BE-394: HashQL: Rework BitMatrix and SparseBitMatrix #8408

indietyp · 2026-02-11T21:25:59Z

🌟 What is the purpose of this PR?

Add benchmarks for BitMatrix and SparseBitMatrix implementations, and moves the implementation into a dedicated module.

Performance

Benchmarked against the previous implementation across sizes 64, 200, and 1000:

Operation	Improvement
Dense insert	At parity (±1%)
Dense contains	5–19% faster
Dense union_rows	1.8–2.0x faster
Dense iter_row	At parity
Sparse insert	6–42% faster
Sparse union_rows	1.5–3.8x faster
Sparse clear+reinsert	2.0–2.3x faster (sizes ≥ 200)

🔍 What does this change?

Adds a new benchmark suite for bit matrix operations
Completely rewrites the BitMatrix and SparseBitMatrix implementations:
- Adds zero-copy row access through RowRef and RowMut view types
- Improves cache locality and vectorization in BitMatrix
- Implements a free-list for SparseBitMatrix to recycle cleared rows
- Adds comprehensive API for bitwise operations (union, subtract, intersect)
- Optimizes transitive closure algorithm
- Adds support for custom allocators
Moves matrix implementations to a dedicated module
Updates dominance frontier code to use the new row view types

🛡 What tests cover this?

Comprehensive unit tests for both BitMatrix and SparseBitMatrix
Property-based tests that verify equivalence between the two implementations
New benchmarks that measure performance across various matrix sizes

Pre-Merge Checklist 🚀

🚢 Has this modified a publishable library?

This PR:

does not modify any publishable blocks or libraries, or modifications do not need publishing

📜 Does this require a change to the docs?

The changes in this PR:

are internal and do not require a docs change

🕸️ Does this require a change to the Turbo Graph?

The changes in this PR:

do not affect the execution graph

vercel · 2026-02-11T21:26:04Z

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
hash	Ready	Preview, Comment	Feb 12, 2026 9:18am

3 Skipped Deployments

Project	Deployment	Actions	Updated (UTC)
hashdotdesign	Ignored	Preview	Feb 12, 2026 9:18am
hashdotdesign-tokens	Ignored	Preview	Feb 12, 2026 9:18am
petrinaut	Skipped		Feb 12, 2026 9:18am

cursor · 2026-02-11T21:26:06Z

PR Summary

Medium Risk
Replaces core bitset matrix data structures used in compiler/graph algorithms; although heavily tested, subtle correctness/perf regressions or memory/allocator edge cases could impact analyses like dominance frontiers and DSE.

Overview
Refactors and replaces the BitMatrix/SparseBitMatrix implementation by moving it into a new id/bit_vec/matrix module, adding zero-copy row view types (RowRef, RowMut) and expanding the API for row-level bitwise ops plus closure helpers.

Changes the sparse representation from per-row DenseBitSet storage to a single arena-backed buffer with an index + free-list, and updates call sites to the new iteration/row APIs (e.g. dominance frontier now stores RowRef, DSE uses iter_row). Adds extensive new unit/property tests for dense+sparse behavior and a new codspeed benchmark (bit_matrix) registered in Cargo.toml.

^{Written by Cursor Bugbot for commit 212c24c. This will update automatically on new commits. Configure here.}

indietyp · 2026-02-11T21:26:33Z

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

This stack of pull requests is managed by Graphite. Learn more about stacking.

augmentcode · 2026-02-11T21:29:16Z

🤖 Augment PR Summary

Summary: This PR reworks the core bit-matrix data structures by moving them into a dedicated module, adding row view types, and introducing a new benchmark suite to measure performance across common sizes.

Changes:

Added a new Criterion/Codspeed benchmark target for dense and sparse bit-matrix operations (benches/bit_matrix.rs).
Introduced id::bit_vec::matrix module containing rewritten BitMatrix and SparseBitMatrix implementations with allocator support.
Added zero-copy row access via RowRef/RowMut, enabling row-level operations without allocating/copying.
Reimplemented SparseBitMatrix as an arena-backed structure with a free-list for recycling cleared rows.
Expanded the API for row-wise bitwise operations (union/subtract/intersect) and (dense) transitive-closure helpers.
Updated dominator frontier code to use the new row view type (RowRef), and updated DSE to use iter_row.
Added comprehensive unit tests plus property-based equivalence tests for dense vs sparse behavior.

