Add a macro to mark code sections that only read from the file system,
along with a config option and documentation.
This facilitates implementation of relatively simple file system level
caches without the need to synchronize with the file system.
Enable read-only sections for 'git status' and preload_index.
Signed-off-by: Karsten Blees <blees@dcon.de>
Reintroduce the 'core.useBuiltinFSMonitor' config setting (originally added
in 0a756b2a25 (fsmonitor: config settings are repository-specific,
2021-03-05)) after its removal from the upstream version of FSMonitor.
Upstream, the 'core.useBuiltinFSMonitor' setting was rendered obsolete by
"overloading" the 'core.fsmonitor' setting to take a boolean value. However,
several applications (e.g., 'scalar') utilize the original config setting,
so it should be preserved for a deprecation period before complete removal:
* if 'core.fsmonitor' is a boolean, the user is correctly using the new
config syntax; do not use 'core.useBuiltinFSMonitor'.
* if 'core.fsmonitor' is unspecified, use 'core.useBuiltinFSMonitor'.
* if 'core.fsmonitor' is a path, override and use the builtin FSMonitor if
'core.useBuiltinFSMonitor' is 'true'; otherwise, use the FSMonitor hook
indicated by the path.
Additionally, for this deprecation period, advise users to switch to using
'core.fsmonitor' to specify their use of the builtin FSMonitor.
Signed-off-by: Victoria Dye <vdye@github.com>
When adding many objects to a repo with `core.fsync=loose-object`,
the cost of fsync'ing each object file can become prohibitive.
One major source of the cost of fsync is the implied flush of the
hardware writeback cache within the disk drive. This commit introduces
a new `core.fsyncMethod=batch` option that batches up hardware flushes.
It hooks into the bulk-checkin odb-transaction functionality, takes
advantage of tmp-objdir, and uses the writeout-only support code.
When the new mode is enabled, we do the following for each new object:
1a. Create the object in a tmp-objdir.
2a. Issue a pagecache writeback request and wait for it to complete.
At the end of the entire transaction when unplugging bulk checkin:
1b. Issue an fsync against a dummy file to flush the log and hardware
writeback cache, which should by now have seen the tmp-objdir writes.
2b. Rename all of the tmp-objdir files to their final names.
3b. When updating the index and/or refs, we assume that Git will issue
another fsync internal to that operation. This is not the default
today, but the user now has the option of syncing the index and there
is a separate patch series to implement syncing of refs.
On a filesystem with a singular journal that is updated during name
operations (e.g. create, link, rename, etc), such as NTFS, HFS+, or XFS
we would expect the fsync to trigger a journal writeout so that this
sequence is enough to ensure that the user's data is durable by the time
the git command returns. This sequence also ensures that no object files
appear in the main object store unless they are fsync-durable.
Batch mode is only enabled if core.fsync includes loose-objects. If
the legacy core.fsyncObjectFiles setting is enabled, but core.fsync does
not include loose-objects, we will use file-by-file fsyncing.
In step (1a) of the sequence, the tmp-objdir is created lazily to avoid
work if no loose objects are ever added to the ODB. We use a tmp-objdir
to maintain the invariant that no loose-objects are visible in the main
ODB unless they are properly fsync-durable. This is important since
future ODB operations that try to create an object with specific
contents will silently drop the new data if an object with the target
hash exists without checking that the loose-object contents match the
hash. Only a full git-fsck would restore the ODB to a functional state
where dataloss doesn't occur.
In step (1b) of the sequence, we issue a fsync against a dummy file
created specifically for the purpose. This method has a little higher
cost than using one of the input object files, but makes adding new
callers of this mechanism easier, since we don't need to figure out
which object file is "last" or risk sharing violations by caching the fd
of the last object file.
_Performance numbers_:
Linux - Hyper-V VM running Kernel 5.11 (Ubuntu 20.04) on a fast SSD.
Mac - macOS 11.5.1 running on a Mac mini on a 1TB Apple SSD.
Windows - Same host as Linux, a preview version of Windows 11.
Adding 500 files to the repo with 'git add' Times reported in seconds.
object file syncing | Linux | Mac | Windows
--------------------|-------|-------|--------
disabled | 0.06 | 0.35 | 0.61
fsync | 1.88 | 11.18 | 2.47
batch | 0.15 | 0.41 | 1.53
Signed-off-by: Neeraj Singh <neerajsi@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This adds support for a new http.sslAutoClientCert config value.
