ci(pnpr): benchmark the install accelerator (new Bencher pnpr testbed)#12154
Merged
Conversation
…elerator Measures the pnpr-accelerated install path end to end. A new `pnpr@<rev>` target in the integrated-benchmark orchestrator builds both the `pacquet` client and the `pnpr` server from the revision's monorepo clone, boots a per-target pnpr server with an isolated `--storage`, and points the client at it via `PNPR_SERVER`. Reusing the existing multi-target hyperfine model gives both comparisons: - `pnpr@HEAD pacquet@HEAD` -> pnpr-vs-direct ratio in one run (same client, with and without the accelerator). - `pnpr@HEAD pnpr@main` -> regression delta tracked in a new Bencher `pnpr` testbed. Two CI workflows mirror the fork-safe two-stage pacquet pattern, triggered on pnpr/**, pacquet/crates/pnpr-client/**, and pacquet/crates/config/** (the pnprServer plumbing), running the hot-cache/hot-store restore and fresh-install scenarios that model a warm long-running server.
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: 📒 Files selected for processing (4)
🚧 Files skipped from review as they are similar to previous changes (4)
📜 Recent review details⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (10)
📝 WalkthroughWalkthroughAdds pnpr as a benchmark target and measures pacquet both directly and via pnpr: CLI parsing, unified repo/build dispatch, pnpr server lifecycle with readiness and env provisioning, conditional install script sourcing, expanded workflow matrix/timeouts, and generation/upload of two Bencher JSONs. ChangesPnpr Install-Accelerator Benchmarking
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
Suggested reviewers
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Contributor
Micro-Benchmark ResultsLinux |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #12154 +/- ##
==========================================
- Coverage 87.61% 87.57% -0.05%
==========================================
Files 268 268
Lines 30712 30761 +49
==========================================
+ Hits 26909 26939 +30
- Misses 3803 3822 +19 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
The pnpr server is built from the pacquet resolver/store/tarball crates,
so any pacquet change can move the pnpr-accelerated numbers as much as the
direct ones. That means the two benchmarks share a trigger surface and
should co-run — so rather than a separate pnpr workflow posting a second
comment on every pacquet PR, measure both in one run.
The pacquet integrated-benchmark workflow now also runs `pnpr@<rev>`
targets in the two hot-cache/hot-store scenarios (a warm long-running
server is pnpr's realistic shape), emits one combined report/comment, and
uploads to two Bencher testbeds: `pacquet` (direct, all scenarios) and
`pnpr` (accelerated, hot scenarios). The trigger gains `pnpr/**`.
Deletes the standalone pnpr-integrated-benchmark{,-comment}.yml added
earlier in this branch.
Run the pnpr targets in the cold-cache/cold-store scenarios too, not just the hot ones. Those scenarios already wipe the client store between iterations while the per-target pnpr server store stays warm, so this measures pnpr's cold-client-vs-warm-server shape — the realistic CI case (empty local store hitting a warm shared server) — alongside the existing hot-client numbers. Both tools now run all four scenarios, so the report tables and both Bencher testbeds (pacquet, pnpr) cover cold and hot. Collapses the two target-list env vars into one and bumps the cold-step timeouts for the extra commands. Table rendering is unchanged.
Review Summary by QodoIntegrate pnpr install-accelerator benchmarking into pacquet workflow
WalkthroughsDescription• Add pnpr@ target kind to benchmark install-accelerator end-to-end - Builds both pacquet client and pnpr server from revision's monorepo - Boots per-target pnpr server with isolated storage, kills on drop - Points client at server via PNPR_SERVER env var from .pnpr-env • Fold pnpr benchmark into existing pacquet integrated-benchmark workflow - One run measures both direct install and accelerated path in same hyperfine - Uploads to two separate Bencher testbeds: pacquet and pnpr - Trigger gains pnpr/** paths alongside pacquet paths • Expand benchmark scenarios to include pnpr targets - Cold-store scenarios wipe client store while server stays warm (realistic CI) - Hot-store scenarios keep client store warm across iterations - Increases timeouts to accommodate additional benchmark commands • Add TCP readiness polling and server lifecycle management - Wait for pnpr server to accept connections before benchmarking - Guards ensure servers killed on drop at end of benchmark run Diagramflowchart LR
A["Target Spec Parser"] -->|"Add pnpr@rev"| B["TargetKind::Pnpr"]
B -->|"build_pnpr"| C["Build pacquet + pnpr"]
C -->|"start_pnpr_servers"| D["Boot per-target server"]
D -->|"TCP poll"| E["Wait for readiness"]
E -->|"write .pnpr-env"| F["install.bash sources env"]
F -->|"PNPR_SERVER set"| G["Client routes through server"]
G -->|"hyperfine runs"| H["Benchmark both tools"]
H -->|"split results"| I["bencher-results.json"]
H -->|"split results"| J["bencher-results-pnpr.json"]
I -->|"upload"| K["pacquet testbed"]
J -->|"upload"| L["pnpr testbed"]
File Changes1. pacquet/tasks/integrated-benchmark/src/cli_args.rs
|
Contributor
There was a problem hiding this comment.
Pull request overview
Adds end-to-end benchmarking for the pnpr install-accelerator path by introducing a new pnpr@<rev> benchmark target that runs the existing pacquet client against a per-target, long-running pnpr server, and publishes results to a separate Bencher pnpr testbed alongside the existing pacquet testbed.
Changes:
- Extend the integrated-benchmark orchestrator with a
pnpr@<rev>target that buildspacquet+pnpr, boots a per-target pnpr server, and injectsPNPR_SERVERvia a generated.pnpr-env. - Update CLI target parsing/tests to accept
pnpr@<rev>and adjust environment validation to treat pnpr like pacquet (needs pacquet repo + cargo). - Update GitHub Actions workflows to co-run
pacquetandpnprbenchmarks, generate two Bencher-shaped result files, and upload to two Bencher testbeds.