Technical Notes: The new dense representation is contiguous (better cache locality/vectorization), while the sparse version reduces per-row allocation overhead by storing all row words in a shared backing buffer.

_{🤖 Was this summary useful? React with 👍 or 👎}

augmentcode

Review completed. 4 suggestions posted.

Comment augment review to trigger a new review at any time.

augmentcode · 2026-02-11T21:29:17Z

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs

+    /// achieving a 64× speedup over the scalar version through bitwise parallelism.
+    pub fn transitive_closure(&mut self) {
+        let size = self.row_domain_size;
+        debug_assert_eq!(size, self.col_domain_size);


transitive_closure only uses debug_assert_eq!(size, self.col_domain_size), so a non-square matrix can yield incorrect behavior in release builds (or panic later via bounds asserts). Consider enforcing the square-matrix precondition non-conditionally (same concern applies to reflexive_transitive_closure).

Severity: medium

Other Locations

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs:707

_{🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.}

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs

augmentcode · 2026-02-11T21:29:17Z

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs

+
+    /// Returns `true` if every bit set in `other` is also set in `row`.
+    #[inline]
+    pub fn superset_row(&self, row: R, other: &DenseBitSet<C>) -> Option<bool> {


superset_row/subset_row don’t enforce that other.domain_size() matches self.col_domain_size (and RowRef::superset_dense only debug_asserts), so mismatched domains can silently produce wrong results in release due to zip truncation. Consider validating domain-size equality in these public methods.

Severity: low

Other Locations

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs:1145

_{🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.}

codecov · 2026-02-11T21:30:20Z

Codecov Report

❌ Patch coverage is 90.63604% with 106 lines in your changes missing coverage. Please review.
✅ Project coverage is 67.94%. Comparing base (e20307f) to head (212c24c).

Files with missing lines	Patch %	Lines
...bs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs	80.98%	102 Missing and 2 partials ⚠️
...l/core/src/graph/algorithms/dominators/frontier.rs	33.33%	2 Missing ⚠️

Additional details and impacted files

@@                                           Coverage Diff                                            @@
##           bm/be-372-hashql-do-not-consider-traversals-during-liveness-analysis    #8408      +/-   ##
========================================================================================================
+ Coverage                                                                 67.66%   67.94%   +0.28%     
========================================================================================================
  Files                                                                       831      833       +2     
  Lines                                                                     75367    76136     +769     
  Branches                                                                   3940     3951      +11     
========================================================================================================
+ Hits                                                                      50995    51731     +736     
- Misses                                                                    23811    23848      +37     
+ Partials                                                                    561      557       -4

Flag	Coverage Δ
apps.hash-ai-worker-ts	`1.41% <ø> (ø)`
apps.hash-api	`0.00% <ø> (ø)`
local.hash-graph-sdk	`6.66% <ø> (ø)`
local.hash-isomorphic-utils	`0.00% <ø> (ø)`
rust.hash-graph-api	`2.88% <ø> (ø)`
rust.hashql-ast	`87.25% <ø> (ø)`
rust.hashql-compiletest	`29.69% <ø> (ø)`
rust.hashql-core	`82.17% <90.62%> (+0.47%)`	⬆️
rust.hashql-eval	`69.13% <ø> (ø)`
rust.hashql-hir	`89.11% <ø> (ø)`
rust.hashql-mir	`90.88% <100.00%> (ø)`
rust.hashql-syntax-jexpr	`94.05% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs

codspeed-hq · 2026-02-11T21:32:37Z

Merging this PR will improve performance by 35.34%

⚡ 6 improved benchmarks
✅ 35 untouched benchmarks
🆕 24 new benchmarks
🗄️ 12 archived benchmarks run¹