In cURL 7.77 or later the schannel backend does not automatically send
client certificates from the Windows Certificate Store anymore.
This config value is only used if http.sslBackend is set to "schannel",
and can be used to opt in to the old behavior and force cURL to send
client certificates.
This fixes https://github.com/git-for-windows/git/issues/3292
Signed-off-by: Pascal Muller <pascalmuller@gmail.com>
The native Windows HTTPS backend is based on Secure Channel which lets
the caller decide how to handle revocation checking problems caused by
missing information in the certificate or offline CRL distribution
points.
Unfortunately, cURL chose to handle these problems differently than
OpenSSL by default: while OpenSSL happily ignores those problems
(essentially saying "¯\_(ツ)_/¯"), the Secure Channel backend will error
out instead.
As a remedy, the "no revoke" mode was introduced, which turns off
revocation checking altogether. This is a bit heavy-handed. We support
this via the `http.schannelCheckRevoke` setting.
In https://github.com/curl/curl/pull/4981, we contributed an opt-in
"best effort" strategy that emulates what OpenSSL seems to do.
In Git for Windows, we actually want this to be the default. This patch
makes it so, introducing it as a new value for the
`http.schannelCheckRevoke" setting, which now becmes a tristate: it
accepts the values "false", "true" or "best-effort" (defaulting to the
last one).
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Since commit 0c499ea60f the send-pack builtin uses the side-band-64k
capability if advertised by the server.
Unfortunately this breaks pushing over the dump git protocol if used
over a network connection.
The detailed reasons for this breakage are (by courtesy of Jeff Preshing,
quoted from ttps://groups.google.com/d/msg/msysgit/at8D7J-h7mw/eaLujILGUWoJ):
----------------------------------------------------------------------------
MinGW wraps Windows sockets in CRT file descriptors in order to mimic the
functionality of POSIX sockets. This causes msvcrt.dll to treat sockets as
Installable File System (IFS) handles, calling ReadFile, WriteFile,
DuplicateHandle and CloseHandle on them. This approach works well in simple
cases on recent versions of Windows, but does not support all usage patterns.
In particular, using this approach, any attempt to read & write concurrently
on the same socket (from one or more processes) will deadlock in a scenario
where the read waits for a response from the server which is only invoked after
the write. This is what send_pack currently attempts to do in the use_sideband
codepath.
----------------------------------------------------------------------------
The new config option "sendpack.sideband" allows to override the side-band-64k
capability of the server, and thus makes the dump git protocol work.
Other transportation methods like ssh and http/https still benefit from
the sideband channel, therefore the default value of "sendpack.sideband"
is still true.
[jes: split out the documentation into Documentation/config/]
Signed-off-by: Thomas Braun <thomas.braun@byte-physics.de>
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Oliver Schneider <oliver@assarbad.net>
"git worktree list --porcelain" did not c-quote pathnames and lock
reasons with unsafe bytes correctly, which is worked around by
introducing NUL terminated output format with "-z".
* pw/worktree-list-with-z:
worktree: add -z option for list subcommand
Built-in fsmonitor (part 2).
* jh/builtin-fsmonitor-part2: (30 commits)
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/perf/p7519: fix coding style
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
...
Give hint when branch tracking cannot be established because fetch
refspecs from multiple remote repositories overlap.
* tk/ambiguous-fetch-refspec:
tracking branches: add advice to ambiguous refspec error
"git fetch --refetch" learned to fetch everything without telling
the other side what we already have, which is useful when you
cannot trust what you have in the local object store.
* rc/fetch-refetch:
docs: mention --refetch fetch option
fetch: after refetch, encourage auto gc repacking
t5615-partial-clone: add test for fetch --refetch
fetch: add --refetch option
builtin/fetch-pack: add --refetch option
fetch-pack: add refetch
fetch-negotiator: add specific noop initializer
Code clean-up.
* ds/partial-bundle-more:
pack-objects: lazily set up "struct rev_info", don't leak
bundle: output hash information in 'verify'
bundle: move capabilities to end of 'verify'
pack-objects: parse --filter directly into revs.filter
pack-objects: move revs out of get_object_list()
list-objects-filter: remove CL_ARG__FILTER
"git ls-tree" learns "--oid-only" option, similar to "--name-only",
and more generalized "--format" option.
