Skip to contentSkip to conversation

What's cooking in git.git (Jan 2022, #03; Thu, 13)

DISCUSSION Junio C Hamano 7 participants 19 messages Jan 14, 2022 – Jan 18, 2022

lore
Share & tools
lore
#0 lore raw

There are a few "oops, what we merged recently is broken" topics that still are not in 'master', but otherwise what we have should be pretty much what we'll have in the final one.

- I am reasonably happy with ab/refs-errno-cleanup (just one patch) that fixes the incorrect state of the code left by the earlier parts of the topic that have already been merged during this cycle.

- I am also OK with ab/reftable-build-fixes (two patches), one for general type correctness fix, the other for helping older sub-C99 compilers.

If there are fixes for regressions that we introduced during this cycle other than these two topics, I certainly am missing them, so please holler loudly and quickly, hopefully in time for me to tag the -rc1 tomorrow.

Here are the topics that have been cooking in my tree. Commits prefixed with '+' are in 'next' (being in 'next' is a sign that a topic is stable enough to be used and are candidate to be in a future release). Commits prefixed with '-' are only in 'seen', which means nothing more than that I have found them of interest for some reason (like "it may have hard-to-resolve conflicts with another topic already in flight" or "this may turn out to be useful"). Do not read too much into a topic being in (or not in) 'seen'. The ones marked with '.' do not appear in any of the integration branches, but I am still holding onto them.

Copies of the source code to Git live in many repositories, and the following is a list of the ones I push into or their mirrors. Some repositories have only a subset of branches.

With maint, master, next, seen, todo:

	git://git.kernel.org/pub/scm/git/git.git/
	git://repo.or.cz/alt-git.git/
	https://kernel.googlesource.com/pub/scm/git/git/
	https://github.com/git/git/
	https://gitlab.com/git-vcs/git/

With all the integration branches and topics broken out:

https://github.com/gitster/git/

Even though the preformatted documentation in HTML and man format are not sources, they are published in these repositories for convenience (replace "htmldocs" with "manpages" for the manual pages):

git://git.kernel.org/pub/scm/git/git-htmldocs.git/
https://github.com/gitster/git-htmldocs.git/

Release tarballs are available at:

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

--------------------------------------------------
[Graduated to 'master']

* ab/do-not-limit-stash-help-to-push (2021-12-16) 1 commit (merged to 'next' on 2022-01-03 at adfc0c1f17)
+ stash: don't show "git stash push" usage on bad "git stash" usage

"git stash" by default triggers its "push" action, but its implementation also made "git stash -h" to show short help only for "git stash push", which has been corrected.
source: <[email protected]>

* ab/makefile-hook-list-dependency-fix (2021-12-25) 1 commit (merged to 'next' on 2022-01-03 at 8c8ca0cbce)
+ Makefile: correct the dependency graph of hook-list.h

Fix dependency rules to generate hook-list.h header file.
source: <[email protected]>

* ab/makefile-msgfmt-wo-stats (2021-12-25) 1 commit (merged to 'next' on 2022-01-03 at 2e2097c310)
+ Makefile: don't invoke msgfmt with --statistics

Make the recipe that runs msgfmt less noisy.
source: <[email protected]>

* ab/makefile-pager-env-is-used-only-by-pager.c (2021-12-25) 1 commit (merged to 'next' on 2022-01-03 at 7ab1e1c5c4)
+ Makefile: move -DPAGER_ENV from BASIC_CFLAGS to EXTRA_CPPFLAGS

source: <[email protected]>

* ab/reflog-prep (2021-12-22) 9 commits
(merged to 'next' on 2022-01-03 at 07d2d78d86)
+ reflog + refs-backend: move "verbose" out of the backend
+ refs files-backend: assume cb->newlog if !EXPIRE_REFLOGS_DRY_RUN
+ reflog: reduce scope of "struct rev_info"
+ reflog expire: don't use lookup_commit_reference_gently()
+ reflog expire: refactor & use "tip_commit" only for UE_NORMAL
+ reflog expire: use "switch" over enum values
+ reflog: change one->many worktree->refnames to use a string_list
+ reflog expire: narrow scope of "cb" in cmd_reflog_expire()
+ reflog delete: narrow scope of "cmd" passed to count_reflog_ent() (this branch is used by jc/reflog-parse-options.)

Code refactoring in the reflog part of refs API.
source: <[email protected]>

* ab/usage-die-message (2021-12-07) 6 commits
(merged to 'next' on 2022-01-03 at 0ef3860099)
+ config API: use get_error_routine(), not vreportf()
+ usage.c + gc: add and use a die_message_errno()
+ gc: return from cmd_gc(), don't call exit()
+ usage.c API users: use die_message() for error() + exit 128
+ usage.c API users: use die_message() for "fatal :" + exit 128
+ usage.c: add a die_message() routine

Code clean-up to hide vreportf() from public API.
source: <[email protected]>

* ak/protect-any-current-branch (2022-01-12) 1 commit (merged to 'next' on 2022-01-12 at 5ad87959d2)
+ branch: missing space fix at line 313

source: <[email protected]>

* ds/fetch-pull-with-sparse-index (2021-12-22) 5 commits (merged to 'next' on 2022-01-03 at f58da4a878)
+ test-read-cache: remove --table, --expand options
+ t1091/t3705: remove 'test-tool read-cache --table'
+ t1092: replace 'read-cache --table' with 'ls-files --sparse'
+ ls-files: add --sparse option
+ fetch/pull: use the sparse index

"git fetch" and "git pull" are now declared sparse-index clean. Also "git ls-files" learns the "--sparse" option to help debugging.
source: <[email protected]>

* ds/sparse-checkout-malformed-pattern-fix (2021-12-30) 3 commits (merged to 'next' on 2021-12-30 at c8b2ade48c)
+ sparse-checkout: refuse to add to bad patterns
+ sparse-checkout: fix OOM error with mixed patterns
+ sparse-checkout: fix segfault on malformed patterns

Certain sparse-checkout patterns that are valid in non-cone mode led to segfault in cone mode, which has been corrected.
source: <[email protected]>

* en/merge-ort-renorm-with-rename-delete-conflict-fix (2021-12-30) 1 commit (merged to 'next' on 2022-01-05 at bb81dd432d)
+ merge-ort: fix bug with renormalization and rename/delete conflicts

A corner case bug in the ort merge strategy has been corrected.
source: <[email protected]>

* en/stash-df-fix (2022-01-04) 1 commit
(merged to 'next' on 2022-01-06 at 22d0c6ccf7)
+ stash: do not return before restoring untracked files

"git stash apply" forgot to attempt restoring untracked files when it failed to restore changes to tracked ones.
source: <[email protected]>

* fs/gpg-unknown-key-test-fix (2022-01-12) 1 commit (merged to 'next' on 2022-01-12 at 80d3416168)
+ t/gpg: simplify test for unknown key

Test simplification.
source: <[email protected]>

* fs/ssh-signing-key-lifetime (2022-01-10) 1 commit (merged to 'next' on 2022-01-12 at 8541f3399c)
+ fmt-merge-msg: prevent use-after-free with signed tags

"git merge $signed_tag" started to drop the tag message from the default merge message it uses by accident, which has been corrected.
source: <6e08b73d602853b3de71257117e85e32b96b5c19.1641849502.git.me@ttaylorr.com>

* hn/ref-api-tests-update (2021-12-22) 4 commits (merged to 'next' on 2022-01-03 at 6815d4d8c1)
+ t7004: use "test-tool ref-store" for reflog inspection
+ t7004: create separate tags for different tests
+ t5550: require REFFILES
+ t5540: require REFFILES

Test updates.
source: <[email protected]>

* hn/refs-debug-update (2021-12-22) 3 commits (merged to 'next' on 2022-01-03 at 1cd451c66c)
+ refs: centralize initialization of the base ref_store.
+ refs: print error message in debug output
+ refs: pass gitdir to packed_ref_store_create

Debugging support for refs API.
source: <[email protected]>

* hn/reftable-fixes (2021-12-23) 3 commits
(merged to 'next' on 2022-01-03 at 532f9bbb27)
+ reftable: support preset file mode for writing
+ reftable: signal overflow
+ reftable: fix typo in header