Reviewed changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
pacquet/tasks/integrated-benchmark/src/work_env.rs |
Adds pnpr target build + server lifecycle management and wires .pnpr-env sourcing into install.bash. |
pacquet/tasks/integrated-benchmark/src/main.rs |
Ensures pacquet repo/cargo prerequisites are enforced when pnpr targets are present. |
pacquet/tasks/integrated-benchmark/src/cli_args.rs |
Adds TargetKind::Pnpr and updates parsing/error messages. |
pacquet/tasks/integrated-benchmark/src/cli_args/tests.rs |
Adds pnpr target parsing tests and updates error expectations. |
.github/workflows/pacquet-integrated-benchmark.yml |
Runs both pacquet and pnpr targets per scenario, produces two bencher-results JSONs, and uploads both testbeds. |
.github/workflows/pacquet-integrated-benchmark-comment.yml |
Extends PR-side Bencher uploads to include the pnpr testbed from the artifact outputs. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@pacquet/tasks/integrated-benchmark/src/cli_args.rs`:
- Around line 65-69: Update the CLI help text to include the new pnpr target
kind: change the description for CliArgs.targets so it documents that targets
can be "pacquet@<rev>", "pnpm@<rev>" or the new "pnpr@<rev>" (this corresponds
to TargetSpec::from_str now accepting pnpr@<rev>). Locate the help string on the
CliArgs struct (the targets field) and add "pnpr@<rev>" to the examples and any
explanatory text so --help reflects the new supported target kind.
In `@pacquet/tasks/integrated-benchmark/src/work_env.rs`:
- Around line 401-423: Don't use pick_unused_port() (TOCTOU); instead start the
pnpr process asking it to bind an ephemeral port (use "--listen 127.0.0.1:0"
when constructing Command in the Command::new(&binary) block) and then
read/parse the child process output (captured via the stdout/stderr files or a
pipe created when spawning) to obtain the actual bound address/port reported by
pnpr; update the code around pick_unused_port(), the Command::new(&binary)
invocation, and the process handling logic to wait for and extract the reported
listening address from the child's output before proceeding.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro Plus
Run ID: 42598102-70aa-4dd1-bc12-335faa897745
📒 Files selected for processing (6)
.github/workflows/pacquet-integrated-benchmark-comment.yml.github/workflows/pacquet-integrated-benchmark.ymlpacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📜 Review details
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (8)
- GitHub Check: Agent
- GitHub Check: Run benchmark on ubuntu-latest
- GitHub Check: Lint and Test (ubuntu-latest)
- GitHub Check: Lint and Test (windows-latest)
- GitHub Check: Lint and Test (macos-latest)
- GitHub Check: Run benchmark on ubuntu-latest
- GitHub Check: Code Coverage
- GitHub Check: Compile & Lint
🧰 Additional context used
📓 Path-based instructions (1)
pacquet/**/*.rs
📄 CodeRabbit inference engine (pacquet/AGENTS.md)
pacquet/**/*.rs: Log emissions are part of matching pnpm — when porting a function that firespnpm:<channel>events throughglobalLogger,logger.debug(...), orstreamParser.write(...), mirror the call site, payload, and ordering so@pnpm/cli.default-reporterparses pacquet's NDJSON the same way
Declare a newtype wrapper for branded string types instead of collapsing the brand into a plainStringor&strin Rust
If upstream TypeScript always validates before construction of a branded string, validate in the Rust wrapper too viaTryFrom<String>and/orFromStrand do not provide an infallible public constructor
If upstream TypeScript never validates a branded string, just brand for type-safety in Rust by exposing an infallibleFrom<String>constructor
If upstream TypeScript occasionally constructs a branded string without validation, exposefrom_str_uncheckedin Rust as an escape hatch alongside the validating constructor
Match upstream serde behavior for branded strings crossing JSON, YAML, or INI boundaries by using#[serde(try_from = "String")]for deserialization and#[serde(into = "String")]for serialization
Derive simple conversions for branded strings using#[derive(derive_more::From)]and#[derive(derive_more::Into)]instead of handwritingimplblocks; use manualimplonly when conversion needs custom logic
Model TypeScript string literal unions (like'auto' | 'always' | 'never') as Rustenums instead of newtype wrappers, since the set of valid values is closed
Treat TypeScript string template literal types (like`${string}@${string}`) the same as branded string types in Rust, using a newtype wrapper with validation
Follow the code style guide inCODE_STYLE_GUIDE.md— imports, modules, naming, ownership and borrowing, parameter type selection, trait bounds, pattern matching,pipe-trait, error handling, test layout, and cloning ofArcandRc
Choose owned vs. borrowed parameters to minimize copies; widen to t...
Files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs
🧠 Learnings (41)
📓 Common learnings
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-05-25T12:36:42.202Z
Learning: User-visible changes (CLI flags, defaults, environment variables, lockfile/manifest/state-file formats, error codes/messages, log emissions, store layout, hook semantics) in pnpm must be mirrored to pacquet in the same PR
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11784
File: pacquet/crates/resolving-deps-resolver/src/hoist_peers.rs:120-133
Timestamp: 2026-05-20T23:08:06.093Z
Learning: Pacquet (pnpm's Rust port) has a cardinal rule: "match pnpm exactly — do not fix pnpm quirks unless the same fix has landed in pnpm first." Review comments should not suggest behavioral deviations from upstream pnpm, even when the upstream behavior appears buggy. If a real bug is identified, it must be fixed upstream first.
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pnpr/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:24.760Z
Learning: Use Conventional Commits with 'pnpr' as the scope in commit messages (e.g., feat(pnpr): ..., fix(pnpr): ...)