* tl/ls-tree-oid-only:
ls-tree: split up "fast path" callbacks
ls-tree: detect and error on --name-only --name-status
ls-tree: support --object-only option for "git-ls-tree"
ls-tree: introduce "--format" option
cocci: allow padding with `strbuf_addf()`
ls-tree: introduce struct "show_tree_data"
ls-tree: slightly refactor `show_tree()`
ls-tree: fix "--name-only" and "--long" combined use bug
ls-tree: simplify nesting if/else logic in "show_tree()"
ls-tree: rename "retval" to "recurse" 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
ls-tree tests: add tests for --name-status
The error "not tracking: ambiguous information for ref" is raised
when we are evaluating what tracking information to set on a branch,
and find that the ref to be added as tracking branch is mapped
under multiple remotes' fetch refspecs.
This can easily happen when a user copy-pastes a remote definition
in their git config, and forgets to change the tracking path.
Add advice in this situation, explicitly highlighting which remotes
are involved and suggesting how to correct the situation. Also
update a test to explicitly expect that advice.
Signed-off-by: Tao Klerks <tao@klerks.biz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a -z option to be used in conjunction with --porcelain that gives
NUL-terminated output. As 'worktree list --porcelain' does not quote
worktree paths this enables it to handle worktree paths that contain
newlines.
Signed-off-by: Phillip Wood <phillip.wood@dunelm.org.uk>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
We raised the weather balloon to see if we can allow the construct
in 44ba10d6 (revision: use C99 declaration of variable in for()
loop, 2021-11-14), which was shipped as a part of Git v2.35.
Document that fact in the coding guidelines, and more importantly,
give ourselves a deadline to revisit and update.
Let's declare that we will officially adopt the variable declaration
in the initializaiton part of "for ()" statement this winter, unless
we find that a platform we care about does not grok it.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git stash" does not allow subcommands it internally runs as its
implementation detail, except for "git reset", to emit messages;
now "git reset" part has also been squelched.
* vd/stash-silence-reset:
reset: show --no-refresh in the short-help
reset: remove 'reset.refresh' config option
reset: remove 'reset.quiet' config option
reset: do not make '--quiet' disable index refresh
stash: make internal resets quiet and refresh index
reset: suppress '--no-refresh' advice if logging is silenced
reset: replace '--quiet' with '--no-refresh' in performance advice
reset: introduce --[no-]refresh option to --mixed
reset: revise index refresh advice
Document it for partial clones as a means to apply a new filter, and
reference it from the remote.<name>.partialclonefilter config parameter.
Signed-off-by: Robert Coup <robert@coup.net.nz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
After invoking `fetch --refetch`, the object db will likely contain many
duplicate objects. If auto-maintenance is enabled, invoke it with
appropriate settings to encourage repacking/consolidation.
* gc.autoPackLimit: unless this is set to 0 (disabled), override the
value to 1 to force pack consolidation.
* maintenance.incremental-repack.auto: unless this is set to 0, override
the value to -1 to force incremental repacking.
Signed-off-by: Robert Coup <robert@coup.net.nz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Teach fetch and transports the --refetch option to force a full fetch
without negotiating common commits with the remote. Use when applying a
new partial clone filter to refetch all matching objects.
Signed-off-by: Robert Coup <robert@coup.net.nz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a refetch option to fetch-pack to force a full fetch. Use when
applying a new partial clone filter to refetch all matching objects.
Signed-off-by: Robert Coup <robert@coup.net.nz>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When "git fetch --recurse-submodules" grabbed submodule commits
that would be needed to recursively check out newly fetched commits
in the superproject, it only paid attention to submodules that are
in the current checkout of the superproject. We now do so for all
submodules that have been run "git submodule init" on.
* gc/recursive-fetch-with-unused-submodules:
submodule: fix latent check_has_commit() bug
fetch: fetch unpopulated, changed submodules
submodule: move logic into fetch_task_create()
submodule: extract get_fetch_task()
submodule: store new submodule commits oid_array in a struct
submodule: inline submodule_commits() into caller
submodule: make static functions read submodules from commits
t5526: create superproject commits with test helper
t5526: stop asserting on stderr literally
t5526: introduce test helper to assert on fetches
Updates to refs traditionally weren't fsync'ed, but we can
configure using core.fsync variable to do so.
* ps/fsync-refs:
core.fsync: new option to harden references
Replace core.fsyncObjectFiles with two new configuration variables,
core.fsync and core.fsyncMethod.
* ns/core-fsyncmethod:
core.fsync: documentation and user-friendly aggregate options
core.fsync: new option to harden the index
core.fsync: add configuration parsing
core.fsync: introduce granular fsync control infrastructure
core.fsyncmethod: add writeout-only mode
wrapper: make inclusion of Windows csprng header tightly scoped
Document how `core.fsmonitor` can be set to a boolean to enable
or disable the builtin FSMonitor.
