perf(encoding): avoid inline bitpacking input copy

[u70b3]

Supersedes #7696. This PR contains the same changes as #7696, but squashed into a single commit with a consolidated commit message.

Summary

• Avoid the unconditional input copy in InlineBitpacking::unchunk by borrowing a typed view over the inline bitpacking chunk words.
• Keep output zero-initialization unchanged; this PR only changes compressed input handling.
• Add focused unchunk roundtrip coverage and a Criterion benchmark comparing the old copy path with the new typed-view path.
• Harden validation by rejecting malformed inline bitpacking chunks up front with clear errors.

Benchmark

Local machine: WSL2 on Intel(R) Core(TM) i7-10700 CPU @ 2.90GHz, 12 logical CPUs.

Command:

cargo bench -p lance-encoding --bench decoder decode_inline_bitpacking_unchunk --features bitpacking -- --noplot

Results:

Case	Path	Criterion time estimate	Throughput
u32_bw12_1024	legacy_copy	285.46 ns	5.0244 GiB/s
u32_bw12_1024	typed_view	247.35 ns	5.7985 GiB/s
u64_bw23_1024	legacy_copy	434.11 ns	6.3331 GiB/s
u64_bw23_1024	typed_view	340.27 ns	8.0797 GiB/s

Path-to-path deltas:

• u32_bw12_1024: typed_view was 13.4% faster than legacy_copy
• u64_bw23_1024: typed_view was 21.6% faster than legacy_copy

The benchmark uses aligned chunks built with LanceBuffer::reinterpret_vec, so it measures the aligned borrow_to_typed_view fast path. Misaligned buffers can still fall back to a copy.

Test Plan

• cargo fmt --all
• cargo test -p lance-encoding --features bitpacking
• cargo clippy -p lance-encoding --all-features --tests --benches -- -D warnings
• cargo clippy --all --tests --benches -- -D warnings
• cargo bench -p lance-encoding --bench decoder decode_inline_bitpacking_unchunk --features bitpacking -- --noplot

Summary by CodeRabbit

• Bug Fixes

  • Improved inline bitpacked data decoding with stronger validation for malformed buffers, invalid value counts, unsupported bit widths, alignment issues, and inconsistent payload sizes.
  • Added clearer corruption errors to help identify invalid encoded data.

• Performance

  • Improved the inline bitpacking decoding path by using typed views, reducing unnecessary data copying.

• Tests

  • Expanded coverage for valid round trips, corruption handling, and decoding performance across multiple data types and bit widths.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: bbe1224b-71d5-465b-bb02-c4f24f4a5672

📥 Commits

Reviewing files that changed from the base of the PR and between 787f759 and d61b50f.

⛔ Files ignored due to path filters (3)

• Cargo.lock is excluded by !**/*.lock
• java/lance-jni/Cargo.lock is excluded by !**/*.lock
• python/Cargo.lock is excluded by !**/*.lock

📒 Files selected for processing (3)

• rust/lance-encoding/Cargo.toml
• rust/lance-encoding/benches/decoder.rs
• rust/lance-encoding/src/encodings/physical/bitpacking.rs

---

📝 Walkthrough

Walkthrough

Inline bitpacking unchunking now uses typed header/payload views with explicit validation and structured corruption errors. Tests cover valid and invalid inputs, while Criterion benchmarks compare legacy copying with typed-view decoding.

Changes

Inline bitpacking decoding

Layer / File(s)	Summary
Typed unchunk validation rust/lance-encoding/Cargo.toml, rust/lance-encoding/src/encodings/physical/bitpacking.rs	Adds the workspace dependency and rewrites inline unchunking around typed views, payload-size checks, bounds validation, and structured corruption errors.
Unchunk round-trip and corruption tests rust/lance-encoding/src/encodings/physical/bitpacking.rs	Adds round-trip coverage for selected bit widths and parameterized checks for malformed headers, alignment, value counts, payload sizes, and bit widths.
Legacy and typed-view benchmarks rust/lance-encoding/benches/decoder.rs	Adds feature-gated benchmarks that generate inline chunks and compare legacy-copy decoding with InlineBitpacking typed-view decoding. Targets are registered on Linux and non-Linux builds.

Estimated code review effort: 4 (Complex) | ~45 minutes

Sequence Diagram(s)

sequenceDiagram
  participant Criterion
  participant ChunkGenerator
  participant LegacyDecoder
  participant InlineBitpacking
  Criterion->>ChunkGenerator: generate inline bitpacked chunk
  Criterion->>LegacyDecoder: decode with legacy_copy
  Criterion->>InlineBitpacking: decode with typed_view
  LegacyDecoder-->>Criterion: return decoded payload
  InlineBitpacking-->>Criterion: return decoded payload
  Criterion->>Criterion: compare decoded buffers and record timings

Loading

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly matches the main change: optimizing inline bitpacking to avoid copying input.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

_{Comment @coderabbitai help to get the list of available commands.}

[codecov]

Codecov Report

❌ Patch coverage is 94.31818% with 5 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...ance-encoding/src/encodings/physical/bitpacking.rs	94.31%	3 Missing and 2 partials ⚠️

📢 Thoughts on this report? Let us know!

[u70b3]

Closed: changes will be force-pushed to the original branch #7696 as a single commit.