BE-515: HashQL: Remove OFFSET 0 materialization fence from generated query

[indietyp]

🌟 What is the purpose of this PR?

HashQL: Remove OFFSET 0 materialization fence from generated SQL

Problem

Each PostgreSQL island is wrapped in a CROSS JOIN LATERAL subquery. The subquery previously included OFFSET 0 as a materialization fence to prevent PostgreSQL's planner from inlining the CASE tree expression and duplicating it across each composite field access in the SELECT list.

The planner's cost estimate favors the fenced plan (lower estimated cost), but empirical benchmarks show the opposite for queries with conditional CASE trees:

Query	Change (fenced → unfenced)
all_entities	−11% (improved)
filter_by_uuid	+10% (regressed)
diamond_cfg	−5% (improved)
sequential_filters	−72% (improved)
stress_test	−77% (improved)

For multi-body queries with conditional branches, the materialization overhead exceeds the duplication cost by 3–4×. The fence helps modestly for simple single-expression queries without branching (filter_by_uuid), but is catastrophic for the common case.

🚢 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

🛡 What tests cover this?

• Existing test coverage for the HashQL eval Postgres module.

❓ How to test this?

1. Checkout the branch.
2. Run the existing test suite.
3. Confirm all tests pass and query behavior is unchanged.

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
hash	Ready	Preview, Comment	May 28, 2026 1:35pm
petrinaut	Ready	Preview	May 28, 2026 1:35pm

1 Skipped Deployment

Project	Deployment	Actions	Updated (UTC)
hashdotdesign-tokens	Ignored	Preview	May 28, 2026 1:35pm

[indietyp]

• BE-537: Remove old filter API backend and wire up HashQL in the API #8827
• BE-513: HashQL: Rework dynamic aggregate size estimation #8697
• BE-512: HashQL: Allow cross-backend transitions for SwitchInt terminators #8696
• BE-524: HashQL: Remove island schedule module #8694
• BE-525: HashQL: Remove implicit type-widening from Integer operations, and make comparison size aware #8693
• BE-515: HashQL: Remove OFFSET 0 materialization fence from generated query #8650 👈 (View in Graphite)
• BE-514: HashQL: implement iterative adjustment passes in placement solver #8635
• BE-500, BE-501: HashQL: Unify mixed parameter resolution in data dependency analysis #8607
• BE-494: HashQL: Loop-breaker selection for recursive inlining #8600
• BE-482: HashQL: Remove logical not from MIR and fix postgres boolean lowering #8595
• BE-474: HashQL: Take into account terminator target eligibility in execution backend placement #8587
• BE-455: HashQL: Introduce evaluation orchestrator #8586
• BE-523: HashQL: Suspendable interpreter with structured values and input system #8675
• BE-522: HashQL: Expand stdlib type system with temporal types #8674
• BE-306: HashQL: PostgreSQL translation #8526
• BE-457: HashQL: MIR execution pipeline extensions for postgres compilation #8525
• BE-456: HashQL: Entity type model enrichment and graph store query extensions #8524
• BE-429: HashQL: Add island dependency graph with data requirement resolution #8501
• BE-428: HashQL: Simplify traversal tracking to path recording #8498
• BE-434: HashQL: Refactor entity storage path resolution with centralized schema mapping #8496
• BE-427: Add hashql-macros crate with procedural macros for ID types and symbols #8495
• main

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

[augmentcode]

🤖 Augment PR Summary

Summary: Removes the OFFSET 0 materialization fence from generated Postgres lateral subqueries in HashQL evaluation.
Why: This avoids unnecessary materialization overhead and aligns the generated SQL with benchmark results showing better performance for multi-branch CASE-tree queries.

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

[augmentcode]

Review completed. 1 suggestion posted.

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

[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.}

^{Reviewed by Cursor Bugbot for commit caff769. Configure here.}

[github-actions]

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	$$26.4 \mathrm{ms} \pm 119 \mathrm{μs}\left({\color{gray}2.32 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.84 \mathrm{ms} \pm 14.1 \mathrm{μs}\left({\color{gray}0.338 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 1001	$$11.6 \mathrm{ms} \pm 65.6 \mathrm{μs}\left({\color{gray}1.71 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 3314	$$37.4 \mathrm{ms} \pm 310 \mathrm{μs}\left({\color{gray}-1.409 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$11.4 \mathrm{ms} \pm 92.6 \mathrm{μs}\left({\color{gray}2.28 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 1526	$$20.6 \mathrm{ms} \pm 140 \mathrm{μs}\left({\color{gray}-2.779 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 2078	$$26.7 \mathrm{ms} \pm 202 \mathrm{μs}\left({\color{gray}-0.927 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$3.11 \mathrm{ms} \pm 14.1 \mathrm{μs}\left({\color{gray}0.042 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 1033	$$12.5 \mathrm{ms} \pm 80.1 \mathrm{μs}\left({\color{gray}1.00 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_medium