Assorted fixlets in reftable code.
source: <[email protected]>

* hn/test-ref-store-show-hash-algo (2021-12-21) 1 commit (merged to 'next' on 2022-01-03 at 3cdbfeeaf1)
+ test-ref-store: print hash algorithm

Debugging support for refs API.
source: <[email protected]>

* ja/i18n-similar-messages (2022-01-05) 11 commits
(merged to 'next' on 2022-01-06 at 6440d8fdf2)
+ i18n: turn even more messages into "cannot be used together" ones
+ i18n: ref-filter: factorize "%(foo) atom used without %(bar) atom"
+ i18n: factorize "--foo outside a repository"
+ i18n: refactor "unrecognized %(foo) argument" strings
+ i18n: factorize "no directory given for --foo"
+ i18n: factorize "--foo requires --bar" and the like
+ i18n: tag.c factorize i18n strings
+ i18n: standardize "cannot open" and "cannot read"
+ i18n: turn "options are incompatible" into "cannot be used together"
+ i18n: refactor "%s, %s and %s are mutually exclusive"
+ i18n: refactor "foo and bar are mutually exclusive"

Similar message templates have been consolidated so that translators need to work on fewer number of messages.
source: <[email protected]>

* ja/perf-use-specified-shell (2021-12-25) 1 commit (merged to 'next' on 2022-01-03 at 8ca35ee3d0)
+ t/perf: do not run tests in user's $SHELL

Perf tests were run with end-user's shell, but it has been corrected to use the shell specified by $TEST_SHELL_PATH.
source: <[email protected]>

* jc/doc-submitting-patches-choice-of-base (2021-12-30) 1 commit (merged to 'next' on 2022-01-05 at eb97c01c24)
+ SubmittingPatchs: clarify choice of base and testing

Extend the guidance to choose the base commit to build your work on, and hint/nudge contributors to read others' changes.
source: <[email protected]>

* jh/p4-human-unit-numbers (2021-12-20) 2 commits
(merged to 'next' on 2022-01-03 at c385766ff4)
+ git-p4: show progress as an integer
+ git-p4: print size values in appropriate units

The way "git p4" shows file sizes in its output has been updated to use human-readable units.
source: <[email protected]>

* jh/p4-remove-unused (2021-12-22) 2 commits
(merged to 'next' on 2022-01-03 at 43f61b2fef)
+ git-p4: remove "rollback" verb
+ git-p4: remove "debug" verb

Remove a few commands from "git p4" that aren't very useful.
source: <[email protected]>

* jl/subtree-check-parents-argument-passing-fix (2022-01-04) 1 commit (merged to 'next' on 2022-01-05 at 7dc995f026)
+ subtree: fix argument handling in check_parents

Fix performance-releated bug in "git subtree" (in contrib/).
source: <[email protected]>

* js/branch-track-inherit (2021-12-20) 3 commits (merged to 'next' on 2022-01-03 at ab5fb34db3)
+ config: require lowercase for branch.*.autosetupmerge
+ branch: add flags and config to inherit tracking
+ branch: accept multiple upstream branches for tracking (this branch is used by gc/branch-recurse-submodules.)

"git -c branch.autosetupmerge=inherit branch new old" makes "new" to have the same upstream as the "old" branch, instead of marking "old" itself as its upstream.
source: <[email protected]>

* js/l10n-mention-ngettext-early-in-readme (2021-12-21) 1 commit (merged to 'next' on 2022-01-03 at a644f898ce)
+ l10n: README: call more attention to plural strings

Localization doc update.
source: <25107068cbbf8c9ce6886e66e25dff19e072583f.1639425295.git.steadmon@google.com>

* jt/pack-header-lshift-overflow (2022-01-12) 1 commit (merged to 'next' on 2022-01-12 at fa27107ab3)
+ packfile: fix off-by-one error in decoding logic

source: <[email protected]>

* jz/apply-3-corner-cases (2021-12-20) 1 commit
(merged to 'next' on 2022-01-03 at cb8e19c304)
+ git-apply: skip threeway in add / rename cases

"git apply --3way" bypasses the attempt to do a three-way application in more cases to address the regression caused by the recent change to use direct application as a fallback.
source: <[email protected]>

* lh/use-gnu-color-in-grep (2022-01-05) 1 commit
(merged to 'next' on 2022-01-05 at a0e2068cb7)
+ grep: align default colors with GNU grep ones

The color palette used by "git grep" has been updated to match that of GNU grep.
source: <[email protected]>

* ma/header-dup-cleanup (2022-01-10) 1 commit (merged to 'next' on 2022-01-12 at de7891f5e8)
+ cache.h: drop duplicate ensure_full_index() declaration

Code clean-up.
source: <[email protected]>

* ma/windows-dynload-fix (2022-01-09) 1 commit
(merged to 'next' on 2022-01-12 at af0a2e6c90)
+ lazyload: use correct calling conventions

Fix calling dynamically loaded functions on Windows.
source: <[email protected]>

* ms/t-readme-typofix (2022-01-04) 1 commit
(merged to 'next' on 2022-01-06 at 59288ebb22)
+ t/README: fix typo

Typofix.
source: <[email protected]>

* ps/lockfile-cleanup-fix (2022-01-07) 1 commit
(merged to 'next' on 2022-01-10 at ff0ae63f07)
+ fetch: fix deadlock when cleaning up lockfiles in async signals

Some lockfile code called free() in signal-death code path, which has been corrected.
source: <555ec6717ecab0fe6ef5660bcf0d61d59f84ef8b.1641552500.git.ps@pks.im>

* rb/nonstop-lacks-uncompress2 (2022-01-12) 1 commit (merged to 'next' on 2022-01-12 at 2387a47655)
+ build: NonStop ships with an older zlib

source: <[email protected]>

* ws/fast-export-with-revision-options (2021-12-21) 1 commit (merged to 'next' on 2022-01-03 at f94f7f70b6)
+ fast-export: fix surprising behavior with --first-parent

Use of certain "git rev-list" options with "git fast-export" created nonsense results (the worst two of which being "--reverse" and "--invert-grep --grep=<foo>"). The use of "--first-parent" is made to behave a bit more sensible than before.
source: <[email protected]>

--------------------------------------------------
[New Topics]

* jh/p4-fix-use-of-process-error-exception (2022-01-06) 1 commit (merged to 'next' on 2022-01-10 at 49d529bfd7)
+ git-p4: fix instantiation of CalledProcessError

Will cook in 'next'.
source: <[email protected]>

* jh/p4-spawning-external-commands-cleanup (2022-01-06) 3 commits (merged to 'next' on 2022-01-10 at 54b36b4e66)
+ git-p4: don't print shell commands as python lists
+ git-p4: pass command arguments as lists instead of using shell
+ git-p4: don't select shell mode using the type of the command argument

Will cook in 'next'.
source: <[email protected]>

* pb/pull-rebase-autostash-fix (2022-01-06) 1 commit
- pull --rebase: honor rebase.autostash even when fast-forwarding

"git pull --rebase" ignored the rebase.autostash configuration variable when the remote history is a descendant of our history, which has been corrected.

Expecting an ack or two.
source: <[email protected]>

* rs/grep-expr-cleanup (2022-01-06) 4 commits (merged to 'next' on 2022-01-10 at b70a3bb0fa)
+ grep: use grep_and_expr() in compile_pattern_and()
+ grep: extract grep_binexp() from grep_or_expr()
+ grep: use grep_not_expr() in compile_pattern_not()
+ grep: use grep_or_expr() in compile_pattern_or()

Code clean-up.

Will cook in 'next'.
source: <[email protected]>

* fs/ssh-signing-crlf (2022-01-07) 1 commit
- gpg-interface: trim CR from ssh-keygen

The code path that verifies signatures made with ssh were made to work better on a system with CRLF line endings.

Will merge to 'next'?
source: <[email protected]>

* jc/qsort-s-alignment-fix (2022-01-07) 2 commits (merged to 'next' on 2022-01-10 at 329fd6e09a)
+ stable-qsort: avoid using potentially unaligned access
+ compat/qsort_s.c: avoid using potentially unaligned access

Fix a hand-rolled alloca() imitation that may have violated alignment requirement of data being sorted in compatibility implementation of qsort_s() and stable qsort().