Performance Changes

	Benchmark	`BASE`	`HEAD`	Efficiency
⚡	`complex`	18 µs	16.2 µs	+11.32%
⚡	`dead stores`	11.5 µs	9.4 µs	+22.63%
⚡	`diamond`	7.3 µs	5.4 µs	+35.34%
⚡	`complex`	8 µs	7 µs	+13.79%
⚡	`linear`	6.7 µs	5.1 µs	+32.27%
⚡	`diamond`	12.2 µs	10.8 µs	+13.29%
🆕	`bit_matrix/dense/contains[1000]`	N/A	4.5 ms	N/A
🆕	`bit_matrix/dense/contains[200]`	N/A	181.4 µs	N/A
🆕	`bit_matrix/dense/iter_row[1000]`	N/A	495.3 ns	N/A
🆕	`bit_matrix/dense/insert[1000]`	N/A	659.8 µs	N/A
🆕	`bit_matrix/dense/contains[64]`	N/A	18.8 µs	N/A
🆕	`bit_matrix/dense/insert[200]`	N/A	29.4 µs	N/A
🆕	`bit_matrix/dense/iter_row[200]`	N/A	213.9 ns	N/A
🆕	`bit_matrix/dense/union_rows[1000]`	N/A	102.4 µs	N/A
🆕	`bit_matrix/dense/insert[64]`	N/A	4.2 µs	N/A
🆕	`bit_matrix/dense/transitive_closure[64]`	N/A	67.8 µs	N/A
🆕	`bit_matrix/dense/iter_row[64]`	N/A	139.7 ns	N/A
🆕	`bit_matrix/dense/transitive_closure[200]`	N/A	862 µs	N/A
🆕	`bit_matrix/dense/union_rows[64]`	N/A	2.3 µs	N/A
🆕	`bit_matrix/dense/union_rows[200]`	N/A	9 µs	N/A
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

_{Comparing bm/be-394-hashql-rework-bitmatrix (212c24c) with bm/be-372-hashql-do-not-consider-traversals-during-liveness-analysis (e20307f)}

12 benchmarks were run, but are now archived. If they were deleted in another branch, consider rebasing to remove them from the report. Instead if they were added back, click here to restore them. ↩

cursor

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs

libs/@local/hashql/core/src/graph/algorithms/dominators/frontier.rs

github-actions · 2026-02-12T10:26:54Z

Benchmark results

@rust/hash-graph-benches – Integrations

policy_resolution_large

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 2002	$$26.9 \mathrm{ms} \pm 205 \mathrm{μs}\left({\color{gray}-0.371 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$3.25 \mathrm{ms} \pm 15.6 \mathrm{μs}\left({\color{gray}0.040 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 1001	$$12.1 \mathrm{ms} \pm 95.3 \mathrm{μs}\left({\color{gray}-2.036 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 3314	$$42.5 \mathrm{ms} \pm 333 \mathrm{μs}\left({\color{gray}0.790 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$13.9 \mathrm{ms} \pm 106 \mathrm{μs}\left({\color{gray}0.064 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 1526	$$23.2 \mathrm{ms} \pm 171 \mathrm{μs}\left({\color{gray}-1.748 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 2078	$$27.7 \mathrm{ms} \pm 181 \mathrm{μs}\left({\color{gray}0.943 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$3.53 \mathrm{ms} \pm 16.7 \mathrm{μs}\left({\color{gray}0.968 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 1033	$$13.4 \mathrm{ms} \pm 112 \mathrm{μs}\left({\color{gray}4.08 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_medium

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 102	$$3.60 \mathrm{ms} \pm 19.5 \mathrm{μs}\left({\color{gray}-0.187 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.82 \mathrm{ms} \pm 16.4 \mathrm{μs}\left({\color{gray}0.591 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 51	$$3.17 \mathrm{ms} \pm 21.3 \mathrm{μs}\left({\color{gray}0.533 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 269	$$4.94 \mathrm{ms} \pm 30.3 \mathrm{μs}\left({\color{gray}-0.419 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$3.38 \mathrm{ms} \pm 17.9 \mathrm{μs}\left({\color{gray}-1.614 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 107	$$3.94 \mathrm{ms} \pm 28.7 \mathrm{μs}\left({\color{gray}-0.356 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 133	$$4.23 \mathrm{ms} \pm 25.5 \mathrm{μs}\left({\color{gray}0.734 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$3.28 \mathrm{ms} \pm 19.3 \mathrm{μs}\left({\color{gray}1.42 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 63	$$3.86 \mathrm{ms} \pm 24.2 \mathrm{μs}\left({\color{gray}0.633 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_none

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 2	$$2.57 \mathrm{ms} \pm 9.35 \mathrm{μs}\left({\color{gray}-2.341 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.54 \mathrm{ms} \pm 12.5 \mathrm{μs}\left({\color{gray}-0.907 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 1	$$2.65 \mathrm{ms} \pm 12.7 \mathrm{μs}\left({\color{gray}-2.020 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 8	$$2.88 \mathrm{ms} \pm 18.4 \mathrm{μs}\left({\color{gray}-1.368 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$2.69 \mathrm{ms} \pm 13.3 \mathrm{μs}\left({\color{gray}-2.101 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 3	$$2.98 \mathrm{ms} \pm 17.6 \mathrm{μs}\left({\color{gray}-0.426 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_small

