'next'.

171

145 - Expecting a reroll.

146 - cf. <xmqqwnmopqqk.fsf@gitster.g>

172 +

173 +* bc/require-c99 (2021-11-14) 1 commit

174 + - git-compat-util: add a test balloon for C99 support

175 +

176 + Weather balloon to break people with compilers that do not support

177 + C99.

178

179

180 +* jc/c99-var-decl-in-for-loop (2021-11-14) 1 commit

181 + - revision: use C99 declaration of variable in for() loop

182 +

183 + Weather balloon to break comiplers that do not grok variable

184 + declaration in the for() loop.

185 +

186 +

187 +

188 +* tl/midx-docfix (2021-11-14) 1 commit

189 + - midx: fix a formatting issue in "multi-pack-index.txt"

190 +

191 + Doc mark-up fix.

192 +

193 + Will merge to 'next'.

194 +

195 +--------------------------------------------------

196 +[Stalled]

197 +

198 * ar/submodule-update (2021-10-13) 9 commits

199 . submodule--helper: rename helper functions

200 . submodule--helper: remove unused helpers

@@ -163,6 +212,21 @@ Release tarballs are available at:

212 which is among the topics this topic stomps on.

213

214

215 +* mp/absorb-submodule-git-dir-upon-deinit (2021-10-07) 1 commit

216 + - submodule: absorb git dir instead of dying on deinit

217 +

218 + "git submodule deinit" for a submodule whose .git metadata

219 + directory is embedded in its working tree refused to work, until

220 + the submodule gets converted to use the "absorbed" form where the

221 + metadata directory is stored in superproject, and a gitfile at the

222 + top-level of the working tree of the submodule points at it. The

223 + command is taught to convert such submodules to the absorbed form

224 + as needed.

225 +

226 + Getting tired of waiting for a reroll; will discard.

227 + cf. <xmqqwnmopqqk.fsf@gitster.g>

228 +

229 +

230 * cf/fetch-set-upstream-while-detached (2021-07-06) 1 commit

231 - fetch: fix segfault on --set-upstream while on a detached HEAD

232

@@ -174,12 +238,45 @@ Release tarballs are available at:

238 --------------------------------------------------

239 [Cooking]

240

241 +* ak/fetch-not-overwrite-any-current-branch (2021-11-09) 4 commits

242 + - branch: protect branches checked out in all worktrees

243 + - receive-pack: protect current branch for bare repository worktree

244 + - receive-pack: clean dead code from update_worktree()

245 + - fetch: protect branches checked out in all worktrees

246 +

247 + "git fetch" without the "--update-head-ok" option ought to protect

248 + a checked out branch from getting updated, to prevent the working

249 + tree that checks it out to go out of sync. The code was written

250 + before the use of "git worktree" got widespread, and only checked

251 + the branch that was checked out in the current worktree, which has

252 + been updated.

253 +

254 +

255 +* jk/test-bitmap-fix (2021-11-05) 1 commit

256 + - test_bitmap_hashes(): handle repository without bitmaps

257 +

258 + Tighten code for testing pack-bitmap.

259 +

260 + Will merge to 'next'.

261 +

262 +

263 +* jk/jump-merge-with-pathspec (2021-11-09) 1 commit

264 + - git-jump: pass "merge" arguments to ls-files

265 +

266 + The "merge" subcommand of "git jump" (in contrib/) silently ignored

267 + pathspec and other parameters.

268 +

269 + Will merge to 'next'.

270 +

271 +

272 * jk/strbuf-addftime-seconds-since-epoch (2021-11-04) 1 commit

273 - strbuf_addftime(): handle "%s" manually

274

275 The "--date=format:<strftime>" gained a workaround for the lack of

276 system support for a non-local timezone to handle "%s" placeholder.

277

278 + Will merge to 'next'.

279 +

280

281 * js/ci-no-directional-formatting (2021-11-04) 1 commit

282 - ci: disallow directional formatting

@@ -187,6 +284,8 @@ Release tarballs are available at:

284 CI has been taught to catch some Unicode directional formatting

285 sequence that can be used in certain mischief.

286

287 + Will merge to 'next'.

288 +

289

290 * tw/var-default-branch (2021-11-03) 1 commit

291 - var: add GIT_DEFAULT_BRANCH variable

@@ -194,6 +293,8 @@ Release tarballs are available at:

293 "git var GIT_DEFAULT_BRANCH" is a way to see what name is used for

294 the newly created branch if "git init" is run.