Will cook in 'next'.
source: <[email protected]>
source: <[email protected]>

* ps/avoid-unnecessary-hook-invocation-with-packed-refs (2022-01-12) 7 commits
- fixup! refs: open-code deletion of packed refs
- refs: skip hooks when deleting uncovered packed refs
- refs: do not execute reference-transaction hook on packing refs
- refs: demonstrate excessive execution of the reference-transaction hook
- refs: allow skipping the reference-transaction hook
- refs: allow passing flags when beginning transactions
- refs: open-code deletion of packed refs

Because a deletion of ref would need to remove it from both the loose ref store and the packed ref store, a delete-ref operation that logically removes one ref may end up invoking ref-transaction hook twice, which has been corrected.

Under review.
source: <[email protected]>

* rs/apply-symlinks-use-strset (2022-01-07) 1 commit (merged to 'next' on 2022-01-10 at 32497a67d5)
+ apply: use strsets to track symlinks

"git apply" (ab)used the util pointer of the string-list to keep track of how each symbolic link needs to be handled, which has been simplified by using strset.

Will cook in 'next'.
source: <[email protected]>

* ld/sparse-index-bash-completion (2022-01-10) 3 commits
- sparse-checkout: limit tab completion to a single level
- sparse-checkout: custom tab completion
- sparse-checkout: custom tab completion tests

The command line completion (in contrib/) learns to complete arguments give to "git sparse-checkout" command.
source: <[email protected]>

* bc/clarify-eol-attr (2022-01-12) 2 commits
- docs: correct documentation about eol attribute
- t0027: add tests for eol without text in .gitattributes

Doc and test update around the eol attribute.
source: <[email protected]>

* jz/rev-list-exclude-first-parent-only (2022-01-12) 1 commit
- git-rev-list: add --exclude-first-parent-only flag

"git log" and friends learned an option --exclude-first-parent-only to propagate UNINTERESTING bit down only along the first-parent chain, just like --first-parent option shows commits that lack the UNINTERESTING bit only along the first-parent chain.
source: <[email protected]>

* ab/refs-errno-cleanup (2022-01-13) 1 commit
- refs API: use "failure_errno", not "errno"

A brown-paper-bag fix on top of a topic that was merged during this cycle.

Will merge to 'next' and then to 'master'.
source: <[email protected]>

* ab/reftable-build-fixes (2022-01-13) 2 commits
- reftable tests: avoid "int" overflow, use "uint64_t"
- reftable: avoid initializing structs from structs

A few portability tweaks.

Will merge to 'next' and then to 'master'?
source: <[email protected]>

* en/present-despite-skipped (2022-01-13) 6 commits
. Accelerate clear_skip_worktree_from_present_files() by caching . Update documentation related to sparsity and the skip-worktree bit . repo_read_index: clear SKIP_WORKTREE bit from files present in worktree . unpack-trees: fix accidental loss of user changes
. t1011: add testcase demonstrating accidental loss of user modifications . Merge branch 'vd/sparse-clean-etc' into en/present-despite-skipped (this branch uses vd/sparse-clean-etc.)

With this topic merged, 'seen' seems to fail t1092.
source: <[email protected]>

* js/t1450-making-it-writable-does-not-need-full-posixperm (2022-01-13) 1 commit
- t1450-fsck: exec-bit is not needed to make loose object writable

Test fix.

Will merge to 'next'.
source: <[email protected]>

--------------------------------------------------
[Stalled]

* ds/sparse-checkout-requires-per-worktree-config (2021-12-21) 5 commits . sparse-checkout: use repo_config_set_worktree_gently()
. config: add repo_config_set_worktree_gently()
. worktree: add upgrade_to_worktree_config()
. config: make some helpers repo-aware
. setup: use a repository when upgrading format

"git sparse-checkout" wants to work with per-worktree configration, but did not work well in a worktree attached to a bare repository.
source: <[email protected]>

* pw/add-p-hunk-split-fix (2022-01-12) 2 commits
- builtin add -p: fix hunk splitting
- t3701: clean up hunk splitting tests

"git add -p" rewritten in C regressed hunk splitting in some cases, which has been corrected.

Will merge to 'next'?
source: <[email protected]>

* gc/fetch-negotiate-only-early-return (2022-01-12) 3 commits
- fetch --negotiate-only: do not update submodules
- fetch: skip tasks related to fetching objects
- fetch: use goto cleanup in cmd_fetch()

"git fetch --nogotiate-only" is an internal command used by "git push" to figure out which part of our history is missing from the other side. It should never recurse into submodules even when fetch.recursesubmodules configuration variable is set, nor it should trigger "gc". The code has been tightened up to ensure it only does common ancestry discovery and nothing else.

Almost there.
source: <[email protected]>

* je/http-better-error-output (2021-12-03) 1 commit
. http-backend: give a hint that web browser access is not supported

When the http-backend program, which is the server-side component for the smart HTTP transport, sends a "404 Not found" error, we deliberately did not say anything to the requesting client. We now send a message back to the browser to tell the user that they do not want to visit the URL via their browser, instead of a totally blank page.

Expecting a reroll.
Breaks its self tests.
cf. <[email protected]>
source: <[email protected]>

* cb/save-term-across-editor-invocation (2021-12-01) 3 commits
- fixup! editor: allow for saving/restoring terminal state
- editor: allow for saving/restoring terminal state
- terminal: teach save_term to fail when not foreground

Some editors are reported to leave the terminal in funny state after they exit on Windows. Work it around by saving and restoring the terminal state when needed.

Expecting a reroll.
cf. <CAPUEsphktbdxeV7hvF52Or3CVHS8oOk5-WV=xfEZa8kfCVVnVg@mail.gmail.com>
source: <[email protected]>

* ar/submodule-update (2021-10-13) 9 commits
. submodule--helper: rename helper functions
. submodule--helper: remove unused helpers
. submodule: move core cmd_update() logic to C
. submodule--helper: run update using child process struct . submodule--helper: allow setting superprefix for init_submodule() . submodule--helper: refactor get_submodule_displaypath() . submodule--helper: rename helpers for update-clone
. submodule--helper: get remote names from any repository . submodule--helper: split up ensure_core_worktree()

Rewrite of "git submodule update" in C.

Expecting a reroll?
cf. <[email protected]>
source: <[email protected]>

--------------------------------------------------
[Cooking]

* bc/csprng-mktemps (2022-01-04) 2 commits
- wrapper: use a CSPRNG to generate random file names
- wrapper: add a helper to generate numbers from a CSPRNG

Pick a better random number generator and use it when we prepare temporary filenames.

Expecting a reroll.
cf. <[email protected]>
source: <[email protected]>

* jc/reflog-parse-options (2022-01-10) 2 commits
(merged to 'next' on 2022-01-12 at 1659e49c4b)
+ builtin/reflog.c: use parse-options api for expire, delete subcommands
+ Merge branch 'ab/reflog-prep' into jc/reflog-parse-options

Use the parse-options API in "git reflog" command.

Will cook in 'next'.
source: <[email protected]>

* vd/sparse-clean-etc (2022-01-13) 9 commits
- update-index: reduce scope of index expansion in do_reupdate
- update-index: integrate with sparse index
- update-index: add tests for sparse-checkout compatibility
- checkout-index: integrate with sparse index
- checkout-index: add --ignore-skip-worktree-bits option
- checkout-index: expand sparse checkout compatibility tests
- clean: integrate with sparse index
- reset: reorder wildcard pathspec conditions
- reset: fix validation in sparse index test
(this branch is used by en/present-despite-skipped.)

"git update-index", "git checkout-index", and "git clean" are taught to work better with the sparse checkout feature.
source: <[email protected]>

* ms/update-index-racy (2022-01-07) 4 commits
- update-index: refresh should rewrite index in case of racy timestamps
- t7508: add tests capturing racy timestamp handling
- t7508: fix bogus mtime verification
- test-lib: introduce API for verifying file mtime

"git update-index --refresh" has been taught to deal better with racy timestamps (just like "git status" already does).

Will merge to 'next'.
source: <[email protected]>

* jc/find-header (2022-01-06) 1 commit
(merged to 'next' on 2022-01-10 at 8a13b4f0b3)
+ receive-pack.c: consolidate find header logic

Code clean-up.

Will cook in 'next'.
source: <[email protected]>

* jc/name-rev-stdin (2022-01-10) 2 commits
- name-rev.c: use strbuf_getline instead of limited size buffer
- name-rev: deprecate --stdin in favor of --annotate-stdin

"git name-rev --stdin" does not behave like usual "--stdin" at all. Start the process of renaming it to "--annotate-stdin".