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 52	$$2.96 \mathrm{ms} \pm 13.2 \mathrm{μs}\left({\color{gray}0.986 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.62 \mathrm{ms} \pm 12.5 \mathrm{μs}\left({\color{gray}-0.355 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 25	$$2.79 \mathrm{ms} \pm 11.5 \mathrm{μs}\left({\color{gray}-0.003 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 94	$$3.31 \mathrm{ms} \pm 17.3 \mathrm{μs}\left({\color{gray}-0.297 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$2.89 \mathrm{ms} \pm 14.9 \mathrm{μs}\left({\color{gray}0.622 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 26	$$3.13 \mathrm{ms} \pm 20.6 \mathrm{μs}\left({\color{gray}-0.636 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 66	$$3.23 \mathrm{ms} \pm 15.8 \mathrm{μs}\left({\color{gray}-1.784 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$2.85 \mathrm{ms} \pm 10.0 \mathrm{μs}\left({\color{gray}-0.136 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 29	$$3.18 \mathrm{ms} \pm 20.3 \mathrm{μs}\left({\color{gray}2.50 \mathrm{\%}}\right) $$	Flame Graph

read_scaling_complete

Function	Value	Mean	Flame graphs
entity_by_id;one_depth	1 entities	$$39.2 \mathrm{ms} \pm 192 \mathrm{μs}\left({\color{gray}-0.898 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	10 entities	$$76.1 \mathrm{ms} \pm 424 \mathrm{μs}\left({\color{gray}-1.277 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	25 entities	$$43.0 \mathrm{ms} \pm 234 \mathrm{μs}\left({\color{gray}-0.068 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	5 entities	$$45.5 \mathrm{ms} \pm 167 \mathrm{μs}\left({\color{gray}-1.095 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	50 entities	$$53.5 \mathrm{ms} \pm 354 \mathrm{μs}\left({\color{gray}-0.424 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	1 entities	$$40.3 \mathrm{ms} \pm 197 \mathrm{μs}\left({\color{gray}0.095 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	10 entities	$$417 \mathrm{ms} \pm 1.29 \mathrm{ms}\left({\color{gray}0.156 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	25 entities	$$93.1 \mathrm{ms} \pm 527 \mathrm{μs}\left({\color{red}5.43 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	5 entities	$$84.1 \mathrm{ms} \pm 371 \mathrm{μs}\left({\color{gray}-1.213 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	50 entities	$$282 \mathrm{ms} \pm 1.09 \mathrm{ms}\left({\color{gray}1.35 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	1 entities	$$14.9 \mathrm{ms} \pm 83.8 \mathrm{μs}\left({\color{gray}3.20 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	10 entities	$$15.0 \mathrm{ms} \pm 80.8 \mathrm{μs}\left({\color{gray}0.533 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	25 entities	$$15.4 \mathrm{ms} \pm 85.4 \mathrm{μs}\left({\color{gray}1.93 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	5 entities	$$14.6 \mathrm{ms} \pm 71.2 \mathrm{μs}\left({\color{gray}0.318 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	50 entities	$$17.9 \mathrm{ms} \pm 99.0 \mathrm{μs}\left({\color{gray}0.864 \mathrm{\%}}\right) $$	Flame Graph

read_scaling_linkless

Function	Value	Mean	Flame graphs
entity_by_id	1 entities	$$15.0 \mathrm{ms} \pm 90.9 \mathrm{μs}\left({\color{gray}3.31 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	10 entities	$$15.0 \mathrm{ms} \pm 83.8 \mathrm{μs}\left({\color{gray}2.49 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	100 entities	$$14.9 \mathrm{ms} \pm 80.5 \mathrm{μs}\left({\color{gray}1.94 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	1000 entities	$$15.4 \mathrm{ms} \pm 75.3 \mathrm{μs}\left({\color{gray}0.812 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	10000 entities	$$23.1 \mathrm{ms} \pm 135 \mathrm{μs}\left({\color{gray}3.93 \mathrm{\%}}\right) $$	Flame Graph

