BE-421: HashQL: Add equality safety checks for Postgres statement placement

[indietyp]

🌟 What is the purpose of this PR?

This PR adds equality safety checks to the Postgres statement placement pass in the HashQL MIR compiler. The changes prevent equality and inequality comparisons from being executed in Postgres when the operand types could produce identical jsonb representations while being semantically distinct in the interpreter.

🔍 What does this change?

• Adds a debug_panic! macro to the hashql-core library for debug-only panics
• Implements is_equality_safe and is_equality_safe_operand functions that check for representational collisions between types in jsonb serialization
• Adds a peel function that strips type wrappers (Opaque, Apply, Generic) to reach underlying concrete types
• Modifies the Postgres statement placement pass to reject == and != operations when operands could have representational collisions (e.g., Dict<String, _> vs struct, List<_> vs tuple)
• Adds comprehensive test coverage for equality safety checks including edge cases with unions, intersections, unknown types, and nested structures
• Updates import statements across multiple files to include newly required types and functions

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

🛡 What tests cover this?

• New unit tests for is_equality_safe function covering various type combinations
• Integration tests for the Postgres statement placement pass with equality operations
• Test cases for representational collisions between Dict/struct and List/tuple types
• Tests for edge cases involving unknown types, unions, intersections, and constants

❓ How to test this?

1. Checkout the branch
2. Run the test suite for the hashql-mir crate
3. Verify that equality comparisons between representationally colliding types are rejected by the Postgres placement pass
4. Confirm that safe equality comparisons are still accepted

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
hash	Ready	Preview, Comment	Feb 26, 2026 5:04pm
petrinaut	Ready	Preview	Feb 26, 2026 5:04pm

2 Skipped Deployments

Project	Deployment	Actions	Updated (UTC)
hashdotdesign	Ignored	Preview	Feb 26, 2026 5:04pm
hashdotdesign-tokens	Ignored	Preview	Feb 26, 2026 5:04pm

[cursor]

PR Summary

Medium Risk
Changes Postgres dispatch eligibility for ==/!= by adding a new type-walking safety check, which can alter where filters execute (DB vs interpreter) and could impact query performance/behavior if the logic is wrong. The change is well-covered by new unit and snapshot tests, but touches core type recursion utilities shared across crates.

Overview
Prevents Postgres statement placement from dispatching ==/!= when operand types could serialize to the same jsonb shape but be semantically different (notably Dict<String, _> vs struct and List<_> vs tuple, plus conservative rejection for ?/Param/Infer).

This introduces peel + is_equality_safe(_operand) with cycle detection via a now-public RecursionBoundary (and RecursiveVisitorGuard::boundary_mut()), plus a debug_panic! macro for debug-only assertions.

Adds extensive new Postgres placement and is_equality_safe tests (including recursive/union/intersection cases) and corresponding UI snapshot fixtures.

^{Written by Cursor Bugbot for commit 44aeaed. This will update automatically on new commits. Configure here.}

[augmentcode]

Review completed. 1 suggestions 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.}

[codecov]

Codecov Report

❌ Patch coverage is 97.43590% with 16 lines in your changes missing coverage. Please review.
✅ Project coverage is 62.94%. Comparing base (e7770d8) to head (44aeaed).
⚠️ Report is 8 commits behind head on main.

Files with missing lines	Patch %	Lines
...pass/execution/statement_placement/postgres/mod.rs	93.43%	6 Missing and 3 partials ⚠️
libs/@local/hashql/core/src/type/recursion.rs	50.00%	4 Missing ⚠️
libs/@local/hashql/core/src/type/visit.rs	0.00%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #8480      +/-   ##
==========================================
+ Coverage   62.70%   62.94%   +0.23%     
==========================================
  Files        1302     1302              
  Lines      131589   132494     +905     
  Branches     5487     5503      +16     
==========================================
+ Hits        82512    83396     +884     
- Misses      48169    48184      +15     
- Partials      908      914       +6     

Flag	Coverage Δ
apps.hash-ai-worker-ts	1.40% <ø> (ø)
apps.hash-api	0.00% <ø> (ø)
local.hash-graph-sdk	7.78% <ø> (ø)
local.hash-isomorphic-utils	0.00% <ø> (ø)
rust.hash-graph-api	2.86% <ø> (ø)
rust.hashql-ast	87.25% <ø> (ø)
rust.hashql-compiletest	29.69% <ø> (ø)
rust.hashql-core	82.26% <36.36%> (-0.03%)	⬇️
rust.hashql-eval	69.13% <ø> (ø)
rust.hashql-hir	89.11% <ø> (ø)
rust.hashql-mir	92.29% <98.53%> (+0.24%)	⬆️
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.

[graphite-app]

Merge activity

• Feb 26, 4:56 PM UTC: Graphite rebased this pull request, because this pull request is set to merge when ready.

[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	$$27.7 \mathrm{ms} \pm 148 \mathrm{μs}\left({\color{gray}-0.178 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$3.53 \mathrm{ms} \pm 20.7 \mathrm{μs}\left({\color{gray}-0.191 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 1001	$$12.6 \mathrm{ms} \pm 86.6 \mathrm{μs}\left({\color{gray}-1.409 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 3314	$$44.4 \mathrm{ms} \pm 362 \mathrm{μs}\left({\color{gray}-0.025 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$15.2 \mathrm{ms} \pm 113 \mathrm{μs}\left({\color{gray}0.887 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 1526	$$25.0 \mathrm{ms} \pm 158 \mathrm{μs}\left({\color{gray}0.055 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 2078	$$28.8 \mathrm{ms} \pm 178 \mathrm{μs}\left({\color{gray}-0.429 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$3.78 \mathrm{ms} \pm 19.4 \mathrm{μs}\left({\color{gray}-1.011 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 1033	$$13.7 \mathrm{ms} \pm 102 \mathrm{μs}\left({\color{gray}-0.789 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_medium