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11915
File: pacquet/crates/resolving-deps-resolver/src/resolve_dependency_tree.rs:553-617
Timestamp: 2026-05-24T21:11:04.272Z
Learning: In the pacquet Rust port (pnpm/pnpm repo), the `ResolvedPackage.optional` AND-folding on revisit intentionally mirrors pnpm's `resolveDependencies.ts:1627-1648` behavior: only the directly-revisited package's `optional` flag is updated; transitive descendants are not re-walked. pnpm CLI corrects stale optional flags downstream via `copyDependencySubGraph` BFS in `lockfile/pruner/src/index.ts:160-205`, which tracks a `nonOptional` set and re-stamps any package reachable by an all-non-optional path. Pacquet does not yet have this pruner equivalent, so the stale flags flow directly through `dependencies_graph_to_lockfile.rs:409` → `create_virtual_store.rs:762` → `installability.rs:394`. A follow-up to port `copyDependencySubGraph` is planned; until then, do not flag the resolver-layer optional propagation gap as a bug in pacquet PRs — it is intentional parity with pnpm's resolver layer.
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Reference the upstream pnpm commit/PR when porting code from pnpm in commit messages
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Use snapshot tests with `insta` and carefully review diffs when intentional changes alter snapshots; accept with `cargo insta review` only after careful review
Learnt from: KSXGitHub
Repo: pnpm/pnpm PR: 12093
File: pacquet/crates/cli/src/cli_args/run/recursive.rs:290-315
Timestamp: 2026-06-01T08:59:42.161Z
Learning: In pacquet's recursive run implementation (`pacquet/crates/cli/src/cli_args/run/recursive.rs`), the `pnpm-exec-summary.json` format for failed package entries correctly includes `prefix` and `message` fields in addition to `status` and `duration`. This matches pnpm's `ActionFailure` variant in `cli/utils/src/recursiveSummary.ts` and the direct serialization in `exec/commands/src/exec.ts`. There is no `ExecutionStatusInSummary` type in pnpm. The only intentional divergence is omitting the JS `error` field, whose `JSON.stringify` output is non-deterministic due to non-enumerable `Error` properties.
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Port relevant pnpm tests to Rust tests whenever they translate when porting behavior from pnpm
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Tests that need the mocked registry should start `pnpr` through `pacquet-testing-utils`; `cargo test` / `cargo nextest run` should not require a separate `just registry-mock launch` step
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pnpr/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:24.760Z
Learning: Applies to pnpr/**/pnpr/**/*.rs : Follow the pacquet contributing guide (../pacquet/CONTRIBUTING.md) for test layout and Rust conventions
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Tests that need the mocked registry should start `pnpr` through `pacquet-testing-utils`; `cargo test` / `cargo nextest run` should not require a separate `just registry-mock launch` step
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Prefer `#[cfg_attr(target_os = "windows", ignore = "...")]` (or matching `#[cfg(unix)]` gates) over runtime probe-and-skip helpers for platform-locked tools
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-24T21:11:04.272Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11915
File: pacquet/crates/resolving-deps-resolver/src/resolve_dependency_tree.rs:553-617
Timestamp: 2026-05-24T21:11:04.272Z
Learning: In the pacquet Rust port (pnpm/pnpm repo), the `ResolvedPackage.optional` AND-folding on revisit intentionally mirrors pnpm's `resolveDependencies.ts:1627-1648` behavior: only the directly-revisited package's `optional` flag is updated; transitive descendants are not re-walked. pnpm CLI corrects stale optional flags downstream via `copyDependencySubGraph` BFS in `lockfile/pruner/src/index.ts:160-205`, which tracks a `nonOptional` set and re-stamps any package reachable by an all-non-optional path. Pacquet does not yet have this pruner equivalent, so the stale flags flow directly through `dependencies_graph_to_lockfile.rs:409` → `create_virtual_store.rs:762` → `installability.rs:394`. A follow-up to port `copyDependencySubGraph` is planned; until then, do not flag the resolver-layer optional propagation gap as a bug in pacquet PRs — it is intentional parity with pnpm's resolver layer.
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Use snapshot tests with `insta` and carefully review diffs when intentional changes alter snapshots; accept with `cargo insta review` only after careful review
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:24.760Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pnpr/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:24.760Z
Learning: Prefer existing pacquet-* crates over writing new code; check pacquet-tarball, pacquet-crypto-hash, pacquet-crypto-shasums-file, pacquet-package-manifest, pacquet-network, pacquet-registry, pacquet-fs, and pacquet-diagnostics before implementing non-trivial functionality
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Tests must not be tolerant of missing build/runtime environment by silently returning early when a tool isn't found — let the test panic when a required tool is absent
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Port relevant pnpm tests to Rust tests whenever they translate when porting behavior from pnpm
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/*.rs : Search for existing shared helpers before writing new code; shared helpers tend to live in `crates/fs`, `crates/testing-utils`, and `crates/diagnostics`
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:24.760Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pnpr/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:24.760Z
Learning: Applies to pnpr/**/pnpr/**/*.rs : Follow the pacquet contributing guide (../pacquet/CONTRIBUTING.md) for test layout and Rust conventions
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:24.760Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pnpr/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:24.760Z
Learning: Applies to pnpr/**/pnpr/**/*.rs : Follow the pacquet code-style guide (../pacquet/CODE_STYLE_GUIDE.md) for Rust-level conventions including imports, naming, ownership, and error handling
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/Cargo.toml : Check whether the workspace already depends on something suitable in `[workspace.dependencies]` in the root `Cargo.toml` before adding a new dependency
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rs
📚 Learning: 2026-05-20T20:41:30.632Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11773
File: pacquet/crates/resolving-tarball-resolver/src/tarball_resolver.rs:115-117
Timestamp: 2026-05-20T20:41:30.632Z
Learning: In the pacquet Rust port of pnpm, the `is_http_url` helper in `pacquet/crates/resolving-tarball-resolver/src/tarball_resolver.rs` intentionally uses `bare.starts_with("http:") || bare.starts_with("https:")` (not `"http://"` / `"https://"`) to match upstream pnpm's `startsWith('http:')` / `startsWith('https:')` check byte-for-byte. Pacquet's cardinal rule (pacquet/AGENTS.md) requires matching pnpm even on quirks; malformed non-URL inputs are rejected downstream by `reqwest::Url::parse` as a `ResolveError`.