Will merge to 'next'?
source: <[email protected]>

* en/remerge-diff (2021-12-30) 10 commits
- merge-ort: mark conflict/warning messages from inner merges as omittable
- show, log: include conflict/warning messages in --remerge-diff headers
- diff: add ability to insert additional headers for paths
- merge-ort: format messages slightly different for use in headers
- merge-ort: mark a few more conflict messages as omittable
- merge-ort: capture and print ll-merge warnings in our preferred fashion
- ll-merge: make callers responsible for showing warnings
- log: clean unneeded objects during log --remerge-diff
- show, log: provide a --remerge-diff capability
- Merge branch 'ns/tmp-objdir' into en/remerge-diff

"git log --remerge-diff" shows the difference from mechanical merge result and the merge result that is actually recorded.

Will merge to 'next'?
source: <[email protected]>

* bs/forbid-i18n-of-protocol-token-in-fetch-pack (2021-12-22) 2 commits
- fixup! fetch-pack: parameterize message containing 'ready' keyword
- fetch-pack: parameterize message containing 'ready' keyword

L10n support for a few error messages.

Expecting an ack for fixup.
source: <[email protected]>

* gc/branch-recurse-submodules (2022-01-10) 6 commits
- branch: add --recurse-submodules option for branch creation
- builtin/branch: clean up action-picking logic in cmd_branch()
- branch: add a dry_run parameter to create_branch()
- branch: make create_branch() always create a branch
- branch: move --set-upstream-to behavior to dwim_and_setup_tracking()
- Merge branch 'js/branch-track-inherit' into gc/branch-recurse-submodules

"git branch" learned the "--recurse-submodules" option.

Will merge to 'next'?
source: <[email protected]>

* hn/reftable-coverity-fixes (2021-12-22) 18 commits
- reftable: be more paranoid about 0-length memcpy calls
- reftable: add print functions to the record types
- reftable: make reftable_record a tagged union
- reftable: remove outdated file reftable.c
- reftable: implement record equality generically
- reftable: make reftable-record.h function signatures const correct
- reftable: handle null refnames in reftable_ref_record_equal
- reftable: drop stray printf in readwrite_test
- reftable: order unittests by complexity
- reftable: all xxx_free() functions accept NULL arguments
- reftable: fix resource warning
- reftable: ignore remove() return value in stack_test.c
- reftable: check reftable_stack_auto_compact() return value
- reftable: fix resource leak blocksource.c
- reftable: fix resource leak in block.c error path
- reftable: fix OOB stack write in print functions
- Merge branch 'hn/create-reflog-simplify' into hn/reftable-coverity-fixes
- Merge branch 'hn/reftable' into hn/reftable-coverity-fixes

Problems identified by Coverity in the reftable code have been corrected.

Expecting a reroll.
cf. <[email protected]>
source: <[email protected]>

* tb/midx-bitmap-corruption-fix (2022-01-04) 9 commits
- pack-bitmap.c: gracefully fallback after opening pack/MIDX
- midx: read RIDX chunk when present
- t/lib-bitmap.sh: parameterize tests over reverse index source
- t5326: move tests to t/lib-bitmap.sh
- t5326: extract test_rev_exists
- t5326: drop unnecessary setup
- pack-revindex.c: instrument loading on-disk reverse index
- midx.c: make changing the preferred pack safe
- t5326: demonstrate bitmap corruption after permutation

A bug that made multi-pack bitmap and the object order out-of-sync (hence the .midx data gets corrupted) has been fixed.

Waiting for a hopefully final review.
cf. <Ydceeo33Yt4N%[email protected]>
source: <[email protected]>

* pw/fix-some-issues-in-reset-head (2021-12-08) 14 commits
- rebase -m: don't fork git checkout
- rebase --apply: set ORIG_HEAD correctly
- rebase --apply: fix reflog
- reset_head(): take struct rebase_head_opts
- rebase: cleanup reset_head() calls
- reset_head(): make default_reflog_action optional
- reset_head(): factor out ref updates
- create_autostash(): remove unneeded parameter
- reset_head(): remove action parameter
- rebase --apply: don't run post-checkout hook if there is an error
- rebase: do not remove untracked files on checkout
- rebase: pass correct arguments to post-checkout hook
- t5403: refactor rebase post-checkout hook tests
- rebase: factor out checkout for up to date branch

Fix "some issues" in a helper function reset_head().

Expecting a reroll.
cf. <[email protected]>
cf. <[email protected]>
cf. <[email protected]>
cf. <[email protected]>
cf. <CABPp-BEHW4VLG18twcM_8iOco1jZ2iuGT+KN8aS+-sAAnBhTnw@mail.gmail.com>
source: <[email protected]>

* ab/cat-file (2022-01-12) 12 commits
(merged to 'next' on 2022-01-12 at ee4d43041d)
+ cat-file: s/_/-/ in typo'd usage_msg_optf() message
+ cat-file: don't whitespace-pad "(...)" in SYNOPSIS and usage output (merged to 'next' on 2022-01-05 at e145efa605)
+ cat-file: use GET_OID_ONLY_TO_DIE in --(textconv|filters)
+ object-name.c: don't have GET_OID_ONLY_TO_DIE imply *_QUIETLY
+ cat-file: correct and improve usage information
+ cat-file: fix remaining usage bugs
+ cat-file: make --batch-all-objects a CMDMODE
+ cat-file: move "usage" variable to cmd_cat_file()
+ cat-file docs: fix SYNOPSIS and "-h" output
+ parse-options API: add a usage_msg_optf()
+ cat-file tests: test messaging on bad objects/paths
+ cat-file tests: test bad usage

Assorted updates to "git cat-file", especially "-h".

Will cook in 'next'.
source: <[email protected]>
source: <[email protected]>

* ab/grep-patterntype (2021-12-27) 7 commits
- grep API: call grep_config() after grep_init()
- grep.c: don't pass along NULL callback value
- built-ins: trust the "prefix" from run_builtin()
- fixup! grep tests: add missing "grep.patternType" config tests
- grep tests: add missing "grep.patternType" config tests
- log tests: check if grep_config() is called by "log"-like cmds
- grep.h: remove unused "regex_t regexp" from grep_opt

Some code clean-up in the "git grep" machinery.

Reroll exists.
source: <[email protected]>

* js/use-builtin-add-i (2021-12-01) 2 commits
- add -i: default to the built-in implementation
- t2016: require the PERL prereq only when necessary

"git add -i" was rewritten in C some time ago and has been in testing; the reimplementation is now exposed to general public by default.

On hold.
There are known breakages on macOS.
cf. <[email protected]>
source: <[email protected]>

* jt/conditional-config-on-remote-url (2021-12-15) 2 commits
- config: include file if remote URL matches a glob
- config: make git_config_include() static

The conditional inclusion mechanism of configuration files using "[includeIf <condition>]" learns to base its decision on the URL of the remote repository the repository interacts with.

How does this one look these days?
source: <[email protected]>

* ab/ambiguous-object-name (2022-01-13) 6 commits
- object-name: re-use "struct strbuf" in show_ambiguous_object()
- object-name: iterate ambiguous objects before showing header
- object-name: show date for ambiguous tag objects
- object-name: make ambiguous object output translatable
- object-name: explicitly handle OBJ_BAD in show_ambiguous_object()
- object-name tests: add tests for ambiguous object blind spots

Error output given in response to an ambiguous object name has been improved.
source: <[email protected]>

* tl/ls-tree-oid-only (2022-01-06) 9 commits
- cocci: allow padding with strbuf_addf()
- ls-tree.c: introduce "--format" option
- ls-tree.c: introduce struct "show_tree_data"
- ls-tree.c: support --object-only option for "git-ls-tree"
- ls-tree: optimize naming and handling of "return" in show_tree()
- ls-tree: use "size_t", not "int" for "struct strbuf"'s "len"
- ls-tree: use "enum object_type", not {blob,tree,commit}_type
- ls-tree: add missing braces to "else" arms
- ls-tree: remove commented-out code

"git ls-tree" learns "--oid-only" option, similar to "--name-only", and more generalized "--format" option.
source: <[email protected]>

* ab/config-based-hooks-2 (2022-01-07) 17 commits
- run-command: remove old run_hook_{le,ve}() hook API
- receive-pack: convert push-to-checkout hook to hook.h
- read-cache: convert post-index-change to use hook.h
- commit: convert {pre-commit,prepare-commit-msg} hook to hook.h
- git-p4: use 'git hook' to run hooks
- send-email: use 'git hook run' for 'sendemail-validate'
- git hook run: add an --ignore-missing flag
- hooks: convert worktree 'post-checkout' hook to hook library
- hooks: convert non-worktree 'post-checkout' hook to hook library
- merge: convert post-merge to use hook.h
- am: convert applypatch-msg to use hook.h
- rebase: convert pre-rebase to use hook.h
- hook API: add a run_hooks_l() wrapper
- am: convert {pre,post}-applypatch to use hook.h
- gc: use hook library for pre-auto-gc hook
- hook API: add a run_hooks() wrapper
- hook: add 'run' subcommand

More "config-based hooks".