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 102	$$3.13 \mathrm{ms} \pm 18.6 \mathrm{μs}\left({\color{gray}-0.256 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.47 \mathrm{ms} \pm 21.2 \mathrm{μs}\left({\color{gray}3.60 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 51	$$2.74 \mathrm{ms} \pm 14.2 \mathrm{μs}\left({\color{gray}0.667 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 269	$$4.45 \mathrm{ms} \pm 19.2 \mathrm{μs}\left({\color{gray}0.975 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$2.89 \mathrm{ms} \pm 11.9 \mathrm{μs}\left({\color{gray}0.448 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 107	$$3.43 \mathrm{ms} \pm 16.9 \mathrm{μs}\left({\color{gray}0.935 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 133	$$3.74 \mathrm{ms} \pm 20.5 \mathrm{μs}\left({\color{gray}0.334 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$2.81 \mathrm{ms} \pm 10.7 \mathrm{μs}\left({\color{gray}-1.841 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 63	$$3.35 \mathrm{ms} \pm 15.0 \mathrm{μs}\left({\color{gray}0.251 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_none

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 2	$$2.13 \mathrm{ms} \pm 11.5 \mathrm{μs}\left({\color{gray}0.766 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.02 \mathrm{ms} \pm 9.69 \mathrm{μs}\left({\color{gray}0.472 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 1	$$2.11 \mathrm{ms} \pm 8.24 \mathrm{μs}\left({\color{gray}-0.122 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 8	$$2.33 \mathrm{ms} \pm 10.1 \mathrm{μs}\left({\color{gray}0.248 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$2.20 \mathrm{ms} \pm 12.3 \mathrm{μs}\left({\color{gray}0.504 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 3	$$2.32 \mathrm{ms} \pm 12.0 \mathrm{μs}\left({\color{gray}0.407 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_small

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 52	$$2.46 \mathrm{ms} \pm 13.5 \mathrm{μs}\left({\color{gray}1.43 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.22 \mathrm{ms} \pm 9.64 \mathrm{μs}\left({\color{gray}0.814 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 25	$$2.40 \mathrm{ms} \pm 10.7 \mathrm{μs}\left({\color{gray}-0.511 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 94	$$2.78 \mathrm{ms} \pm 16.0 \mathrm{μs}\left({\color{gray}0.816 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$2.38 \mathrm{ms} \pm 9.85 \mathrm{μs}\left({\color{gray}0.705 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 26	$$2.64 \mathrm{ms} \pm 13.2 \mathrm{μs}\left({\color{gray}0.980 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 66	$$2.67 \mathrm{ms} \pm 11.9 \mathrm{μs}\left({\color{gray}-0.239 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$2.39 \mathrm{ms} \pm 13.2 \mathrm{μs}\left({\color{gray}1.12 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 29	$$2.67 \mathrm{ms} \pm 12.9 \mathrm{μs}\left({\color{gray}0.709 \mathrm{\%}}\right) $$	Flame Graph

read_scaling_complete

Function	Value	Mean	Flame graphs
entity_by_id;one_depth	1 entities	$$47.7 \mathrm{ms} \pm 178 \mathrm{μs}\left({\color{gray}-0.693 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	10 entities	$$39.6 \mathrm{ms} \pm 153 \mathrm{μs}\left({\color{gray}0.294 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	25 entities	$$42.7 \mathrm{ms} \pm 177 \mathrm{μs}\left({\color{gray}-0.910 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	5 entities	$$37.5 \mathrm{ms} \pm 170 \mathrm{μs}\left({\color{gray}-1.279 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	50 entities	$$54.5 \mathrm{ms} \pm 240 \mathrm{μs}\left({\color{gray}0.825 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	1 entities	$$53.9 \mathrm{ms} \pm 188 \mathrm{μs}\left({\color{gray}-4.059 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	10 entities	$$48.0 \mathrm{ms} \pm 191 \mathrm{μs}\left({\color{gray}1.26 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	25 entities	$$85.1 \mathrm{ms} \pm 322 \mathrm{μs}\left({\color{lightgreen}-7.476 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	5 entities	$$38.8 \mathrm{ms} \pm 149 \mathrm{μs}\left({\color{gray}-1.812 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	50 entities	$$280 \mathrm{ms} \pm 887 \mathrm{μs}\left({\color{gray}-3.002 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	1 entities	$$16.7 \mathrm{ms} \pm 82.5 \mathrm{μs}\left({\color{gray}3.22 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	10 entities	$$17.4 \mathrm{ms} \pm 75.1 \mathrm{μs}\left({\color{gray}0.597 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	25 entities	$$17.6 \mathrm{ms} \pm 102 \mathrm{μs}\left({\color{gray}1.27 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	5 entities	$$16.5 \mathrm{ms} \pm 83.6 \mathrm{μs}\left({\color{gray}-2.517 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	50 entities	$$20.8 \mathrm{ms} \pm 183 \mathrm{μs}\left({\color{gray}2.51 \mathrm{\%}}\right) $$	Flame Graph