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rs
📚 Learning: 2026-06-02T13:18:26.437Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 12134
File: pacquet/crates/resolving-npm-resolver/src/create_npm_resolution_verifier.rs:311-325
Timestamp: 2026-06-02T13:18:26.437Z
Learning: In pacquet's lockfile resolution verifier (`pacquet/crates/resolving-npm-resolver/src/create_npm_resolution_verifier.rs`), URL-keyed tarball dependencies do NOT need a separate `non_semver_version` field in `VerifyCtx`. Unlike the TypeScript side (which derives `version` from `snapshot.version` and threads `nonSemverVersion` separately), pacquet's `collect_candidates` takes `version` from the lockfile key suffix. For a URL-keyed dep the key is `name@<url>`, so `ctx.version` is the URL string, which fails `node_semver::Version::parse(ctx.version)` and the existing guard `if node_semver::Version::parse(ctx.version).is_err() { return ResolutionVerification::Ok; }` already skips the registry lookup correctly. Adding a `non_semver_version` field to `VerifyCtx` for this purpose would be inert.
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rs
📚 Learning: 2026-05-24T21:11:04.272Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11915
File: pacquet/crates/resolving-deps-resolver/src/resolve_dependency_tree.rs:553-617
Timestamp: 2026-05-24T21:11:04.272Z
Learning: In pacquet (pnpm/pnpm repo), `ResolvedPackage.optional` AND-folding intentionally mirrors pnpm's resolveDependencies.ts:1627-1648 revisit behavior: only the directly-visited package's `optional` flag is updated on revisit, not transitive descendants. pnpm CLI corrects stale optional flags via `copyDependencySubGraph` BFS in `lockfile/pruner/src/index.ts:160-205`. Pacquet does not yet have this pruner equivalent, so raw `node.optional` flows directly into snapshot/virtual-store via `dependencies_graph_to_lockfile.rs:409` → `create_virtual_store.rs:762` → `installability.rs:394`. A follow-up issue to port `copyDependencySubGraph` is planned.
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rs
📚 Learning: 2026-05-25T14:58:11.105Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11931
File: pacquet/crates/resolving-npm-resolver/src/create_npm_resolution_verifier.rs:560-589
Timestamp: 2026-05-25T14:58:11.105Z
Learning: In `pacquet/crates/resolving-npm-resolver/src/create_npm_resolution_verifier.rs`, all per-`(registry, name[, version])` caches in `NpmResolutionVerifier` (`published_at`, `full_meta`, `full_meta_for_trust`, `abbreviated_meta`, `local_meta`) intentionally use the same pattern: lock → miss-check → release lock → await fetch/load → re-acquire lock → insert. This uniform pattern is deliberate; do not flag individual caches for using it. The known follow-up improvement (replacing the pattern with `tokio::sync::OnceCell` per key inside a `Mutex<HashMap<…>>`) is tracked as a future structural change to cover all five caches simultaneously.
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rs
📚 Learning: 2026-05-20T23:07:58.444Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11784
File: pacquet/crates/resolving-deps-resolver/src/hoist_peers.rs:120-133
Timestamp: 2026-05-20T23:07:58.444Z
Learning: When reviewing code in this pacquet Rust port, follow the upstream pnpm compatibility rule: only match pnpm’s behavior exactly. Do not propose review changes that intentionally deviate from pnpm’s documented/observed behavior, even if pnpm appears buggy. If you identify a real bug in pnpm behavior, the review should prioritize fixing it upstream in pnpm first, and avoid implementing a pnpm-behavior workaround here unless the same fix has already landed upstream.
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: The `pre-push` hook runs `pacquet/scripts/pre-push-rust.sh` which checks `rustfmt`, `taplo`, `cargo doc` with `-D warnings`, and `cargo dylint`
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Match how the same feature is implemented in the TypeScript pnpm CLI — any change in pacquet must match pnpm's behavior, logic, edge cases, config resolution, error messages, file/lockfile formats, and existing tests
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-20T20:41:50.322Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11773
File: pacquet/crates/package-manager/src/install_package_from_registry.rs:111-114
Timestamp: 2026-05-20T20:41:50.322Z
Learning: In pacquet (pnpm/pnpm repo, Rust codebase), `install_package_from_registry` is the npm-only install path. The npm resolver always stamps `ResolveResult.id` (a `PkgResolutionId`) as `nameversion`. Parsing it back through `PkgNameVer` with `.expect()` is intentional — a parse failure means a mis-dispatch bug, not malformed external input. Per pacquet's CLAUDE.md: "Don't add error handling, fallbacks, or validation for scenarios that can't happen. Trust internal code and framework guarantees." Do not suggest replacing such `.expect()` calls with graceful error handling.
Applied to files:
pacquet/tasks/integrated-benchmark/src/main.rspacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/*.rs : Tests are documentation — do not duplicate test scenarios, edge cases, failure modes, or worked examples in prose when they are already captured by tests
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args/tests.rs
📚 Learning: 2026-05-20T21:18:56.391Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11778
File: pacquet/crates/resolving-local-resolver/tests/resolve.rs:365-372
Timestamp: 2026-05-20T21:18:56.391Z
Learning: In `pacquet/crates/resolving-local-resolver/tests/resolve.rs`, the test `fail_when_resolving_from_not_existing_directory_an_injected_dependency` intentionally uses `injected: false`. The test is a verbatim port of the upstream pnpm TypeScript test (resolving/local-resolver/test/index.ts at ef87f3ccff). The `injected` flag only affects the file/link protocol choice for plain directory paths; when the `file:` scheme is explicit in the bare specifier, the flag has no effect on the resolution code path. The misleading test name is inherited from upstream.