295

296 + Will merge to 'next'.

297 +

298

299 * ab/mark-leak-free-tests-even-more (2021-11-01) 15 commits

300 - leak tests: mark some fast-import tests as passing with SANITIZE=leak

@@ -214,13 +315,17 @@ Release tarballs are available at:

315

316 More tests are marked as leak-free.

317

318 + Will merge to 'next'?

319 +

320

218 -* if/redact-packfile-uri (2021-10-29) 2 commits

321 +* if/redact-packfile-uri (2021-11-11) 2 commits

322 - http-fetch: redact url on die() message

323 - fetch-pack: redact packfile urls in traces

324

325 Redact the path part of packfile URI that appears in the trace output.

326

327 + Will merge to 'next'.

328 +

329

330 * ew/test-wo-fsync (2021-10-29) 1 commit

331 - tests: disable fsync everywhere

@@ -252,6 +357,8 @@ Release tarballs are available at:

357

358 "default" and "reset" colors have been added to our palette.

359

360 + Will merge to 'next'?

361 +

362

363 * jc/fix-first-object-walk (2021-10-29) 2 commits

364 - docs: add headers in MyFirstObjectWalk

@@ -259,6 +366,8 @@ Release tarballs are available at:

366

367 Doc update.

368

369 + Will merge to 'next'?

370 +

371

372 * jc/unsetenv-returns-an-int (2021-10-29) 1 commit

373 (merged to 'next' on 2021-11-01 at a11670a5a5)

@@ -324,19 +433,6 @@ Release tarballs are available at:

433 Build optimization.

434

435

327 -* jc/fix-pull-ff-only-when-already-up-to-date (2021-10-29) 1 commit

328 - (merged to 'next' on 2021-10-29 at ad4753e668)

329 - + pull: --ff-only should make it a noop when already-up-to-date

330 -

331 - "git pull --ff-only" and "git pull --rebase --ff-only" should make

332 - it a no-op to attempt pulling from a remote that is behind us, but

333 - instead the command errored out by saying it was impossible to

334 - fast-forward, which may technically be true, but not a useful thing

335 - to diagnose as an error. This has been corrected.

336 -

337 - Will cook in 'next'.

338 -

339 -

436 * ab/sh-retire-helper-functions (2021-10-21) 6 commits

437 (merged to 'next' on 2021-11-03 at 4b49d9f4be)

438 + git-sh-setup: remove "sane_grep", it's not needed anymore

@@ -388,6 +484,8 @@ Release tarballs are available at:

484 selected command-line options to the underlying "git describe"

485 command.

486

487 + Will merge to 'next'?

488 +

489

490 * fs/ssh-signing-key-lifetime (2021-10-27) 8 commits

491 - ssh signing: make fmt-merge-msg consider key lifetime

@@ -402,6 +500,8 @@ Release tarballs are available at:

500 Extend the signing of objects with SSH keys and learn to pay

501 attention to the key validity time range when verifying.

502

503 + Will merge to 'next'?

504 +

505

506 * jc/doc-submitting-patches-choice-of-base (2021-10-25) 2 commits

507 - (wip) reword the final review part

@@ -687,6 +787,8 @@ Release tarballs are available at:

787

788 "Zealous diff3" style of merge conflict presentation has been added.

789

790 + What's the status of this thing?

791 +

792

793 * js/scalar (2021-10-27) 15 commits

794 - scalar: accept -C and -c options before the subcommand

@@ -707,8 +809,6 @@ Release tarballs are available at:

809

810 Add pieces from "scalar" to contrib/.

811

710 - What's the status of this thing?

711 -

812

813 * ms/customizable-ident-expansion (2021-09-01) 1 commit

814 - keyword expansion: make "$Id$" string configurable

@@ -716,8 +816,6 @@ Release tarballs are available at:

816 Instead of "$Id$", user-specified string (like $FreeBSD$) can be

817 used as an in-blob placeholder for keyword expansion.

818

719 - What's the status of this one? Meh?

720 -

819

820 * ab/refs-errno-cleanup (2021-10-16) 21 commits

821 (merged to 'next' on 2021-10-29 at 3f57147176)

@@ -784,3 +882,5 @@ Release tarballs are available at:

882

883 The "reftable" backend for the refs API, without integrating into

884 the refs subsystem.

885 +

886 + Will merge to 'next'?

What's cooking (2021/11 #04)