Will merge to 'next', with minor nits?
source: <[email protected]>

* jh/builtin-fsmonitor-part2 (2021-12-25) 31 commits
- fixup! t7527: create test for fsmonitor--daemon
- fixup! t/perf/p7519: speed up test on Windows
- t7527: test status with untracked-cache and fsmonitor--daemon
- fsmonitor: force update index after large responses
- fsmonitor--daemon: use a cookie file to sync with file system
- fsmonitor--daemon: periodically truncate list of modified files
- t/perf/p7519: add fsmonitor--daemon test cases
- t/perf/p7519: speed up test on Windows
- t/helper/test-chmtime: skip directories on Windows
- t/perf: avoid copying builtin fsmonitor files into test repo
- t7527: create test for fsmonitor--daemon
- t/helper/fsmonitor-client: create IPC client to talk to FSMonitor Daemon
- help: include fsmonitor--daemon feature flag in version info
- fsmonitor--daemon: implement handle_client callback
- compat/fsmonitor/fsm-listen-darwin: implement FSEvent listener on MacOS
- compat/fsmonitor/fsm-listen-darwin: add macos header files for FSEvent
- compat/fsmonitor/fsm-listen-win32: implement FSMonitor backend on Windows
- fsmonitor--daemon: create token-based changed path cache
- fsmonitor--daemon: define token-ids
- fsmonitor--daemon: add pathname classification
- fsmonitor--daemon: implement 'start' command
- fsmonitor--daemon: implement 'run' command
- compat/fsmonitor/fsm-listen-darwin: stub in backend for Darwin
- compat/fsmonitor/fsm-listen-win32: stub in backend for Windows
- fsmonitor--daemon: implement 'stop' and 'status' commands
- fsmonitor--daemon: add a built-in fsmonitor daemon
- fsmonitor: document builtin fsmonitor
- fsmonitor: use IPC to query the builtin FSMonitor daemon
- fsmonitor: config settings are repository-specific
- fsmonitor-ipc: create client routines for git-fsmonitor--daemon
- fsmonitor: enhance existing comments

Built-in fsmonitor (part 2).

Expecting a reroll.
Seems that the discussion stalled.
cf. <[email protected]>
source: <[email protected]>

* es/superproject-aware-submodules (2021-11-18) 5 commits
- submodule: use config to find superproject worktree
- submodule: record superproject gitdir during 'update'
- submodule: record superproject gitdir during absorbgitdirs
- introduce submodule.superprojectGitDir record
- t7400-submodule-basic: modernize inspect() helper

A configuration variable in a submodule points at the location of the superproject it is bound to (RFC).

Expecting a reroll.
cf. <[email protected]>
source: <[email protected]>

* ab/only-single-progress-at-once (2022-01-07) 7 commits
- *.c: use isatty(0|2), not isatty(STDIN_FILENO|STDERR_FILENO)
- pack-bitmap-write.c: don't return without stop_progress()
- progress.c: add temporary variable from progress struct
- progress.c tests: test some invalid usage
- progress.c tests: make start/stop commands on stdin
- progress.c test helper: add missing braces
- leak tests: fix a memory leak in "test-progress" helper

Further tweaks on progress API.

Getting there.
source: <[email protected]>

--------------------------------------------------
[Discarded]

* ms/customizable-ident-expansion (2021-12-25) 2 commits . fixup! keyword expansion: make "$Id$" string configurable . keyword expansion: make "$Id$" string configurable

Instead of "$Id$", user-specified string (like $FreeBSD$) can be used as an in-blob placeholder for keyword expansion.

Will discard.
Stalled for too long.
cf. <[email protected]>
cf. <[email protected]>
source: <[email protected]>

* tb/cruft-packs (2021-11-29) 17 commits
. sha1-file.c: don't freshen cruft packs
. builtin/gc.c: conditionally avoid pruning objects via loose . builtin/repack.c: add cruft packs to MIDX during geometric repack . builtin/repack.c: use named flags for existing_packs
. builtin/repack.c: allow configuring cruft pack generation . builtin/repack.c: support generating a cruft pack
. builtin/pack-objects.c: --cruft with expiration
. reachable: report precise timestamps from objects in cruft packs . reachable: add options to add_unseen_recent_objects_to_traversal . builtin/pack-objects.c: --cruft without expiration
. builtin/pack-objects.c: return from create_object_entry() . t/helper: add 'pack-mtimes' test-tool
. pack-mtimes: support writing pack .mtimes files
. chunk-format.h: extract oid_version()
. pack-write: pass 'struct packing_data' to 'stage_tmp_packfiles' . pack-mtimes: support reading .mtimes files
. Documentation/technical: add cruft-packs.txt

Instead of leaving unreachable objects in loose form when packing, or ejecting them into loose form when repacking, gather them in a packfile with an auxiliary file that records the last-use time of these objects.

Dropped per author request.
cf. <[email protected]>
cf. <[email protected]> cf. <[email protected]> cf. <[email protected]>
source: <[email protected]>

#1 lore raw
Quoted 9 lines · Junio C Hamano

There are a few "oops, what we merged recently is broken" topics that still are not in 'master', but otherwise what we have should be pretty much what we'll have in the final one.

- I am reasonably happy with ab/refs-errno-cleanup (just one patch) that fixes the incorrect state of the code left by the earlier parts of the topic that have already been merged during this cycle.

Quoted 8 lines

- I am also OK with ab/reftable-build-fixes (two patches), one for general type correctness fix, the other for helping older sub-C99 compilers.

If there are fixes for regressions that we introduced during this cycle other than these two topics, I certainly am missing them, so please holler loudly and quickly, hopefully in time for me to tag the -rc1 tomorrow.

Oh, by the way, the tip of 'seen' has consistently failing the leak-check test. I didn't have chance, time or energy to see if they are failing merely because an existing test script that used to be leak-clean gained a use of command that has been known to be leak-unclean without introducing any new leaks, or our recent change did introduce new leaks to commands that have been leak-clean. Somebody with too much time on their hand should go in and check to help, before CI testing on 'seen' becomes useful again.

Thanks.

#2 lore raw
Quoted 21 lines · Junio C Hamano
Quoted 9 lines · Junio C Hamano

There are a few "oops, what we merged recently is broken" topics that still are not in 'master', but otherwise what we have should be pretty much what we'll have in the final one.

- I am reasonably happy with ab/refs-errno-cleanup (just one patch) that fixes the incorrect state of the code left by the earlier parts of the topic that have already been merged during this cycle.

Quoted 8 lines

- I am also OK with ab/reftable-build-fixes (two patches), one for general type correctness fix, the other for helping older sub-C99 compilers.

If there are fixes for regressions that we introduced during this cycle other than these two topics, I certainly am missing them, so please holler loudly and quickly, hopefully in time for me to tag the -rc1 tomorrow.

Quoted 6 lines

Oh, by the way, the tip of 'seen' has consistently failing the leak-check test. I didn't have chance, time or energy to see if they are failing merely because an existing test script that used to be leak-clean gained a use of command that has been known to be leak-unclean without introducing any new leaks, or our recent change did introduce new leaks to commands that have been leak-clean.

It's apparently the latter, because there have been no test script changes in the relevant tests.

Quoted 2 lines

Somebody with too much time on their hand should go in and check to help, before CI testing on 'seen' becomes useful again.