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args/tests.rs
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Consult `plans/TEST_PORTING.md` before adding ported tests and update its checkboxes as items land; use `known_failures` modules and `pacquet_testing_utils::allow_known_failure!` at the not-yet-implemented boundary
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args/tests.rs
📚 Learning: 2026-05-29T18:03:24.760Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pnpr/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:24.760Z
Learning: Applies to pnpr/**/pnpr/crates/**/Cargo.toml : Cargo.toml files for new registry-only crates must use the pnpr- prefix for the package name
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-20T23:23:43.636Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11788
File: pacquet/crates/lockfile/src/snapshot_dep_ref.rs:52-56
Timestamp: 2026-05-20T23:23:43.636Z
Learning: In the pacquet lockfile crate, both `SnapshotDepRef::Link` and `ImporterDepVersion::Link` intentionally use raw `Link(String)` (not a newtype) for the link target. Upstream pnpm stores `link:` targets verbatim without validation (`refToRelative` only checks for the literal `link:` prefix). A `LinkTarget` newtype should not be suggested for either variant; if branding ever makes sense it should be applied to both sites together in a dedicated refactor PR, not in a bug-fix PR.
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-29T18:03:24.760Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pnpr/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:24.760Z
Learning: Applies to pnpr/**/pnpr/crates/**/Cargo.toml : New registry-only crates must be placed under pnpr/crates/<short-name>/ and named pnpr-<short-name> in Cargo.toml, never using the pacquet- prefix
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args/tests.rspacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Use `bench` as a commit type for benchmark-only changes in addition to standard Conventional Commit types
Applied to files:
.github/workflows/pacquet-integrated-benchmark-comment.yml.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-25T12:36:42.202Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: AGENTS.md:0-0
Timestamp: 2026-05-25T12:36:42.202Z
Learning: User-visible changes (CLI flags, defaults, environment variables, lockfile/manifest/state-file formats, error codes/messages, log emissions, store layout, hook semantics) in pnpm must be mirrored to pacquet in the same PR
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/*.rs : Declare a newtype wrapper for branded string types instead of collapsing the brand into a plain `String` or `&str` in Rust
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-06-01T08:59:42.161Z
Learnt from: KSXGitHub
Repo: pnpm/pnpm PR: 12093
File: pacquet/crates/cli/src/cli_args/run/recursive.rs:290-315
Timestamp: 2026-06-01T08:59:42.161Z
Learning: In pacquet's recursive run implementation (`pacquet/crates/cli/src/cli_args/run/recursive.rs`), the `pnpm-exec-summary.json` format for failed package entries correctly includes `prefix` and `message` fields in addition to `status` and `duration`. This matches pnpm's `ActionFailure` variant in `cli/utils/src/recursiveSummary.ts` and the direct serialization in `exec/commands/src/exec.ts`. There is no `ExecutionStatusInSummary` type in pnpm. The only intentional divergence is omitting the JS `error` field, whose `JSON.stringify` output is non-deterministic due to non-enumerable `Error` properties.
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args.rs
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/*.rs : Follow Rust API Guidelines for naming
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-20T10:06:55.749Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11760
File: pacquet/crates/resolving-deps-resolver/src/resolved_tree.rs:15-18
Timestamp: 2026-05-20T10:06:55.749Z
Learning: In `pacquet/crates/resolving-deps-resolver/src/resolved_tree.rs`, the `DirectDep.id`, `ResolvedPackage.id`, and `ResolvedTree.packages` HashMap keys are intentionally plain `String` for now. The natural branded type would be `pacquet_lockfile::PkgNameVer`, but it cannot be used as a HashMap key because `node_semver::Version` does not derive `Hash`. The upstream parity type is `PkgResolutionId` (which carries an optional peer-dep suffix), and the branded type should be introduced alongside peer-dep resolution and lockfile generation work to avoid locking the seam too early.
Applied to files:
pacquet/tasks/integrated-benchmark/src/cli_args.rspacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-29T18:03:24.760Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pnpr/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:24.760Z
Learning: Applies to pnpr/**/pnpr/**/Cargo.toml : Declare new shared dependencies in the root [workspace.dependencies] and use { workspace = true } in pnpr crate's Cargo.toml
Applied to files:
pacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-06-02T14:39:19.809Z
Learnt from: KSXGitHub
Repo: pnpm/pnpm PR: 11938
File: pacquet/crates/config/src/lib.rs:959-966
Timestamp: 2026-06-02T14:39:19.809Z
Learning: In the pnpm/pnpm pacquet Rust port (`pacquet/crates/config/src/lib.rs`), `Config.extra_bin_paths` is intentionally left empty (`Vec::new()`) until workspace fan-out support for `run`/`exec` is implemented. It mirrors pnpm's `Config.extraBinPaths` (the workspace-root `node_modules/.bin`), which is also empty outside a workspace. Populating it before the workspace-root is resolved would put a non-existent path on `PATH`, so it should only be derived once workspace support lands. Do not flag this as a bug or missing derivation in reviews.
