3503 commits

Author	SHA1	Message	Date
Yuya Nishihara	57806ee8c2	test_commit_builder: reload repo per settings ... I'm going to remove the settings argument from start_transaction(), new_commit(), and rewrite_commit(), and these functions will use the settings passed in to the constructor.	2024-12-31 10:51:57 +09:00
Yuya Nishihara	475ac4e86a	repo: keep copy of UserSettings, remove RepoSettings ... The idea is that ReadonlyRepo/MutableRepo hold UserSettings to accomplish their operations (such as transaction commit.) Later patches will remove the "settings" argument from a couple of repo methods, which will greatly reduce the amount of settings refs we had to pass around. The current UserSettings is set up for the repo, so we wouldn't need a separate RepoSettings type.	2024-12-31 10:51:57 +09:00
Yuya Nishihara	feb032ee34	settings: make UserSettings cheap to clone ... I think the issue #5144 can be fixed cleanly if UserSettings object is loaded/resolved per repo/workspace, not per CommandHelper. This change makes it clear that UserSettings is supposed to be shared immutably.	2024-12-31 10:51:57 +09:00
Yuya Nishihara	43ac4faac4	config: preserve key formatting on set_value() ... I just noticed toml_edit::Table has a format-preserving insert() API. I think it's better to retain the user-specified quoting style.	2024-12-29 09:57:04 +09:00
Yuya Nishihara	10783f9e70	config: don't leave empty tables visible by set_value() ... It's unlikely that user would want to define all intermediate tables by "jj config set foo.bar.baz ..".	2024-12-29 09:57:04 +09:00
Yuya Nishihara	4af39e2038	cli: config list: mark values overridden by table or parent value ... This fixes the output of "jj config list --include-overridden --include-defaults ui.pager" and ".. ui.pager.command". The default ui.pager.* should be marked as overridden by ui.pager if set in user configuration.	2024-12-28 10:28:43 +09:00
Yuya Nishihara	e5361c6cc0	fsmonitor: move default settings to config/misc.toml	2024-12-25 10:44:37 +09:00
Yuya Nishihara	9b60a9df90	signing: move default backend settings to config/misc.toml	2024-12-25 10:44:37 +09:00
Yuya Nishihara	e4a350fcaa	config: extract jj-lib's default values to embedded TOML file ... It's nice that "jj config list --include-defaults" can show these default values. I just copied the jj-cli directory structure, but it's unlikely we'll add more config/*.toml files.	2024-12-25 10:44:37 +09:00
Yuya Nishihara	5bd669e892	settings: propagate configuration error of commit and operation parameters ... Note that infallible version of whoami::username() would return "Unknown" on error. I just made it error out, but it's also an option to fall back to an empty string.	2024-12-23 22:57:57 +09:00
Yuya Nishihara	4a69d0178c	settings: cache commit and operation parameters by UserSettings ... This helps propagate configuration error. RevsetParseContext is also updated because it was easier to pass &str in to it.	2024-12-23 22:57:57 +09:00
Yuya Nishihara	6f00c565b2	graph: inline ReverseGraphIterator to callers	2024-12-23 09:28:03 +09:00
Yuya Nishihara	ec853027be	graph: make reverse_graph() return nodes in "reversed" order	2024-12-23 09:28:03 +09:00
Yuya Nishihara	d6b84da382	graph: extract function that reverses graph edges ... ReverseGraphIterator will be inlined. It doesn't make sense the iterator yields Item = Result<..> whereas all possible errors are confined by constructor.	2024-12-23 09:28:03 +09:00