Update references to `core.fsmonitor` and `core.fsmonitorHookVersion` and
pointers to `Watchman` to refer to it.
Create `git-fsmonitor--daemon` manual page and describe its features.
Signed-off-by: Jeff Hostetler <jeffhost@microsoft.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Remove the 'reset.refresh' option, requiring that users explicitly specify
'--no-refresh' if they want to skip refreshing the index.
The 'reset.refresh' option was introduced in 101cee42dd (reset: introduce
--[no-]refresh option to --mixed, 2022-03-11) as a replacement for the
refresh-skipping behavior originally controlled by 'reset.quiet'.
Although 'reset.refresh=false' functionally served the same purpose as
'reset.quiet=true', it exposed [1] the fact that the existence of a global
"skip refresh" option could potentially cause problems for users. Allowing a
global config option to avoid refreshing the index forces scripts using 'git
reset --mixed' to defensively use '--refresh' if index refresh is expected;
if that option is missing, behavior of a script could vary from user-to-user
without explanation.
Furthermore, globally disabling index refresh in 'reset --mixed' was
initially devised as a passive performance improvement; since the
introduction of the option, other changes have been made to Git (e.g., the
sparse index) with a greater potential performance impact without
sacrificing index correctness. Therefore, we can more aggressively err on
the side of correctness and limit the cases of skipping index refresh to
only when a user specifies the '--no-refresh' option.
[1] https://lore.kernel.org/git/xmqqy2179o3c.fsf@gitster.g/
Signed-off-by: Victoria Dye <vdye@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Remove the 'reset.quiet' config option, remove '--no-quiet' documentation in
'Documentation/git-reset.txt'. In 4c3abd0551 (reset: add new reset.quiet
config setting, 2018-10-23), 'reset.quiet' was introduced as a way to
globally change the default behavior of 'git reset --mixed' to skip index
refresh.
However, now that '--quiet' does not affect index refresh, 'reset.quiet'
would only serve to globally silence logging. This was not the original
intention of the config setting, and there's no precedent for such a setting
in other commands with a '--quiet' option, so it appears to be obsolete.
In addition to the options & its documentation, remove 'reset.quiet' from
the recommended config for 'scalar'.
Signed-off-by: Victoria Dye <vdye@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Update '--quiet' to no longer implicitly skip refreshing the index in a
mixed reset. Users now have the ability to explicitly disable refreshing the
index with the '--no-refresh' option, so they no longer need to use
'--quiet' to do so. Moreover, we explicitly remove the refresh-skipping
behavior from '--quiet' because it is completely unrelated to the stated
purpose of the option: "Be quiet, only report errors."
Helped-by: Phillip Wood <phillip.wood@dunelm.org.uk>
Signed-off-by: Victoria Dye <vdye@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
"git repack" learned a new configuration to disable triggering of
age-old "update-server-info" command, which is rarely useful these
days.
* ps/repack-with-server-info:
repack: add config to skip updating server info
repack: refactor to avoid double-negation of update-server-info
The 'filter' capability was added in 105c6f14a (bundle: parse filter
capability, 2022-03-09), but was added in a strange place in the 'git
bundle verify' output.
The tests for this show output like the following:
The bundle contains these 2 refs:
<COMMIT1> <REF1>
<COMMIT2> <REF2>
The bundle uses this filter: blob:none
The bundle records a complete history.
This looks very odd if we have a thin bundle that contains boundary
commits instead of a complete history:
The bundle contains these 2 refs:
<COMMIT1> <REF1>
<COMMIT2> <REF2>
The bundle uses this filter: blob:none
The bundle requires these 2 refs:
<COMMIT3>
<COMMIT4>
This separation between tip refs and boundary refs is unfortunate. Move
the filter capability output to the end of the output. Update the
documentation to match.
Signed-off-by: Derrick Stolee <derrickstolee@github.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
'--object-only' is an alias for '--format=%(objectname)'. It cannot
be used together other format-altering options like '--name-only',
'--long' or '--format', they are mutually exclusive.
The "--name-only" option outputs <filepath> only. Likewise, <objectName>
is another high frequency used field, so implement '--object-only' option
will bring intuitive and clear semantics for this scenario. Using
'--format=%(objectname)' we can achieve a similar effect, but the former
is with a lower learning cost(without knowing the format requirement
of '--format' option).