Applied to files:
pacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/tests/**/*.rs : Prefer real fixtures over dependency-injection seams — use `tempfile::TempDir`, the mocked registry, or integration tests spawning the actual binary for happy paths and most error paths; use the DI seam only for filesystem error kinds, deterministic time, shared process-global state, or external-service happy paths
Applied to files:
pacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-06-02T14:39:46.617Z
Learnt from: KSXGitHub
Repo: pnpm/pnpm PR: 11938
File: pacquet/crates/cli/src/cli_args/dlx.rs:281-288
Timestamp: 2026-06-02T14:39:46.617Z
Learning: In pacquet's `pacquet/crates/cli/src/cli_args/dlx.rs` (and the analogous exec path), the `shell_mode` branch intentionally uses `joined.join(" ")` (no per-token shell-escaping) when building the command string passed to the platform shell. This matches pnpm's observable behavior: pnpm passes the command to execa with `shell: true`, and execa explicitly does NOT escape arguments in shell mode — it joins them and delegates entirely to the shell. Applying any shell-escaping here would diverge from pnpm's behavior and violates the pnpm-compatibility rule for the pacquet Rust port.
Applied to files:
pacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-23T09:14:43.635Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11867
File: pacquet/crates/package-manager/src/install_with_fresh_lockfile.rs:726-730
Timestamp: 2026-05-23T09:14:43.635Z
Learning: In `pacquet/crates/package-manager/src/install_with_fresh_lockfile.rs`, the fresh-lockfile path intentionally does not run lifecycle scripts (`BuildModules` is not invoked, and `side_effects_maps_by_snapshot` from `CreateVirtualStoreOutput` is discarded). This is pre-existing behavior documented by an in-code comment mirroring upstream `link.ts:167-170`. The frozen-lockfile path wires `BuildModules` end-to-end as normal. Wiring lifecycle scripts into the fresh-lockfile path is tracked as future work, separate from this file's concern.
Applied to files:
pacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-23T09:14:43.635Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11867
File: pacquet/crates/package-manager/src/install_with_fresh_lockfile.rs:726-730
Timestamp: 2026-05-23T09:14:43.635Z
Learning: In `pacquet/crates/package-manager/src/install_with_fresh_lockfile.rs`, the fresh-lockfile path intentionally does not invoke `BuildModules` and discards `side_effects_maps_by_snapshot` from `CreateVirtualStoreOutput`. This is pre-existing, documented behavior (mirroring upstream `link.ts:167-170`): `importing_done` fires once extraction and symlink linking are complete, and the fresh-lockfile path does not run lifecycle scripts. The frozen-lockfile path wires `BuildModules` end-to-end as normal. Do not flag this omission as a bug; wiring lifecycle scripts into the fresh-lockfile path is tracked as future work separate from perf refactors.
Applied to files:
pacquet/tasks/integrated-benchmark/src/work_env.rs
📚 Learning: 2026-05-04T17:01:30.322Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11452
File: exec/commands/src/dlx.ts:298-305
Timestamp: 2026-05-04T17:01:30.322Z
Learning: In `exec/commands/src/dlx.ts`, when `promptApproveDlxBuilds` exits early (no `approve-builds` command or non-interactive stdin), the `cachedDir` is intentionally still linked into the shared dlx cache even if some build scripts were skipped. This is by design: (1) it mirrors `pnpm add -g` behavior which also commits the install with skipped builds in non-interactive mode; (2) the cache key already encodes `allowBuild`, so users can recover by re-invoking with `--allow-build=<pkg>` which generates a fresh cache entry; (3) re-installing on every non-interactive invocation would be a perf regression for CI. Do not suggest preventing the symlink in this code path.
Applied to files:
pacquet/tasks/integrated-benchmark/src/work_env.rs.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-29T18:03:15.354Z
Learnt from: CR
Repo: pnpm/pnpm PR: 0
File: pacquet/AGENTS.md:0-0
Timestamp: 2026-05-29T18:03:15.354Z
Learning: Applies to pacquet/**/*.rs : Follow the code style guide in `CODE_STYLE_GUIDE.md` — imports, modules, naming, ownership and borrowing, parameter type selection, trait bounds, pattern matching, `pipe-trait`, error handling, test layout, and cloning of `Arc` and `Rc`
Applied to files:
.github/workflows/pacquet-integrated-benchmark.yml
📚 Learning: 2026-05-24T08:18:06.019Z
Learnt from: zkochan
Repo: pnpm/pnpm PR: 11895
File: pnpm/test/deploy.ts:91-95
Timestamp: 2026-05-24T08:18:06.019Z
Learning: In the pnpm/pnpm repository, integration tests that hit the real `registry.npmjs.org` (e.g., for `pacquet` or `pnpm/pacquet`) do NOT use a runtime env-var gate (such as `PNPM_RUN_PUBLIC_REGISTRY_TESTS`). They simply pass `--config.registry=https://registry.npmjs.org/` directly to `execPnpm` and set a higher timeout. This is the established pattern, as seen in `pnpm/test/install/pacquet.ts` and `pnpm/test/deploy.ts`. Do not suggest adding env-var guards for these tests.