read_scaling_linkless

Function	Value	Mean	Flame graphs
entity_by_id	1 entities	$$16.4 \mathrm{ms} \pm 66.1 \mathrm{μs}\left({\color{gray}1.35 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	10 entities	$$16.7 \mathrm{ms} \pm 85.8 \mathrm{μs}\left({\color{gray}2.27 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	100 entities	$$16.5 \mathrm{ms} \pm 85.1 \mathrm{μs}\left({\color{gray}-0.156 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	1000 entities	$$17.2 \mathrm{ms} \pm 67.1 \mathrm{μs}\left({\color{gray}-1.500 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	10000 entities	$$24.0 \mathrm{ms} \pm 151 \mathrm{μs}\left({\color{gray}1.54 \mathrm{\%}}\right) $$	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	$$31.5 \mathrm{ms} \pm 230 \mathrm{μs}\left({\color{gray}2.07 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/book/v/1	$$31.7 \mathrm{ms} \pm 245 \mathrm{μs}\left({\color{gray}3.72 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/building/v/1	$$29.9 \mathrm{ms} \pm 244 \mathrm{μs}\left({\color{lightgreen}-5.757 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/organization/v/1	$$30.7 \mathrm{ms} \pm 277 \mathrm{μs}\left({\color{gray}-0.301 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/page/v/2	$$30.8 \mathrm{ms} \pm 266 \mathrm{μs}\left({\color{gray}-3.183 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/person/v/1	$$31.0 \mathrm{ms} \pm 286 \mathrm{μs}\left({\color{gray}-2.494 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/playlist/v/1	$$31.6 \mathrm{ms} \pm 327 \mathrm{μs}\left({\color{gray}2.65 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/song/v/1	$$32.0 \mathrm{ms} \pm 328 \mathrm{μs}\left({\color{gray}-2.764 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/uk-address/v/1	$$31.5 \mathrm{ms} \pm 297 \mathrm{μs}\left({\color{gray}-1.020 \mathrm{\%}}\right) $$	Flame Graph

representative_read_entity_type

Function	Value	Mean	Flame graphs
get_entity_type_by_id	Account ID: bf5a9ef5-dc3b-43cf-a291-6210c0321eba	$$6.92 \mathrm{ms} \pm 30.1 \mathrm{μs}\left({\color{gray}-1.928 \mathrm{\%}}\right) $$	Flame Graph

representative_read_multiple_entities

Function	Value	Mean	Flame graphs
entity_by_property	traversal_paths=0	0	$$91.6 \mathrm{ms} \pm 393 \mathrm{μs}\left({\color{gray}-1.702 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=255	1,resolve_depths=inherit:1;values:255;properties:255;links:127;link_dests:126;type:true	$$137 \mathrm{ms} \pm 600 \mathrm{μs}\left({\color{gray}0.151 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:0;link_dests:0;type:false	$$98.1 \mathrm{ms} \pm 386 \mathrm{μs}\left({\color{gray}0.639 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:1;link_dests:0;type:true	$$110 \mathrm{ms} \pm 383 \mathrm{μs}\left({\color{gray}3.92 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:2;links:1;link_dests:0;type:true	$$113 \mathrm{ms} \pm 498 \mathrm{μs}\left({\color{gray}1.60 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:2;properties:2;links:1;link_dests:0;type:true	$$119 \mathrm{ms} \pm 631 \mathrm{μs}\left({\color{gray}-2.127 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=0	0	$$97.5 \mathrm{ms} \pm 359 \mathrm{μs}\left({\color{gray}-0.790 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=255	1,resolve_depths=inherit:1;values:255;properties:255;links:127;link_dests:126;type:true	$$123 \mathrm{ms} \pm 626 \mathrm{μs}\left({\color{gray}-0.543 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:0;link_dests:0;type:false	$$104 \mathrm{ms} \pm 459 \mathrm{μs}\left({\color{gray}-1.952 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:1;link_dests:0;type:true	$$112 \mathrm{ms} \pm 495 \mathrm{μs}\left({\color{gray}-0.463 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:2;links:1;link_dests:0;type:true	$$113 \mathrm{ms} \pm 510 \mathrm{μs}\left({\color{gray}-1.020 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:2;properties:2;links:1;link_dests:0;type:true	$$112 \mathrm{ms} \pm 465 \mathrm{μs}\left({\color{gray}-1.924 \mathrm{\%}}\right) $$

scenarios

Function	Value	Mean	Flame graphs
full_test	query-limited	$$161 \mathrm{ms} \pm 702 \mathrm{μs}\left({\color{gray}-2.337 \mathrm{\%}}\right) $$	Flame Graph
full_test	query-unlimited	$$131 \mathrm{ms} \pm 479 \mathrm{μs}\left({\color{gray}-2.120 \mathrm{\%}}\right) $$	Flame Graph
linked_queries	query-limited	$$35.4 \mathrm{ms} \pm 129 \mathrm{μs}\left({\color{gray}-1.476 \mathrm{\%}}\right) $$	Flame Graph
linked_queries	query-unlimited	$$493 \mathrm{ms} \pm 819 \mathrm{μs}\left({\color{gray}1.36 \mathrm{\%}}\right) $$	Flame Graph