This "fixes" seen:
https://lore.kernel.org/git/[email protected]/

I briefly looked at a couple leak traces and thought they looked ref related, but I don't have time to go hunt down memory leaks right now. I figure this thread has reported them, so let's just get "seen" back to green.

#3 lore raw
Quoted 6 lines · Elijah Newren

It's apparently the latter, because there have been no test script changes in the relevant tests.

Quoted 2 lines

Somebody with too much time on their hand should go in and check to help, before CI testing on 'seen' becomes useful again.

Quoted 7 lines

This "fixes" seen:
https://lore.kernel.org/git/[email protected]/

I briefly looked at a couple leak traces and thought they looked ref related, but I don't have time to go hunt down memory leaks right now. I figure this thread has reported them, so let's just get "seen" back to green.

If it were "we added a use of known-to-leak command in an otherwise clean test, without adding a new leak", I would wholeheartedly support such a change, but if it is the other way around, it may make sense to leave it broken as an incentive for people who care about leaks to go in and fix them up.

If we toggle it off any time leak-checker CI job starts complaining on a test script, the leak-checker CI job serves no useful purpose, no?

An obvious alternative, based on the same attitude, is to rip out the whole fragile leak-checker thing from the CI. I've mentioned an ideal alternative (disregarding feasibility) already elsewhere so I won't repeat it.

Thanks.

#4 lore raw
Quoted 17 lines · Junio C Hamano
Quoted 6 lines · Elijah Newren

It's apparently the latter, because there have been no test script changes in the relevant tests.

Quoted 2 lines

Somebody with too much time on their hand should go in and check to help, before CI testing on 'seen' becomes useful again.

Quoted 7 lines

This "fixes" seen:
https://lore.kernel.org/git/[email protected]/

I briefly looked at a couple leak traces and thought they looked ref related, but I don't have time to go hunt down memory leaks right now. I figure this thread has reported them, so let's just get "seen" back to green.

Quoted 5 lines

If it were "we added a use of known-to-leak command in an otherwise clean test, without adding a new leak", I would wholeheartedly support such a change, but if it is the other way around, it may make sense to leave it broken as an incentive for people who care about leaks to go in and fix them up.

Perhaps. Waiting can make sense up to a point.

Quoted 3 lines

If we toggle it off any time leak-checker CI job starts complaining on a test script, the leak-checker CI job serves no useful purpose, no?

Folks who use PRs for the purpose of getting the cross-platform CI testing before submitting to the list can still get early notification of potential leaks in their own series, due to the remaining tests being marked as leak-free. They can then fix up their series before submitting them to the list. That seems like a useful purpose to me.

Further, these CI jobs did notify us of an issue in someone else's patches (we don't yet know whose), and we were able to report it much like any other bug report. That gives people a heads up and allows them to take action on it. (And if they do so, they can remark the test as leak-free.) That also seems like a useful purpose to me.

In contrast, if we leave the leak-checker failing and the failing job spreads to next and master, then we'll just end up training everyone to ignore it -- both for their own PRs and in general. To me, that's what making the leak-checker serve no useful purpose would look like.

#5 lore raw
Quoted 4 lines · Elijah Newren

In contrast, if we leave the leak-checker failing and the failing job spreads to next and master, then we'll just end up training everyone to ignore it -- both for their own PRs and in general. To me, that's what making the leak-checker serve no useful purpose would look like.

What you proposed is no better than that.

Marking a test as "OK to fail", because somebody added a new leak, is a small step of removing the leak-checking job from the CI. Among 226 such tests, you killed one of them and 225 more to go.

And after you are done, nobody's PR will be blocked because they do not see a leak-checker breakage.

#6 lore raw
Quoted 4 lines · Junio C Hamano
Quoted 2 lines · Elijah Newren

In contrast, if we leave the leak-checker failing and the failing job spreads to next and master,...

In any case, I think Ævar identified an offending topic, so I kicked it out from 'seen'. If it makes the leak-checker job pass, that would be a much better outcome than anything else we have been discussing on this thread so far, and we'd hopefully set a good precedent to follow. (1) if your change adds a new leak, you'll not hit 'next', (2) as an exception, if new test added by you use a tool that wasn't used in it, and if the test is known to be leaky, it is OK to mark the test leak-checker-unclean.

Let's see how well the rest of the topics do.

#7 lore raw
In reply to Junio C Hamano
Quoted 20 lines · Junio C Hamano
Quoted 9 lines · Junio C Hamano

There are a few "oops, what we merged recently is broken" topics that still are not in 'master', but otherwise what we have should be pretty much what we'll have in the final one.

- I am reasonably happy with ab/refs-errno-cleanup (just one patch) that fixes the incorrect state of the code left by the earlier parts of the topic that have already been merged during this cycle.

Quoted 8 lines

- I am also OK with ab/reftable-build-fixes (two patches), one for general type correctness fix, the other for helping older sub-C99 compilers.

If there are fixes for regressions that we introduced during this cycle other than these two topics, I certainly am missing them, so please holler loudly and quickly, hopefully in time for me to tag the -rc1 tomorrow.

Quoted 8 lines

Oh, by the way, the tip of 'seen' has consistently failing the leak-check test. I didn't have chance, time or energy to see if they are failing merely because an existing test script that used to be leak-clean gained a use of command that has been known to be leak-unclean without introducing any new leaks, or our recent change did introduce new leaks to commands that have been leak-clean. Somebody with too much time on their hand should go in and check to help, before CI testing on 'seen' becomes useful again.

It's a regression in ps/avoid-unnecessary-hook-invocation-with-packed-refs, Patrick could you look into it? On your current "seen" doing a:

    git revert -m 1 48b388cbf31

Will make those 3x failing tests pass:
https://github.com/git/git/runs/4811683950?check_suite_focus=true

(That commit being: 48b388cbf31 (Merge branch 'ps/avoid-unnecessary-hook-invocation-with-packed-refs' into seen, 2022-01-13))

I didn't have much time to look now, but this mostly untested fix-up fixes up the topic under SANITIZE=leak (but may break something else). I ran the broken tests with SANITIZE=leak, and the normal tests without SANITIZE=leak, but didn't have time for further testing:

refs/packed-backend.c +1 -0
index ff96ee482a0..b8012f97009 100644
@@ -1577,6 +1577,7 @@ int packed_refs_delete_refs(struct ref_store *ref_store,
                        error(_("could not delete references: %s"), err.buf);
        }
 
+       ref_transaction_free(transaction);
        strbuf_release(&err);
        return ret;
 }

I.e. the moving around of the ref_transaction_free() is at fault somehow, probably...

Junio: Is there something you're doing to hide
https://github.com/gitster/git/actions ? My version of that is visible without being logged-in, but yours is a 404 (maybe it needs to be set up?)

If CI is setup to run on your topic push-outs it'll become easier both for you & others to see what topics in "seen" may be broken. Usually it would be a matter of browsing the page of pushed-out topics for failures (except in cases where the merger causes the faulure...)

#8 lore raw
Quoted 63 lines · Ævar Arnfjörð Bjarmason
Quoted 20 lines · Junio C Hamano
Quoted 9 lines · Junio C Hamano

There are a few "oops, what we merged recently is broken" topics that still are not in 'master', but otherwise what we have should be pretty much what we'll have in the final one.

- I am reasonably happy with ab/refs-errno-cleanup (just one patch) that fixes the incorrect state of the code left by the earlier parts of the topic that have already been merged during this cycle.

Quoted 8 lines

- I am also OK with ab/reftable-build-fixes (two patches), one for general type correctness fix, the other for helping older sub-C99 compilers.

If there are fixes for regressions that we introduced during this cycle other than these two topics, I certainly am missing them, so please holler loudly and quickly, hopefully in time for me to tag the -rc1 tomorrow.

Quoted 8 lines

Oh, by the way, the tip of 'seen' has consistently failing the leak-check test. I didn't have chance, time or energy to see if they are failing merely because an existing test script that used to be leak-clean gained a use of command that has been known to be leak-unclean without introducing any new leaks, or our recent change did introduce new leaks to commands that have been leak-clean. Somebody with too much time on their hand should go in and check to help, before CI testing on 'seen' becomes useful again.