Yuya Nishihara	78b5766993	repo, workspace: use dunce::canonicalize() to normalize paths ... These paths may be printed, compared with user inputs, or passed to external programs. It's probably better to avoid unusual "\\?\C:\" paths on Windows. Fixes #5143	2024-12-22 09:45:37 +09:00
Scott Taylor	6baa43624c	local_working_copy: store materialized conflict marker length ... Storing the conflict marker length in the working copy makes conflict parsing more consistent, and it allows us to parse valid conflict hunks even if the user left some invalid conflict markers in the file while resolving the conflicts.	2024-12-21 11:36:30 -06:00
Scott Taylor	b11ce6bd28	conflicts: escape conflict markers by making them longer ... If a file contains lines which look like conflict markers, then we need to make the real conflict markers longer so that the materialized conflicts can be parsed unambiguously. When parsing the conflict, we require that the conflict markers are at least as long as the materialized conflict markers based on the current tree. This can lead to some unintuitive edge cases which will be solved in the next commit. For instance, if we have a file explaining the differences between Jujutsu's conflict markers and Git's conflict markers, it could produce a conflict with long markers like this: ``` <<<<<<<<<<< Conflict 1 of 1 %%%%%%%%%%% Changes from base to side #1 Jujutsu uses different conflict markers than Git, which just shows the -sides of a conflict without a diff. +sides of a conflict without a diff: + +<<<<<<< +left +||||||| +base +======= +right +>>>>>>> +++++++++++ Contents of side #2 Jujutsu uses different conflict markers than Git: <<<<<<< %%%%%%% -base +left +++++++ right >>>>>>> >>>>>>>>>>> Conflict 1 of 1 ends ``` We should support these options for "git" conflict marker style as well, since Git actually does support producing longer conflict markers in some cases through .gitattributes: https://git-scm.com/docs/gitattributes#_conflict_marker_size We may also want to support passing the conflict marker length to merge tools as well in the future, since Git supports a "%L" parameter to pass the conflict marker length to merge drivers: https://git-scm.com/docs/gitattributes#_defining_a_custom_merge_driver	2024-12-21 11:36:30 -06:00
Scott Taylor	369e8ea057	conflicts: refactor conflict marker writing and parsing ... These changes make the code a bit more readable, and they will make it easier to have conflict markers of different lengths in the next commit.	2024-12-21 11:36:30 -06:00
Yuya Nishihara	75ce7f6b7f	absorb: abandon source commit if it becomes discardable ... I don't think we need --keep-emptied flag. IIRC, "jj squash" has that flag in order not to squash commit description to the destination commits. Since "jj absorb" never moves commit description, the source commit is preserved in that situation. Closes #5141	2024-12-21 09:19:54 +09:00
Yuya Nishihara	bb6b551a4c	commit_builder: add is_empty(), is_discardable(), and .abandon() ... It's convenient to test if rewriter.reparent()-ed commit is empty than implementing the same check based on rewriter.new_parents(). CommitRewriter is an intermediate state, and it doesn't know whether the commit will be rebased or reparented.	2024-12-21 09:19:54 +09:00
Yuya Nishihara	f7fd523dd8	rewrite: remove unneeded Result from CommitRewriter::reparent() ... reparent() just configures new CommitBuilder instance, which should never fail.	2024-12-21 09:19:54 +09:00
Yuya Nishihara	d0f0e8dcff	settings: propagate error from git_settings() and signing_backend()	2024-12-21 09:19:44 +09:00