Even so, if a user is prefer to use "--format=%(objectname)", this is entirely
welcome because they are not only equivalent in function, but also have almost
identical performance. The reason is this commit also add the specific of
"--format=%(objectname)" to the current fast-pathes (builtin formats) to
avoid running unnecessary parsing mechanisms.
The following performance benchmarks are based on torvalds/linux.git:
When hit the fast-path:
Benchmark 1: /opt/git/ls-tree-oid-only/bin/git ls-tree -r --object-only HEAD
Time (mean ± σ): 83.6 ms ± 2.0 ms [User: 59.4 ms, System: 24.1 ms]
Range (min … max): 80.4 ms … 87.2 ms 35 runs
Benchmark 1: /opt/git/ls-tree-oid-only/bin/git ls-tree -r --format='%(objectname)' HEAD
Time (mean ± σ): 84.1 ms ± 1.8 ms [User: 61.7 ms, System: 22.3 ms]
Range (min … max): 80.9 ms … 87.5 ms 35 runs
But for a customized format, it will be slower:
Benchmark 1: /opt/git/ls-tree-oid-only/bin/git ls-tree -r --format='oid: %(objectname)' HEAD
Time (mean ± σ): 96.5 ms ± 2.5 ms [User: 72.9 ms, System: 23.5 ms]
Range (min … max): 93.1 ms … 104.1 ms 31 runs
Helped-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Teng Long <dyroneteng@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add a --format option to ls-tree. It has an existing default output,
and then --long and --name-only options to emit the default output
along with the objectsize and, or to only emit object paths.
Rather than add --type-only, --object-only etc. we can just support a
--format using a strbuf_expand() similar to "for-each-ref
--format". We might still add such options in the future for
convenience.
The --format implementation is slower than the existing code, but this
change does not cause any performance regressions. We'll leave the
existing show_tree() unchanged, and only run show_tree_fmt() in if
a --format different than the hardcoded built-in ones corresponding to
the existing modes is provided.
I.e. something like the "--long" output would be much slower with
this, mainly due to how we need to allocate various things to do with
quote.c instead of spewing the output directly to stdout.
The new option of '--format' comes from Ævar Arnfjörð Bjarmasonn's
idea and suggestion, this commit makes modifications in terms of the
original discussion on community [1].
In [1] there was a "GIT_TEST_LS_TREE_FORMAT_BACKEND" variable to
ensure that we had test coverage for passing tests that would
otherwise use show_tree() through show_tree_fmt(), and thus that the
formatting mechanism could handle all the same cases as the
non-formatting options.
Somewhere in subsequent re-rolls of that we seem to have drifted away
from what the goal of these tests should be. We're trying to ensure
correctness of show_tree_fmt(). We can't tell if we "hit [the]
fast-path" here, and instead of having an explicit test for that, we
can just add it to something our "test_ls_tree_format" tests for.
Here is the statistics about performance tests:
1. Default format (hitten the builtin formats):
"git ls-tree <tree-ish>" vs "--format='%(mode) %(type) %(object)%x09%(file)'"
$hyperfine --warmup=10 "/opt/git/master/bin/git ls-tree -r HEAD"
Benchmark 1: /opt/git/master/bin/git ls-tree -r HEAD
Time (mean ± σ): 105.2 ms ± 3.3 ms [User: 84.3 ms, System: 20.8 ms]
Range (min … max): 99.2 ms … 113.2 ms 28 runs
$hyperfine --warmup=10 "/opt/git/ls-tree-oid-only/bin/git ls-tree -r --format='%(mode) %(type) %(object)%x09%(file)' HEAD"
Benchmark 1: /opt/git/ls-tree-oid-only/bin/git ls-tree -r --format='%(mode) %(type) %(object)%x09%(file)' HEAD
Time (mean ± σ): 106.4 ms ± 2.7 ms [User: 86.1 ms, System: 20.2 ms]
Range (min … max): 100.2 ms … 110.5 ms 29 runs
2. Default format includes object size (hitten the builtin formats):
"git ls-tree -l <tree-ish>" vs "--format='%(mode) %(type) %(object) %(size:padded)%x09%(file)'"
$hyperfine --warmup=10 "/opt/git/master/bin/git ls-tree -r -l HEAD"
Benchmark 1: /opt/git/master/bin/git ls-tree -r -l HEAD
Time (mean ± σ): 335.1 ms ± 6.5 ms [User: 304.6 ms, System: 30.4 ms]
Range (min … max): 327.5 ms … 348.4 ms 10 runs
$hyperfine --warmup=10 "/opt/git/ls-tree-oid-only/bin/git ls-tree -r --format='%(mode) %(type) %(object) %(size:padded)%x09%(file)' HEAD"
Benchmark 1: /opt/git/ls-tree-oid-only/bin/git ls-tree -r --format='%(mode) %(type) %(object) %(size:padded)%x09%(file)' HEAD
Time (mean ± σ): 337.2 ms ± 8.2 ms [User: 309.2 ms, System: 27.9 ms]
Range (min … max): 328.8 ms … 349.4 ms 10 runs
Links:
[1] https://public-inbox.org/git/RFC-patch-6.7-eac299f06ff-20211217T131635Z-avarab@gmail.com/
[2] https://lore.kernel.org/git/cb717d08be87e3239117c6c667cb32caabaad33d.1646390152.git.dyroneteng@gmail.com/
Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: Teng Long <dyroneteng@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Bundle file format gets extended to allow a partial bundle,
filtered by similar criteria you would give when making a
partial/lazy clone.