It's a regression in ps/avoid-unnecessary-hook-invocation-with-packed-refs, Patrick could you look into it? On your current "seen" doing a:

    git revert -m 1 48b388cbf31

Will make those 3x failing tests pass:
https://github.com/git/git/runs/4811683950?check_suite_focus=true

(That commit being: 48b388cbf31 (Merge branch 'ps/avoid-unnecessary-hook-invocation-with-packed-refs' into seen, 2022-01-13))

I didn't have much time to look now, but this mostly untested fix-up fixes up the topic under SANITIZE=leak (but may break something else). I ran the broken tests with SANITIZE=leak, and the normal tests without SANITIZE=leak, but didn't have time for further testing:

refs/packed-backend.c +1 -0
index ff96ee482a0..b8012f97009 100644
@@ -1577,6 +1577,7 @@ int packed_refs_delete_refs(struct ref_store *ref_store,
                        error(_("could not delete references: %s"), err.buf);
        }
 
+       ref_transaction_free(transaction);
        strbuf_release(&err);
        return ret;
 }
Quoted 2 lines

I.e. the moving around of the ref_transaction_free() is at fault somehow, probably...

Thanks for digging! The bug is actually in the files backend, where files_delete_refs() has two different exit paths, but I added the free of the packed-refs backend only to one of both. So the following patch fixes it:

refs/files-backend.c +1 -0
index 9a20cb8fa8..844918cbd8 100644
@@ -1280,6 +1280,7 @@ static int files_delete_refs(struct ref_store *ref_store, const char *msg,
 			result |= error(_("could not remove reference %s"), refname);
 	}
 
+	ref_transaction_free(transaction);
 	strbuf_release(&err);
 	return result;
 

I'll send a reroll of my series.

Patrick

#9 lore raw

Mistakes in the stalled category? (Was: Re: What's cooking in git.git (Jan 2022, #03; Thu, 13))

In reply to Junio C Hamano

Hi,

Are there some errors with the stalled category this time around? In particular...

Quoted 11 lines · Junio C Hamano

--------------------------------------------------
[Stalled]

* ds/sparse-checkout-requires-per-worktree-config (2021-12-21) 5 commits . sparse-checkout: use repo_config_set_worktree_gently()
. config: add repo_config_set_worktree_gently()
. worktree: add upgrade_to_worktree_config()
. config: make some helpers repo-aware
. setup: use a repository when upgrading format

Quoted 3 lines

"git sparse-checkout" wants to work with per-worktree configration, but did not work well in a worktree attached to a bare repository.
source: <[email protected]>

It has been two weeks since the last submission and emails about this topic, so maybe you put this one in "stalled" intentionally. (If so, and Stolee if this really is stalled, would you like me to try updating? I know it has expanded quite a bit from the early simple fix you were trying to provide, but you've got most the code I think you need and some important fixes I wouldn't want to see dropped.) But I'm wondering if marking this topic as stalled was intentional, because:

Quoted 23 lines

* pw/add-p-hunk-split-fix (2022-01-12) 2 commits
- builtin add -p: fix hunk splitting
- t3701: clean up hunk splitting tests

"git add -p" rewritten in C regressed hunk splitting in some cases, which has been corrected.

Will merge to 'next'?
source: <[email protected]>

* gc/fetch-negotiate-only-early-return (2022-01-12) 3 commits
- fetch --negotiate-only: do not update submodules
- fetch: skip tasks related to fetching objects
- fetch: use goto cleanup in cmd_fetch()

"git fetch --nogotiate-only" is an internal command used by "git push" to figure out which part of our history is missing from the other side. It should never recurse into submodules even when fetch.recursesubmodules configuration variable is set, nor it should trigger "gc". The code has been tightened up to ensure it only does common ancestry discovery and nothing else.

Quoted 2 lines

Almost there.
source: <[email protected]>

These last two series were submitted in the two days; they don't seem stalled to me.

(The other series in "stalled" besides these three all seemed to belong there to me.)

#10 lore raw

Re: Mistakes in the stalled category?

Quoted 2 lines · Elijah Newren

Are there some errors with the stalled category this time around? In particular...

Yeah, I agree that the more recent ones would not belong there.

#11 lore raw

Re: Mistakes in the stalled category?

In reply to Elijah Newren
Quoted 13 lines · Elijah Newren
Quoted 9 lines

[Stalled]

* ds/sparse-checkout-requires-per-worktree-config (2021-12-21) 5 commits . sparse-checkout: use repo_config_set_worktree_gently()
. config: add repo_config_set_worktree_gently()
. worktree: add upgrade_to_worktree_config()
. config: make some helpers repo-aware
. setup: use a repository when upgrading format

Quoted 3 lines

"git sparse-checkout" wants to work with per-worktree configration, but did not work well in a worktree attached to a bare repository.
source: <[email protected]>

Quoted 6 lines

It has been two weeks since the last submission and emails about this topic, so maybe you put this one in "stalled" intentionally. (If so, and Stolee if this really is stalled, would you like me to try updating? I know it has expanded quite a bit from the early simple fix you were trying to provide, but you've got most the code I think you need and some important fixes I wouldn't want to see dropped.)

[Stalled] being early in the report is primarily a way to remind us that an update is overdue.

<[email protected]> is what prompted me to consider that I can put it on backburner and spend my time on other topics.

Quoted 1 line
Quoted 1 line

* pw/add-p-hunk-split-fix (2022-01-12) 2 commits

I think what happened was that the previous iteration of this was on 2021-12-20 and marked as "Expecting a reroll" for a few issues of the "What's cooking" report. And then I've moved it to [Stalled] section, but before I had a chance to issue another "What's cooking" report, an update came, and when I updated the list of commits, I forgot to move the topic out of the [Stalled] category.

Quoted 1 line
Quoted 1 line

* gc/fetch-negotiate-only-early-return (2022-01-12) 3 commits

Likewise, the previous round was on 2021-12-21.

#12 lore raw

Re: Mistakes in the stalled category?

Quoted 4 lines · Junio C Hamano
Quoted 2 lines · Elijah Newren
Quoted 2 lines

[Stalled]

Quoted 8 lines
Quoted 1 line
Quoted 1 line

* ds/sparse-checkout-requires-per-worktree-config (2021-12-21) 5 commits

[Stalled] being early in the report is primarily a way to remind us that an update is overdue.

<[email protected]> is what prompted me to consider that I can put it on backburner and spend my time on other topics.

The pushout of 'seen' I did after -rc1 last night included this topic again; https://github.com/git/git/actions/runs/1700157728

The CI job triggered Leak Sanitizer in "git init" (the first thing tests do is to prepare a test repository with working tree, so there may be other new breakages, but we wouldn't know) and broke CI.

This morning, just to test, I ejected this topic from 'seen' and pushed it out; https://github.com/git/git/actions/runs/1702373554

As the only difference between these two runs is this topic, it may be worth looking into locating and fixing new leaks introduced by it.

Thanks.

#13 lore raw

Re: Mistakes in the stalled category?

Quoted 22 lines · Junio C Hamano
Quoted 4 lines · Junio C Hamano
Quoted 2 lines · Elijah Newren
Quoted 2 lines

[Stalled]

Quoted 8 lines
Quoted 1 line
Quoted 1 line

* ds/sparse-checkout-requires-per-worktree-config (2021-12-21) 5 commits

[Stalled] being early in the report is primarily a way to remind us that an update is overdue.

<[email protected]> is what prompted me to consider that I can put it on backburner and spend my time on other topics.

The pushout of 'seen' I did after -rc1 last night included this topic again; https://github.com/git/git/actions/runs/1700157728

The CI job triggered Leak Sanitizer in "git init" (the first thing tests do is to prepare a test repository with working tree, so there may be other new breakages, but we wouldn't know) and broke CI.

Quoted 6 lines

This morning, just to test, I ejected this topic from 'seen' and pushed it out; https://github.com/git/git/actions/runs/1702373554

As the only difference between these two runs is this topic, it may be worth looking into locating and fixing new leaks introduced by it.

Thanks for pointing this out. I will work to fix this leak in the next version. Feel free to eject it until the new version is available.