Yuya Nishihara	52511f491e	settings: inline CLI options to callers, propagate type error ... This patch moves max_new_file_size() and conflict_marker_style() to CLI, but there isn't a clear boundary whether the configuration should be managed by UserSettings or not. I decided to move them to CLI just because we can eliminate .optional() handling. The default parameters are defined in config/misc.toml.	2024-12-21 09:19:44 +09:00
Benjamin Tan	dddeb8b526	lib: fix various typos ... This commit fixes typos unintentionally introduced in d9c68e08, when renaming `jj branch` to `jj bookmark`.	2024-12-21 02:46:24 +08:00
Yuya Nishihara	89d3a8b8b7	config: remove check for relative path patterns, add doc about path equivalence ... I originally added the check so we would never canonicalize path relative to random cwd, but Windows CI failed because "/" is a relative path. Suppose user would want to share the same configuration file between native Windows and WSL, this check would be too nitpicky.	2024-12-20 09:23:46 +09:00
George Tsiamasiotis	843e0edcf7	jj-lib: fix typo in tests	2024-12-19 11:35:44 -08:00
Yuya Nishihara	dc9caa5b0a	config: add function to resolve conditional tables ... This provides an inline version of Git's includeIf. The idea is that each config layer has an optional key "--when" to enable the layer, and a layer may have multiple sub-layer tables "--scope" to inline stuff. Layers can be nested, but that wouldn't be useful in practice. I choose "--" prefix to make these meta keys look special (and are placed earlier in lexicographical order), but I don't have strong opinion about that. We can use whatever names that are unlikely to conflict with the other config sections. resolve() isn't exported as a StackedConfig method because it doesn't make sense to call resolve() on "resolved" StackedConfig. I'll add a newtype in CLI layer to distinguish raw config from resolved one. Most consumers will just see the "resolved" config. #616	2024-12-19 11:09:02 +09:00
Yuya Nishihara	7d46207fa6	config: wrap ConfigLayer in Arc so it can be cheaply cloned ... ConfigLayers will be cloned to new StackedConfig instance at resolution stage. ConfigFile could own ConfigLayer instead of Arc<_>, but copy-on-write behavior is probably better for the current users.	2024-12-19 11:09:02 +09:00
Yuya Nishihara	6bbaf79ca5	settings: parse TOML date-time value as well as string timestamp ... We can usually omit quotes in --config=NAME=VALUE, but an RFC3339 string is a valid TOML date-time expression. It's weird that quoting is required to specify a date-time value.	2024-12-18 09:51:56 +09:00
Yuya Nishihara	4f08c62fe5	settings: propagate error from UserSettings::from_config() ... All variables parsed here are debug options, but it would be annoying if timestamp options were silently ignored because of a typo.	2024-12-18 09:51:56 +09:00
Yuya Nishihara	8fc06690d2	settings: cache "debug.operation-timestamp" value ... This will make error handling easier in later commit.	2024-12-18 09:51:56 +09:00
Martin von Zweigbergk	b836e0ae95	docs/cli: update URLs to from martinvonz user to jj-vcs org ... We just migrated to the jj-vcs GitHub org, so we should point to the new GitHub URLs.	2024-12-17 12:44:44 -08:00
dploch	396833f803	absorb: move core functionality from cli to lib	2024-12-17 14:42:30 -05:00
Yuya Nishihara	abf48576ea	cli: replace --config-toml=TOML with --config=NAME=VALUE	2024-12-17 20:12:12 +09:00