representative_read_entity

Function	Value	Mean	Flame graphs
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/block/v/1`	$$31.1 \mathrm{ms} \pm 290 \mathrm{μs}\left({\color{gray}-0.814 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/book/v/1`	$$31.7 \mathrm{ms} \pm 280 \mathrm{μs}\left({\color{gray}-1.379 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/building/v/1`	$$30.4 \mathrm{ms} \pm 292 \mathrm{μs}\left({\color{gray}-2.438 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/organization/v/1`	$$30.2 \mathrm{ms} \pm 333 \mathrm{μs}\left({\color{lightgreen}-6.626 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/page/v/2`	$$30.5 \mathrm{ms} \pm 265 \mathrm{μs}\left({\color{gray}-1.620 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/person/v/1`	$$30.6 \mathrm{ms} \pm 287 \mathrm{μs}\left({\color{gray}-2.012 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/playlist/v/1`	$$30.8 \mathrm{ms} \pm 293 \mathrm{μs}\left({\color{gray}-0.867 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/song/v/1`	$$29.7 \mathrm{ms} \pm 310 \mathrm{μs}\left({\color{gray}-4.914 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: `https://blockprotocol.org/@alice/types/entity-type/uk-address/v/1`	$$30.8 \mathrm{ms} \pm 301 \mathrm{μs}\left({\color{gray}-3.013 \mathrm{\%}}\right) $$	Flame Graph

representative_read_entity_type

Function	Value	Mean	Flame graphs
get_entity_type_by_id	Account ID: `bf5a9ef5-dc3b-43cf-a291-6210c0321eba`	$$8.27 \mathrm{ms} \pm 39.5 \mathrm{μs}\left({\color{gray}-4.087 \mathrm{\%}}\right) $$	Flame Graph

representative_read_multiple_entities

Function	Value	Mean	Flame graphs
entity_by_property	traversal_paths=0	0	$$91.7 \mathrm{ms} \pm 535 \mathrm{μs}\left({\color{gray}-4.905 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=255	1,resolve_depths=inherit:1;values:255;properties:255;links:127;link_dests:126;type:true	$$139 \mathrm{ms} \pm 520 \mathrm{μs}\left({\color{gray}0.036 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:0;link_dests:0;type:false	$$98.4 \mathrm{ms} \pm 548 \mathrm{μs}\left({\color{gray}-2.871 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:1;link_dests:0;type:true	$$106 \mathrm{ms} \pm 494 \mathrm{μs}\left({\color{gray}-2.610 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:2;links:1;link_dests:0;type:true	$$114 \mathrm{ms} \pm 471 \mathrm{μs}\left({\color{gray}-0.666 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:2;properties:2;links:1;link_dests:0;type:true	$$122 \mathrm{ms} \pm 561 \mathrm{μs}\left({\color{gray}0.189 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=0	0	$$88.0 \mathrm{ms} \pm 506 \mathrm{μs}\left({\color{gray}2.63 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=255	1,resolve_depths=inherit:1;values:255;properties:255;links:127;link_dests:126;type:true	$$115 \mathrm{ms} \pm 488 \mathrm{μs}\left({\color{gray}1.63 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:0;link_dests:0;type:false	$$94.6 \mathrm{ms} \pm 505 \mathrm{μs}\left({\color{gray}1.83 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:1;link_dests:0;type:true	$$102 \mathrm{ms} \pm 432 \mathrm{μs}\left({\color{gray}1.81 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:2;links:1;link_dests:0;type:true	$$105 \mathrm{ms} \pm 507 \mathrm{μs}\left({\color{gray}2.17 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:2;properties:2;links:1;link_dests:0;type:true	$$104 \mathrm{ms} \pm 575 \mathrm{μs}\left({\color{gray}1.06 \mathrm{\%}}\right) $$

scenarios

Function	Value	Mean	Flame graphs
full_test	query-limited	$$137 \mathrm{ms} \pm 415 \mathrm{μs}\left({\color{gray}3.00 \mathrm{\%}}\right) $$	Flame Graph
full_test	query-unlimited	$$139 \mathrm{ms} \pm 483 \mathrm{μs}\left({\color{gray}3.22 \mathrm{\%}}\right) $$	Flame Graph
linked_queries	query-limited	$$103 \mathrm{ms} \pm 556 \mathrm{μs}\left({\color{gray}-0.655 \mathrm{\%}}\right) $$	Flame Graph
linked_queries	query-unlimited	$$573 \mathrm{ms} \pm 3.18 \mathrm{ms}\left({\color{gray}0.272 \mathrm{\%}}\right) $$	Flame Graph

indietyp added 5 commits February 11, 2026 20:03

feat: implement new bitmatrix

9d37fc5

feat: implement new bitmatrix

9f35562

feat: implement new bitmatrix

ce2ffa1

chore: convert callers

2019b9b

chore: remove stray markdown

b1e1a0c

This was referenced Feb 11, 2026

BE-360: HashQL: Rework compiletest UI #8340

Open

BE-303: HashQL: Split MIR basic blocks by target support #8354

Open

indietyp mentioned this pull request Feb 11, 2026

BE-372: HashQL: Add traversal-aware liveness analysis for optimizing entity projections #8363