Applied to files:
.github/workflows/pacquet-integrated-benchmark.yml
Contributor
Integrated-Benchmark Report (Linux)Each scenario has pacquet rows (direct install) and pnpr rows (the same client through the pnpr install accelerator), so pnpr@HEAD vs pacquet@HEAD is the pnpr-vs-direct ratio. Cold-store scenarios wipe the client store between runs (warm server); hot-store scenarios keep it warm. The pacquet@HEAD rows feed the pacquet Bencher testbed; the pnpr@HEAD rows feed the pnpr testbed. Scenario: Isolated linker: fresh restore, cold cache + cold store
BENCHMARK_REPORT.json{
"results": [
{
"command": "pacquet@HEAD",
"mean": 2.0861554052000004,
"stddev": 0.12925520876560043,
"median": 2.0288604063,
"user": 2.5435962799999996,
"system": 3.2980932600000004,
"min": 1.9869582653000002,
"max": 2.3732088473000004,
"times": [
1.9884014223000002,
2.0077769633,
2.0291247163,
2.2558754073,
2.0496852883,
1.9869582653000002,
2.3732088473000004,
2.1222533843,
2.0285960963000003,
2.0196736613
]
},
{
"command": "pacquet@main",
"mean": 2.0299339596999997,
"stddev": 0.06708796402913302,
"median": 2.0337242308,
"user": 2.5406828799999994,
"system": 3.2817418600000003,
"min": 1.9059394443000002,
"max": 2.1073763143,
"times": [
2.0036902093,
2.0933590923,
2.1073763143,
1.9962874723000001,
2.0697342233000002,
2.0637582523,
1.9059394443000002,
1.9688778913,
1.9887133453000003,
2.1016033523
]
},
{
"command": "pnpr@HEAD",
"mean": 2.0175024352,
"stddev": 0.07791484682321347,
"median": 1.9981918093000002,
"user": 2.54448208,
"system": 3.29467826,
"min": 1.9327927433000003,
"max": 2.1881747513,
"times": [
1.9927742313000003,
2.0036093873,
1.9716969163000002,
1.9454365373000002,
1.9327927433000003,
2.0225519873,
2.1881747513,
2.0477413583,
2.1009818723,
1.9692645673000002
]
},
{
"command": "pnpr@main",
"mean": 2.0019971756,
"stddev": 0.04967089924243278,
"median": 1.9855038288000002,
"user": 2.5615398799999993,
"system": 3.30721146,
"min": 1.9562590933000001,
"max": 2.0986291153,
"times": [
1.9996319203000001,
1.9653064303,
2.0986291153,
1.9713757373000003,
2.0062494893,
1.9562590933000001,
2.0239286703,
2.0723461473,
1.9658687593000002,
1.9603763933000002
]
}
]
}Scenario: Isolated linker: fresh restore, hot cache + hot store
BENCHMARK_REPORT.json{
"results": [
{
"command": "pacquet@HEAD",
"mean": 0.6369365807000001,
"stddev": 0.015778823600057047,
"median": 0.6330426294,
"user": 0.3504649,
"system": 1.2996630200000001,
"min": 0.6215896604000001,
"max": 0.6798281994,
"times": [
0.6798281994,
0.6304626444,
0.6358275044,
0.6215896604000001,
0.6292434174,
0.6321474074,
0.6339378514,
0.6315902054,
0.6362950204000001,
0.6384438964
]
},
{
"command": "pacquet@main",
"mean": 0.6511617211999999,
"stddev": 0.04437032991311834,
"median": 0.6371889694,
"user": 0.35265040000000003,
"system": 1.31253262,
"min": 0.6291449804,
"max": 0.7762677334,
"times": [
0.7762677334,
0.6366917694,
0.6390219594000001,
0.6504025434,
0.6317980254000001,
0.6359531544,
0.6425324334,
0.6321184434,
0.6291449804,
0.6376861694
]
},
{
"command": "pnpr@HEAD",
"mean": 0.6875611163,
"stddev": 0.10544614400144374,
"median": 0.6519109519,
"user": 0.3539667,
"system": 1.2982522199999997,
"min": 0.6304772494,
"max": 0.9789412694,
"times": [
0.7126744344,
0.6526992204000001,
0.9789412694,
0.6511226834,
0.6724860324,
0.6719062124,
0.6304772494,
0.6341378294000001,
0.6339330624,
0.6372331694000001
]
},
{
"command": "pnpr@main",
"mean": 0.7211107651,
"stddev": 0.05781711294641612,
"median": 0.7234326839,
"user": 0.3466363,
"system": 1.31503682,
"min": 0.6363625644,
"max": 0.8313554144,
"times": [
0.8313554144,
0.6865132874000001,
0.6708674054,
0.7786030404000001,
0.7233144234000001,
0.7235509444,
0.6363625644,
0.6750252954,
0.7247211604,
0.7607941154000001
]
}
]
}Scenario: Isolated linker: fresh install, cold cache + cold store
BENCHMARK_REPORT.json{
"results": [
{
"command": "pacquet@HEAD",
"mean": 2.28988288714,
"stddev": 0.0171159143278725,
"median": 2.28879967814,
"user": 3.6896820200000002,
"system": 3.0997950199999997,
"min": 2.26793653264,
"max": 2.31975152464,
"times": [
2.29143065264,
2.30475407764,
2.26793653264,
2.27583457464,
2.31975152464,
2.30919427064,
2.27383335664,
2.27632606864,
2.28616870364,
2.2935991096399997
]
},
{
"command": "pacquet@main",
"mean": 2.29759498494,
"stddev": 0.025077737403242723,
"median": 2.2960424011400002,
"user": 3.7169062199999994,
"system": 3.084124319999999,
"min": 2.25929663164,
"max": 2.34077993064,
"times": [
2.3256298536399997,
2.29567080264,
2.26677408364,
2.25929663164,
2.28019928864,
2.29385530764,
2.31224750064,
2.34077993064,
2.30508245064,
2.29641399964
]
},
{
"command": "pnpr@HEAD",
"mean": 2.30209557664,
"stddev": 0.04406301469605407,
"median": 2.30031841464,
"user": 3.6995176199999995,
"system": 3.0888319199999996,
"min": 2.23820226864,
"max": 2.36932517464,
"times": [
2.34475898564,
2.32072721364,
2.34697071064,
2.32120339564,
2.36932517464,
2.26477183064,
2.23820226864,
2.27990961564,
2.27267058764,
2.26241598364
]
},
{
"command": "pnpr@main",
"mean": 2.28172563434,
"stddev": 0.01826467606648467,
"median": 2.27929243664,
"user": 3.6906895200000003,
"system": 3.06558522,
"min": 2.26241219264,
"max": 2.31455338664,
"times": [
2.31455338664,
2.28170709164,
2.31029657964,
2.2735464846399998,
2.26241219264,