Scott Taylor	d75cc94c64	gitignore: remove FIXME comment about passing path ... I looked through the code for the `ignore` crate, and this optional path is not used anywhere. The only reason to pass it would be to be able to get the path from the `Glob` when we call `Gitignore::matched` or `Gitignore::matched_path_or_any_parents`, but we ignore the returned `Glob` completely anyway. Passing the path would require an unnecessary clone of the path for each line in every .gitignore file, so it's better not to pass it since we don't need it.	2024-12-16 21:02:56 -06:00
Scott Taylor	9591523db4	gitignore: add more detail to errors on invalid pattern	2024-12-16 18:49:48 -06:00
Yuya Nishihara	ac52e43435	cli: let "config set"/"unset" continue if file to edit can be disambiguated ... If the user config path was an empty directory, these commands would fail with EISDIR instead of "can't set config in path (dirs not supported)". I think this can be changed later to create new "$dir/config.toml" file. Maybe we can also change the default path to ~/.config/jj directory, and load all *.toml files from there?	2024-12-15 16:36:29 +09:00
Yuya Nishihara	d6ca0c9940	config: add convenient ConfigLayer wrapper that provides .save() method ... I'm going to remove write/remove_config_value_to/from_file() functions, but I don't want to copy layer.path.expect(..) to all callers.	2024-12-14 22:12:41 +09:00
Yuya Nishihara	215c82e975	config: add layer.delete_value(name) method ... Since .get("path.to.non-table.children") returns NotFound, I made .delete_value() not fail in that case. The path doesn't exist, so .delete_value() should be noop. remove_config_value_from_file() will be inlined to callers.	2024-12-14 22:12:41 +09:00
Yuya Nishihara	fca92f1e16	cli: add --config-file=PATH argument ... This would be useful for scripting purpose. Maybe we can also replace the current --config-toml=<TOML> use cases by --config-file=<PATH> and simpler --config=<KEY>=<VALUE>. https://github.com/martinvonz/jj/issues/4926#issuecomment-2506672165 If we want to add more source variants (such as fd number), it might be better to add --config-from=<type>:<path|fd|..>. In any case, we'll probably want --config=<KEY>=<VALUE>, and therefore, we'll need to merge more than one --config* arguments.	2024-12-13 10:27:03 +09:00
Yuya Nishihara	c29b5a2aa3	cli: set global args to config table without re-parsing as TOML ... This should be safer than constructing a parsable TOML form.	2024-12-13 10:27:03 +09:00
Yuya Nishihara	0de36918e4	config: merge and print inline tables as values ... Before, "jj config get"/"list" and .get() functions processed inline tables as tables (or directories in filesystem analogy), whereas "set"/"unset" processed ones as values (or files.) This patch makes all commands and functions process inline tables as values. We rarely use the inline table syntax, and it's very hard to pack many (unrelated) values into an inline table. TOML doesn't allow newlines between { .. }. Our common use case is to define color styles, which wouldn't be meant to inherit attributes from the default settings. The default pager setting is flattened in case user overrides pager.env without changing the command args.	2024-12-12 10:11:51 +09:00
Yuya Nishihara	168c7979fe	working_copy: on snapshot, warn new large files and continue ... I think this provides a better UX than refusing any operation due to large files. Because untracked files won't be overwritten, it's usually safe to continue operation ignoring the untracked files. One caveat is that new large files can become tracked if files of the same name checked out. (see the test case) FWIW, the warning will be printed only once if watchman is enabled. If we use the snapshot stats to print untracked paths in "jj status", this will be a problem. Closes #3616, #3912	2024-12-11 20:19:51 +09:00