Open

3 tasks

This was referenced Feb 11, 2026

BE-254: Add CodSpeed benchmarking for HashQL MIR transformation passes #8204

Merged

BE-231: HashQL: Add InstSimplify pass for MIR instruction simplification and constant folding #8206

Merged

graphite-app bot assigned indietyp Feb 11, 2026

github-actions bot added area/libs Relates to first-party libraries/crates/packages (area) type/eng > backend Owned by the @backend team labels Feb 11, 2026

augmentcode bot reviewed Feb 11, 2026

View reviewed changes

graphite-app bot reviewed Feb 11, 2026

View reviewed changes

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs Show resolved Hide resolved

vercel bot deployed to Preview – hash February 11, 2026 21:37 View deployment

cursor bot reviewed Feb 11, 2026

View reviewed changes

libs/@local/hashql/core/src/id/bit_vec/matrix/mod.rs Show resolved Hide resolved

indietyp force-pushed the bm/be-372-hashql-do-not-consider-traversals-during-liveness-analysis branch from 5351b95 to e20307f Compare February 11, 2026 22:03

indietyp force-pushed the bm/be-394-hashql-rework-bitmatrix branch from fa38f9f to b1e1a0c Compare February 11, 2026 22:03

vercel bot temporarily deployed to Preview – petrinaut February 11, 2026 22:03 Inactive

vercel bot deployed to Preview – hash February 11, 2026 22:10 View deployment

indietyp changed the title ~~BE-394: Rework BitMatrix and SparseBitMatrix~~ BE-394: HashQL: Rework BitMatrix and SparseBitMatrix Feb 12, 2026

graphite-app bot reviewed Feb 12, 2026

View reviewed changes

libs/@local/hashql/core/src/graph/algorithms/dominators/frontier.rs Show resolved Hide resolved

chore: fix docs

212c24c

vercel bot temporarily deployed to Preview – petrinaut February 12, 2026 09:11 Inactive

vercel bot deployed to Preview – hash February 12, 2026 09:18 View deployment

graphite-app bot requested review from a team February 12, 2026 09:32

BE-394: HashQL: Rework BitMatrix and SparseBitMatrix #8408

Are you sure you want to change the base?

BE-394: HashQL: Rework BitMatrix and SparseBitMatrix #8408

Conversation

indietyp commented Feb 11, 2026

🌟 What is the purpose of this PR?

Performance

🔍 What does this change?

🛡 What tests cover this?

Pre-Merge Checklist 🚀

🚢 Has this modified a publishable library?

📜 Does this require a change to the docs?

🕸️ Does this require a change to the Turbo Graph?

Uh oh!

vercel bot commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cursor bot commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Summary

Uh oh!

indietyp commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

augmentcode bot commented Feb 11, 2026

Uh oh!

augmentcode bot left a comment

Choose a reason for hiding this comment

Uh oh!

augmentcode bot Feb 11, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

augmentcode bot Feb 11, 2026

Choose a reason for hiding this comment

Uh oh!

codecov bot commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Uh oh!

codspeed-hq bot commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Merging this PR will improve performance by 35.34%

Performance Changes

Footnotes

Uh oh!

cursor bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

github-actions bot commented Feb 12, 2026

Benchmark results

@rust/hash-graph-benches – Integrations

policy_resolution_large

policy_resolution_medium

policy_resolution_none

policy_resolution_small

read_scaling_complete

read_scaling_linkless

representative_read_entity

representative_read_entity_type

representative_read_multiple_entities

scenarios

Uh oh!

Reviewers

Assignees

Labels

Milestone

Development

Uh oh!

1 participant

vercel bot commented Feb 11, 2026 •

edited

Loading

cursor bot commented Feb 11, 2026 •

edited

Loading

indietyp commented Feb 11, 2026 •

edited

Loading

codecov bot commented Feb 11, 2026 •

edited

Loading

codspeed-hq bot commented Feb 11, 2026 •

edited

Loading