2.28162662164,
2.26449645864,
2.2876493176399997,
2.27695825164,
2.26400995864
]
}
]
}Scenario: Isolated linker: fresh install, hot cache + hot store
BENCHMARK_REPORT.json{
"results": [
{
"command": "pacquet@HEAD",
"mean": 1.46024397508,
"stddev": 0.024606781803267995,
"median": 1.45258086798,
"user": 1.64208264,
"system": 1.8124288599999996,
"min": 1.43237795048,
"max": 1.50077750148,
"times": [
1.43237795048,
1.47067056848,
1.44689160148,
1.50077750148,
1.4506888254799999,
1.43359873948,
1.45447291048,
1.4992827924799998,
1.44340113048,
1.4702777304799999
]
},
{
"command": "pacquet@main",
"mean": 1.4853076224800001,
"stddev": 0.07435164418902246,
"median": 1.46418768898,
"user": 1.66085654,
"system": 1.8221989600000001,
"min": 1.43280297848,
"max": 1.68818242548,
"times": [
1.44822395748,
1.43280297848,
1.68818242548,
1.4784057234799999,
1.4735941664799999,
1.50344318648,
1.48028647548,
1.44466807148,
1.45478121148,
1.44868802848
]
},
{
"command": "pnpr@HEAD",
"mean": 1.46130182928,
"stddev": 0.038060809838510026,
"median": 1.4450608319799998,
"user": 1.63489854,
"system": 1.8182050599999997,
"min": 1.43214944848,
"max": 1.5438753804799998,
"times": [
1.44878722948,
1.43975124748,
1.4413344344799999,
1.45414370248,
1.5438753804799998,
1.45530786348,
1.43921428448,
1.43214944848,
1.43965488448,
1.5187998174800001
]
},
{
"command": "pnpr@main",
"mean": 1.48244680488,
"stddev": 0.04613213460391995,
"median": 1.46875333548,
"user": 1.6524617400000001,
"system": 1.8402756599999996,
"min": 1.44103347648,
"max": 1.59039193648,
"times": [
1.5285854644799999,
1.49610435148,
1.4679589984799999,
1.59039193648,
1.4448732254799999,
1.46802419048,
1.47268973548,
1.44532418948,
1.44103347648,
1.46948248048
]
}
]
} |
Contributor
|
Bencher Report| Branch | pr/12154 |
| Testbed | pacquet |
Click to view all benchmark results
| Benchmark | Latency | Benchmark Result milliseconds (ms) (Result Δ%) | Upper Boundary milliseconds (ms) (Limit %) |
|---|---|---|---|
| isolated-linker.fresh-install.cold-cache.cold-store | 📈 view plot 🚷 view threshold | 2,289.88 ms(-2.19%)Baseline: 2,341.15 ms | 2,809.38 ms (81.51%) |
| isolated-linker.fresh-install.hot-cache.hot-store | 📈 view plot 🚷 view threshold | 1,460.24 ms(-3.73%)Baseline: 1,516.86 ms | 1,820.23 ms (80.22%) |
| isolated-linker.fresh-restore.cold-cache.cold-store | 📈 view plot 🚷 view threshold | 2,086.16 ms(+1.67%)Baseline: 2,051.91 ms | 2,462.29 ms (84.72%) |
| isolated-linker.fresh-restore.hot-cache.hot-store | 📈 view plot 🚷 view threshold | 636.94 ms(-2.01%)Baseline: 650.03 ms | 780.03 ms (81.66%) |
- work_env: wrap the spawned pnpr child in its PnprServer guard before the readiness wait and .pnpr-env write, so an early panic kills the process on unwind instead of leaking an orphaned server (Copilot). - cli_args: document pnpr@<rev> in the `targets` --help text (CodeRabbit). - workflows: guard each bencher upload on its file existing, so a missing optional results file logs a notice instead of failing the step (Copilot).
This was referenced Jun 3, 2026
Merged
This was referenced Jun 22, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Benchmarks the pnpr install-accelerator path end to end and tracks it in a new Bencher
pnprtestbed, folded into the existing pacquet integrated-benchmark so the two co-run.Orchestrator: a new
pnpr@<rev>targetpacquet/tasks/integrated-benchmarkgains apnpr@<rev>target kind. It builds both thepacquetclient and thepnprserver from the revision's monorepo clone, boots a per-target pnpr server with an isolated--storage, and points the client at it viaPNPR_SERVER(sourced from a generated.pnpr-env). The server is killed on drop; its store is wiped once before the run so the hyperfine warmup primes a warm long-running server.One workflow, two testbeds
The pnpr server is built from the
pacquet-*resolver/store/tarball crates, so a pacquet change can move the pnpr-accelerated numbers as much as the direct ones. They share a trigger surface and should co-run — so rather than a second workflow posting a second comment on every pacquet PR, the existingpacquet-integrated-benchmarkworkflow now measures both in one run:pacquet@<rev>(direct) andpnpr@<rev>(accelerated), sopnpr@HEADvspacquet@HEADis the pnpr-vs-direct ratio in one hyperfine table.pacquetandpnpr— so each keeps its own HEAD-vs-main regression gate.pnpr/**.Cache axes
Both tools run all four scenarios. The cold-store scenarios wipe the client store between iterations while the per-target pnpr server stays warm — i.e. pnpr's cold-client-vs-warm-server shape, the realistic CI case (empty local store hitting a warm shared server). The hot-store scenarios keep the client store warm. A cold-server axis (server restarted per iteration) is a deliberate follow-up — it needs per-iteration server restart machinery.
Validation
cargo check,clippy --deny warnings,cargo fmt --check, and the orchestrator's unit tests pass; the pre-push Rust hook passed.actionlintis at parity with the prior workflow (only the pre-existing intentionalSC2086word-split notes on\$BENCHMARK_TARGETS).Manual step before merge
The Bencher
pnprtestbed must be created in thepnpmBencher project;--testbed pnprdoes not auto-create it.Written by an agent (Claude Code, claude-opus-4-8).
Summary by CodeRabbit
New Features
CI / Tests