Yuya Nishihara	f4fdc19d9e	working_copy: plumbing to propagate untracked paths to caller	2024-12-11 20:19:51 +09:00
Yuya Nishihara	0975cb5374	cargo: drop dependency on config crate	2024-12-10 16:08:50 +09:00
Yuya Nishihara	118e76e5c2	config: add ConfigLoadError, replace uses of config::ConfigError	2024-12-10 16:08:50 +09:00
Yuya Nishihara	e2be4fa1ac	config: migrate underlying data structure to toml_edit ... This patch does not change the handling of inline tables yet. Both inline and non-inline tables are merged as before. OTOH, .set_value() is strict about table types because it should refuse to overwrite a table whereas an inline table should be overwritten as a value. This matches "jj config set"/"unset" semantics. rules_from_config() in formatter.rs uses .as_inline_table(), which is valid because toml_edit::Value type never contains non-inline table. Since toml_edit::Value doesn't implement PartialEq, stacking tests now use insta::assert_snapshot!().	2024-12-10 12:11:39 +09:00
Yuya Nishihara	45c80c2a5f	local_working_copy: optimize path comparison in prefixed file states ... Since all entries in filtered file states share the same directory prefix, we don't need to compare full file paths. The added functions take (path, name) instead of (path, sub_path) because the comparison can be slightly faster if the name is guaranteed to be a single path component. Benchmark: 1. original (omitted) 2. per-directory spawn (omitted) 3. per-directory deleted files (previous patch) 4. shorter path comparison (this patch) gecko-dev (~357k files, ~25k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 3: target/release-with-debug/jj-3 -R ~/mirrors/gecko-dev debug snapshot Time (mean ± σ): 480.1 ms ± 8.8 ms [User: 3190.5 ms, System: 2127.2 ms] Range (min … max): 471.2 ms … 509.8 ms 30 runs Benchmark 4: target/release-with-debug/jj-4 -R ~/mirrors/gecko-dev debug snapshot Time (mean ± σ): 404.0 ms ± 4.4 ms [User: 1933.4 ms, System: 2148.8 ms] Range (min … max): 396.4 ms … 416.9 ms 30 runs Relative speed comparison 1.19 ± 0.03 target/release-with-debug/jj-3 -R ~/mirrors/gecko-dev debug snapshot 1.00 target/release-with-debug/jj-4 -R ~/mirrors/gecko-dev debug snapshot ``` linux (~87k files, ~6k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 3: target/release-with-debug/jj-3 -R ~/mirrors/linux debug snapshot Time (mean ± σ): 204.2 ms ± 3.0 ms [User: 667.3 ms, System: 545.6 ms] Range (min … max): 197.1 ms … 209.2 ms 30 runs Benchmark 4: target/release-with-debug/jj-4 -R ~/mirrors/linux debug snapshot Time (mean ± σ): 191.3 ms ± 3.3 ms [User: 467.4 ms, System: 542.2 ms] Range (min … max): 186.1 ms … 200.6 ms 30 runs Relative speed comparison 1.07 ± 0.02 target/release-with-debug/jj-3 -R ~/mirrors/linux debug snapshot 1.00 target/release-with-debug/jj-4 -R ~/mirrors/linux debug snapshot ``` nixpkgs (~45k files, ~31k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 3: target/release-with-debug/jj-3 -R ~/mirrors/nixpkgs debug snapshot Time (mean ± σ): 173.3 ms ± 6.7 ms [User: 899.4 ms, System: 889.0 ms] Range (min … max): 166.5 ms … 197.9 ms 30 runs Benchmark 4: target/release-with-debug/jj-4 -R ~/mirrors/nixpkgs debug snapshot Time (mean ± σ): 161.7 ms ± 2.5 ms [User: 739.1 ms, System: 881.7 ms] Range (min … max): 156.5 ms … 166.4 ms 30 runs Relative speed comparison 1.07 ± 0.04 target/release-with-debug/jj-3 -R ~/mirrors/nixpkgs debug snapshot 1.00 target/release-with-debug/jj-4 -R ~/mirrors/nixpkgs debug snapshot ``` git (~4.5k files, 0.2k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 30 --runs 50 .. Benchmark 3: target/release-with-debug/jj-3 -R ~/mirrors/git debug snapshot Time (mean ± σ): 28.8 ms ± 1.0 ms [User: 33.0 ms, System: 37.6 ms] Range (min … max): 26.8 ms … 31.3 ms 50 runs Benchmark 4: target/release-with-debug/jj-4 -R ~/mirrors/git debug snapshot Time (mean ± σ): 28.8 ms ± 1.9 ms [User: 30.3 ms, System: 36.5 ms] Range (min … max): 26.0 ms … 39.2 ms 50 runs Relative speed comparison 1.00 target/release-with-debug/jj-3 -R ~/mirrors/git debug snapshot 1.00 ± 0.08 target/release-with-debug/jj-4 -R ~/mirrors/git debug snapshot ```	2024-12-10 10:51:04 +09:00
Yuya Nishihara	8caec186c1	local_working_copy: filter deleted files per directory (or job) ... This greatly reduces the amount of paths to be sent over the channel and the strings to be hashed. Benchmark: 1. original (omitted) 2. per-directory spawn (previous patch) 3. per-directory deleted files (this patch) 4. shorter path comparison (omitted) gecko-dev (~357k files, ~25k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 2: target/release-with-debug/jj-2 -R ~/mirrors/gecko-dev debug snapshot Time (mean ± σ): 710.7 ms ± 9.1 ms [User: 3070.7 ms, System: 2142.6 ms] Range (min … max): 695.9 ms … 740.1 ms 30 runs Benchmark 3: target/release-with-debug/jj-3 -R ~/mirrors/gecko-dev debug snapshot Time (mean ± σ): 480.1 ms ± 8.8 ms [User: 3190.5 ms, System: 2127.2 ms] Range (min … max): 471.2 ms … 509.8 ms 30 runs Relative speed comparison 1.76 ± 0.03 target/release-with-debug/jj-2 -R ~/mirrors/gecko-dev debug snapshot 1.19 ± 0.03 target/release-with-debug/jj-3 -R ~/mirrors/gecko-dev debug snapshot ``` linux (~87k files, ~6k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 2: target/release-with-debug/jj-2 -R ~/mirrors/linux debug snapshot Time (mean ± σ): 242.3 ms ± 3.3 ms [User: 656.8 ms, System: 538.0 ms] Range (min … max): 236.9 ms … 252.3 ms 30 runs Benchmark 3: target/release-with-debug/jj-3 -R ~/mirrors/linux debug snapshot Time (mean ± σ): 204.2 ms ± 3.0 ms [User: 667.3 ms, System: 545.6 ms] Range (min … max): 197.1 ms … 209.2 ms 30 runs Relative speed comparison 1.27 ± 0.03 target/release-with-debug/jj-2 -R ~/mirrors/linux debug snapshot 1.07 ± 0.02 target/release-with-debug/jj-3 -R ~/mirrors/linux debug snapshot ``` nixpkgs (~45k files, ~31k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 2: target/release-with-debug/jj-2 -R ~/mirrors/nixpkgs debug snapshot Time (mean ± σ): 190.7 ms ± 4.1 ms [User: 859.3 ms, System: 881.1 ms] Range (min … max): 184.6 ms … 202.4 ms 30 runs Benchmark 3: target/release-with-debug/jj-3 -R ~/mirrors/nixpkgs debug snapshot Time (mean ± σ): 173.3 ms ± 6.7 ms [User: 899.4 ms, System: 889.0 ms] Range (min … max): 166.5 ms … 197.9 ms 30 runs Relative speed comparison 1.18 ± 0.03 target/release-with-debug/jj-2 -R ~/mirrors/nixpkgs debug snapshot 1.07 ± 0.04 target/release-with-debug/jj-3 -R ~/mirrors/nixpkgs debug snapshot ``` git (~4.5k files, 0.2k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 30 --runs 50 .. Benchmark 2: target/release-with-debug/jj-2 -R ~/mirrors/git debug snapshot Time (mean ± σ): 30.6 ms ± 1.1 ms [User: 33.8 ms, System: 39.0 ms] Range (min … max): 29.0 ms … 35.0 ms 50 runs Benchmark 3: target/release-with-debug/jj-3 -R ~/mirrors/git debug snapshot Time (mean ± σ): 28.8 ms ± 1.0 ms [User: 33.0 ms, System: 37.6 ms] Range (min … max): 26.8 ms … 31.3 ms 50 runs Relative speed comparison 1.06 ± 0.05 target/release-with-debug/jj-2 -R ~/mirrors/git debug snapshot 1.00 target/release-with-debug/jj-3 -R ~/mirrors/git debug snapshot ```	2024-12-10 10:51:04 +09:00