Thanks,
-Stolee

#14 lore raw

en/present-despite-skipped & en/remerge-diff (Was: Re: What's cooking in git.git (Jan 2022, #03; Thu, 13))

In reply to Junio C Hamano
Quoted 10 lines · Junio C Hamano

* en/present-despite-skipped (2022-01-13) 6 commits
. Accelerate clear_skip_worktree_from_present_files() by caching . Update documentation related to sparsity and the skip-worktree bit . repo_read_index: clear SKIP_WORKTREE bit from files present in worktree . unpack-trees: fix accidental loss of user changes
. t1011: add testcase demonstrating accidental loss of user modifications . Merge branch 'vd/sparse-clean-etc' into en/present-despite-skipped (this branch uses vd/sparse-clean-etc.)

Quoted 2 lines

With this topic merged, 'seen' seems to fail t1092.
source: <[email protected]>

Doh! Sorry about that. I tried to merge with seen to check for semantic conflicts, because this series was more likely to have them than others. That's how I noticed the issue with vd/sparse-clean-etc. But I apparently missed ds/fetch-pull-with-sparse-index. Since the latter has now merged to master and your version of vd/sparse-clean-etc builds on top of a version of master that has ds/fetch-pull-with-sparse-index in it, I can simply submit an updated series and you won't have to merge anything extra into that topic. I'll submit it soon.

Quoted 12 lines

* en/remerge-diff (2021-12-30) 10 commits
- merge-ort: mark conflict/warning messages from inner merges as omittable
- show, log: include conflict/warning messages in --remerge-diff headers
- diff: add ability to insert additional headers for paths
- merge-ort: format messages slightly different for use in headers
- merge-ort: mark a few more conflict messages as omittable
- merge-ort: capture and print ll-merge warnings in our preferred fashion
- ll-merge: make callers responsible for showing warnings
- log: clean unneeded objects during log --remerge-diff
- show, log: provide a --remerge-diff capability
- Merge branch 'ns/tmp-objdir' into en/remerge-diff

Quoted 5 lines

"git log --remerge-diff" shows the difference from mechanical merge result and the merge result that is actually recorded.

Will merge to 'next'?
source: <[email protected]>

Back in September, when you last asked this about an earlier version of this series, I suggested waiting off[1]. Crazy how many things have happened since then. Anyway, this time we've gotten good feedback from multiple reviewers and I've addressed all of it, so this time I think it's ready to merge down.

[1] https://lore.kernel.org/git/CABPp-BH2C5HFw_=NtK71oNGXkntzcJd2un0TgMb=UnJ5yJxAAA@mail.gmail.com/

#15 lore raw

tb/midx-bitmap-corruption-fix (was: Re: What's cooking in git.git (Jan 2022, #03; Thu, 13))

In reply to Junio C Hamano
Quoted 11 lines · Junio C Hamano

* tb/midx-bitmap-corruption-fix (2022-01-04) 9 commits
- pack-bitmap.c: gracefully fallback after opening pack/MIDX
- midx: read RIDX chunk when present
- t/lib-bitmap.sh: parameterize tests over reverse index source
- t5326: move tests to t/lib-bitmap.sh
- t5326: extract test_rev_exists
- t5326: drop unnecessary setup
- pack-revindex.c: instrument loading on-disk reverse index
- midx.c: make changing the preferred pack safe
- t5326: demonstrate bitmap corruption after permutation

Quoted 6 lines

A bug that made multi-pack bitmap and the object order out-of-sync (hence the .midx data gets corrupted) has been fixed.

Waiting for a hopefully final review.
cf. <Ydceeo33Yt4N%[email protected]>
source: <[email protected]>

I would really like to get this into 2.35 since it's fixing an important source of repository corruption, but I think it should have a careful round of review before merging. And it is pretty late into the cycle anyway, so we may be too late to merge the whole thing.

But the first two patches:

- midx.c: make changing the preferred pack safe
- t5326: demonstrate bitmap corruption after permutation

could be applied as-is and the rest of the series left for later, which should be a safer approach (and would be sufficient to resolve the bug at the expense of some redundant bytes on disk[1]).

I think Stolee is probably the most familiar with this topic, but he is off currently and I'm not sure whether or not he'll be back with enough time to get this merged before 2.35.

On the other hand, the bug is pretty difficult to trigger, is affecting a very new feature, shouldn't ever cause permanent damage, and can be recovered from fairly easily (by dropping existing bitmaps). So perhaps it's OK to let it sit out for another release like this...

Thanks,
Taylor

[1]: More or less storing the contents of the multi-pack-index-$HASH.rev
file twice: once in the .rev file itself, and again as an optional write-only chunk in the MIDX.

#16 lore raw
In reply to Junio C Hamano
Quoted 5 lines · Junio C Hamano

There are a few "oops, what we merged recently is broken" topics that still are not in 'master', but otherwise what we have should be pretty much what we'll have in the final one.

Quoted 8 lines

- I am reasonably happy with ab/refs-errno-cleanup (just one patch) that fixes the incorrect state of the code left by the earlier parts of the topic that have already been merged during this cycle.

- I am also OK with ab/reftable-build-fixes (two patches), one for general type correctness fix, the other for helping older sub-C99 compilers.

I ran into the c99 test balloon. I get the (expected?) c99 build errors and an uncompress2 build error with RHEL7.9's stock gcc 4.8.5.

Are we dropping support for RHEL7.9?

Using NO_UNCOMPRESS2=YesPlease and CFLAGS += -std=c99 in config.mak works but I'm not sure if we actually expect users to deal with those details themselves.

Sorry for being out of the loop on the plan for older toolchains, but I wasn't sure if we should be expecting a patch that configures the default flags so that things build out of the box again or if this is considered the new status quo and we're okay regressing?

$ rpm -q zlib-devel
zlib-devel-1.2.7-19.el7_9.x86_64
$ gcc -v
...
gcc version 4.8.5 20150623 (Red Hat 4.8.5-44) (GCC)

Was part of the intention of the test balloon to get reports of this nature? I checked the commit logs and related discussion but that didn't really clarify things.

It seems like that was the intention based on the note of, "any warnings will eventually surface." in the commit message for 5f46385309 (config.mak.dev: specify -std=gnu99 for gcc/clang, 2021-12-08).

I'll keep an eye out for any patches that I should test in this area.

cheers,
--
David

#17 lore raw
Quoted 4 lines · David Aguilar

I ran into the c99 test balloon. I get the (expected?) c99 build errors and an uncompress2 build error with RHEL7.9's stock gcc 4.8.5.

Are we dropping support for RHEL7.9?

No. Our expectation is that those who care about various distros and platforms supply patches like ...

Quoted 1 line

Using NO_UNCOMPRESS2=YesPlease and CFLAGS += -std=c99 in config.mak

... this, but not locallly but to config.mak.uname, to help the users and builders on their favorite systems. It is better done sooner rather than later.

Quoted 2 lines

works but I'm not sure if we actually expect users to deal with those details themselves.

Users would not have to, as long as somebody else on the list uses the smea system as theirs cares enough to supply such patches to config.mak.uname.

Quoted 4 lines

Sorry for being out of the loop on the plan for older toolchains, but I wasn't sure if we should be expecting a patch that configures the default flags so that things build out of the box again or if this is considered the new status quo and we're okay regressing?

Obviously the former. This is a team effort and everybody who finds new issues is very appreciated to chip in.

#18 lore raw
Quoted 24 lines · Junio C Hamano
Quoted 4 lines · David Aguilar

I ran into the c99 test balloon. I get the (expected?) c99 build errors and an uncompress2 build error with RHEL7.9's stock gcc 4.8.5.

Are we dropping support for RHEL7.9?

No. Our expectation is that those who care about various distros and platforms supply patches like ...

Quoted 1 line

Using NO_UNCOMPRESS2=YesPlease and CFLAGS += -std=c99 in config.mak

... this, but not locallly but to config.mak.uname, to help the users and builders on their favorite systems. It is better done sooner rather than later.

Quoted 2 lines

works but I'm not sure if we actually expect users to deal with those details themselves.

Users would not have to, as long as somebody else on the list uses the smea system as theirs cares enough to supply such patches to config.mak.uname.

Quoted 8 lines
Quoted 4 lines

Sorry for being out of the loop on the plan for older toolchains, but I wasn't sure if we should be expecting a patch that configures the default flags so that things build out of the box again or if this is considered the new status quo and we're okay regressing?

Obviously the former. This is a team effort and everybody who finds new issues is very appreciated to chip in.

Thanks for verifying; patch sent. A simple $(uname_R) check worked for me so hopefully that's not too much cruft to carry around for a few years.

cheers,

Signature · 1 line
David