* ds/partial-bundles:
clone: fail gracefully when cloning filtered bundle
bundle: unbundle promisor packs
bundle: create filtered bundles
rev-list: move --filter parsing into revision.c
bundle: parse filter capability
list-objects: handle NULL function pointers
MyFirstObjectWalk: update recommended usage
list-objects: consolidate traverse_commit_list[_filtered]
pack-bitmap: drop filter in prepare_bitmap_walk()
pack-objects: use rev.filter when possible
revision: put object filter into struct rev_info
list-objects-filter-options: create copy helper
index-pack: document and test the --promisor option
Fixes to the way generation number v2 in the commit-graph files are
(not) handled.
* ds/commit-graph-gen-v2-fixes:
commit-graph: declare bankruptcy on GDAT chunks
commit-graph: fix generation number v2 overflow values
commit-graph: start parsing generation v2 (again)
commit-graph: fix ordering bug in generation numbers
t5318: extract helpers to lib-commit-graph.sh
test-read-graph: include extra post-parse info
"git remote rename A B", depending on the number of remote-tracking
refs involved, takes long time renaming them. The command has been
taught to show progress bar while making the user wait.
* tb/rename-remote-progress:
builtin/remote.c: show progress when renaming remote references
builtin/remote.c: parse options in 'rename'
"git fetch --recurse-submodules" only considers populated
submodules (i.e. submodules that can be found by iterating the index),
which makes "git fetch" behave differently based on which commit is
checked out. As a result, even if the user has initialized all submodules
correctly, they may not fetch the necessary submodule commits, and
commands like "git checkout --recurse-submodules" might fail.
Teach "git fetch" to fetch cloned, changed submodules regardless of
whether they are populated. This is in addition to the current behavior
of fetching populated submodules (which is always attempted regardless
of what was fetched in the superproject, or even if nothing was fetched
in the superproject).
A submodule may be encountered multiple times (via the list of
populated submodules or via the list of changed submodules). When this
happens, "git fetch" only reads the 'populated copy' and ignores the
'changed copy'. Amend the verify_fetch_result() test helper so that we
can assert on which 'copy' is being read.
Signed-off-by: Glen Choo <chooglen@google.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When writing both loose and packed references to disk we first create a
lockfile, write the updated values into that lockfile, and on commit we
rename the file into place. According to filesystem developers, this
behaviour is broken because applications should always sync data to disk
before doing the final rename to ensure data consistency [1][2][3]. If
applications fail to do this correctly, a hard crash of the machine can
easily result in corrupted on-disk data.
This kind of corruption can in fact be easily observed with Git when the
machine hard-resets shortly after writing references to disk. On
machines with ext4, this will likely lead to the "empty files" problem:
the file has been renamed, but its data has not been synced to disk. The
result is that the reference is corrupt, and in the worst case this can
lead to data loss.
Implement a new option to harden references so that users and admins can
avoid this scenario by syncing locked loose and packed references to
disk before we rename them into place.
[1]: https://thunk.org/tytso/blog/2009/03/15/dont-fear-the-fsync/
[2]: https://btrfs.wiki.kernel.org/index.php/FAQ (What are the crash guarantees of overwrite-by-rename)
[3]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/admin-guide/ext4.rst (see auto_da_alloc)
Signed-off-by: Patrick Steinhardt <ps@pks.im>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