Yuya Nishihara	99d8703d3b	local_working_copy: spawn snapshot job per directory with file count threshold ... This change basically means two things: a. a directory scan isn't split into too many small jobs, and b. a directory scan isn't blocked by recursive visit_directory() calls. Before, small jobs were created at each recursion depth, so there were silent time slice before these jobs started producing work. I don't know if this mitigates the issue #4508, but it's slightly faster on my Linux machine. matcher.visit(dir) is moved to caller because it's silly to spawn an empty job. TreeState::snapshot() already checks that for the root path. Benchmark: 1. original 2. per-directory spawn (this patch) 3. per-directory deleted files (omitted) 4. shorter path comparison (omitted) gecko-dev (~357k files, ~25k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 1: target/release-with-debug/jj-1 -R ~/mirrors/gecko-dev debug snapshot Time (mean ± σ): 764.9 ms ± 16.7 ms [User: 3274.7 ms, System: 2183.3 ms] Range (min … max): 731.9 ms … 814.2 ms 30 runs Benchmark 2: target/release-with-debug/jj-2 -R ~/mirrors/gecko-dev debug snapshot Time (mean ± σ): 710.7 ms ± 9.1 ms [User: 3070.7 ms, System: 2142.6 ms] Range (min … max): 695.9 ms … 740.1 ms 30 runs Relative speed comparison 1.89 ± 0.05 target/release-with-debug/jj-1 -R ~/mirrors/gecko-dev debug snapshot 1.76 ± 0.03 target/release-with-debug/jj-2 -R ~/mirrors/gecko-dev debug snapshot ``` linux (~87k files, ~6k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 1: target/release-with-debug/jj-1 -R ~/mirrors/linux debug snapshot Time (mean ± σ): 268.2 ms ± 11.3 ms [User: 636.6 ms, System: 518.5 ms] Range (min … max): 247.5 ms … 295.2 ms 30 runs Benchmark 2: target/release-with-debug/jj-2 -R ~/mirrors/linux debug snapshot Time (mean ± σ): 242.3 ms ± 3.3 ms [User: 656.8 ms, System: 538.0 ms] Range (min … max): 236.9 ms … 252.3 ms 30 runs Relative speed comparison 1.40 ± 0.06 target/release-with-debug/jj-1 -R ~/mirrors/linux debug snapshot 1.27 ± 0.03 target/release-with-debug/jj-2 -R ~/mirrors/linux debug snapshot ``` nixpkgs (~45k files, ~31k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 3 --runs 30 .. Benchmark 1: target/release-with-debug/jj-1 -R ~/mirrors/nixpkgs debug snapshot Time (mean ± σ): 201.0 ms ± 8.5 ms [User: 929.3 ms, System: 917.6 ms] Range (min … max): 170.3 ms … 218.5 ms 30 runs Benchmark 2: target/release-with-debug/jj-2 -R ~/mirrors/nixpkgs debug snapshot Time (mean ± σ): 190.7 ms ± 4.1 ms [User: 859.3 ms, System: 881.1 ms] Range (min … max): 184.6 ms … 202.4 ms 30 runs Relative speed comparison 1.24 ± 0.06 target/release-with-debug/jj-1 -R ~/mirrors/nixpkgs debug snapshot 1.18 ± 0.03 target/release-with-debug/jj-2 -R ~/mirrors/nixpkgs debug snapshot ``` git (~4.5k files, 0.2k dirs) ``` % JJ_CONFIG=/dev/null hyperfine --sort command --warmup 30 --runs 50 .. Benchmark 1: target/release-with-debug/jj-1 -R ~/mirrors/git debug snapshot Time (mean ± σ): 30.3 ms ± 1.1 ms [User: 40.5 ms, System: 39.4 ms] Range (min … max): 28.3 ms … 35.7 ms 50 runs Benchmark 2: target/release-with-debug/jj-2 -R ~/mirrors/git debug snapshot Time (mean ± σ): 30.6 ms ± 1.1 ms [User: 33.8 ms, System: 39.0 ms] Range (min … max): 29.0 ms … 35.0 ms 50 runs Relative speed comparison 1.05 ± 0.05 target/release-with-debug/jj-1 -R ~/mirrors/git debug snapshot 1.06 ± 0.05 target/release-with-debug/jj-2 -R ~/mirrors/git debug snapshot ``` - CPU: 8-core AMD Ryzen 7 PRO 4750U with Radeon Graphics (-MT MCP-) - speed/min/max: 1600/1400/1700 MHz Kernel: 6.11.10-amd64 x86_64 - Filesystem: ext4	2024-12-10 10:51:04 +09:00