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 102	$$3.93 \mathrm{ms} \pm 18.9 \mathrm{μs}\left({\color{gray}0.198 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$3.10 \mathrm{ms} \pm 12.2 \mathrm{μs}\left({\color{gray}-0.787 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 51	$$3.46 \mathrm{ms} \pm 16.1 \mathrm{μs}\left({\color{gray}0.034 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 269	$$5.34 \mathrm{ms} \pm 29.9 \mathrm{μs}\left({\color{gray}-0.075 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$3.71 \mathrm{ms} \pm 17.5 \mathrm{μs}\left({\color{gray}0.134 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 107	$$4.28 \mathrm{ms} \pm 26.4 \mathrm{μs}\left({\color{gray}-0.390 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 133	$$4.61 \mathrm{ms} \pm 24.2 \mathrm{μs}\left({\color{gray}0.012 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$3.59 \mathrm{ms} \pm 17.1 \mathrm{μs}\left({\color{gray}0.382 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 63	$$4.20 \mathrm{ms} \pm 22.5 \mathrm{μs}\left({\color{gray}-0.715 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_none

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 2	$$2.81 \mathrm{ms} \pm 11.7 \mathrm{μs}\left({\color{gray}-1.260 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.76 \mathrm{ms} \pm 10.7 \mathrm{μs}\left({\color{gray}-1.128 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 1	$$2.89 \mathrm{ms} \pm 14.8 \mathrm{μs}\left({\color{gray}-0.812 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 8	$$3.14 \mathrm{ms} \pm 14.3 \mathrm{μs}\left({\color{gray}-0.850 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$2.95 \mathrm{ms} \pm 13.8 \mathrm{μs}\left({\color{gray}-1.644 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 3	$$3.25 \mathrm{ms} \pm 15.8 \mathrm{μs}\left({\color{gray}-1.236 \mathrm{\%}}\right) $$	Flame Graph

policy_resolution_small

Function	Value	Mean	Flame graphs
resolve_policies_for_actor	user: empty, selectivity: high, policies: 52	$$3.21 \mathrm{ms} \pm 12.7 \mathrm{μs}\left({\color{gray}-0.043 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: low, policies: 1	$$2.92 \mathrm{ms} \pm 13.8 \mathrm{μs}\left({\color{gray}-0.741 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: empty, selectivity: medium, policies: 25	$$3.08 \mathrm{ms} \pm 14.0 \mathrm{μs}\left({\color{gray}-0.074 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: high, policies: 94	$$3.69 \mathrm{ms} \pm 23.7 \mathrm{μs}\left({\color{gray}1.15 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: low, policies: 1	$$3.22 \mathrm{ms} \pm 17.1 \mathrm{μs}\left({\color{gray}0.404 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: seeded, selectivity: medium, policies: 26	$$3.46 \mathrm{ms} \pm 18.3 \mathrm{μs}\left({\color{gray}0.583 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: high, policies: 66	$$3.60 \mathrm{ms} \pm 21.4 \mathrm{μs}\left({\color{gray}0.127 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: low, policies: 1	$$3.18 \mathrm{ms} \pm 17.9 \mathrm{μs}\left({\color{gray}-0.553 \mathrm{\%}}\right) $$	Flame Graph
resolve_policies_for_actor	user: system, selectivity: medium, policies: 29	$$3.49 \mathrm{ms} \pm 17.8 \mathrm{μs}\left({\color{gray}0.589 \mathrm{\%}}\right) $$	Flame Graph

read_scaling_complete

Function	Value	Mean	Flame graphs
entity_by_id;one_depth	1 entities	$$42.1 \mathrm{ms} \pm 178 \mathrm{μs}\left({\color{gray}-0.025 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	10 entities	$$79.3 \mathrm{ms} \pm 394 \mathrm{μs}\left({\color{gray}0.252 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	25 entities	$$46.6 \mathrm{ms} \pm 203 \mathrm{μs}\left({\color{gray}1.29 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	5 entities	$$49.9 \mathrm{ms} \pm 262 \mathrm{μs}\left({\color{gray}2.28 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;one_depth	50 entities	$$57.4 \mathrm{ms} \pm 326 \mathrm{μs}\left({\color{gray}-0.686 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	1 entities	$$44.3 \mathrm{ms} \pm 225 \mathrm{μs}\left({\color{gray}1.13 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	10 entities	$$424 \mathrm{ms} \pm 929 \mathrm{μs}\left({\color{gray}1.54 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	25 entities	$$101 \mathrm{ms} \pm 501 \mathrm{μs}\left({\color{gray}2.06 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	5 entities	$$89.8 \mathrm{ms} \pm 460 \mathrm{μs}\left({\color{gray}3.22 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;two_depth	50 entities	$$327 \mathrm{ms} \pm 961 \mathrm{μs}\left({\color{gray}4.12 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	1 entities	$$17.2 \mathrm{ms} \pm 93.9 \mathrm{μs}\left({\color{red}8.01 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	10 entities	$$17.0 \mathrm{ms} \pm 94.0 \mathrm{μs}\left({\color{gray}4.70 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	25 entities	$$17.5 \mathrm{ms} \pm 96.3 \mathrm{μs}\left({\color{red}6.69 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	5 entities	$$16.9 \mathrm{ms} \pm 97.3 \mathrm{μs}\left({\color{gray}4.78 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id;zero_depth	50 entities	$$19.7 \mathrm{ms} \pm 135 \mathrm{μs}\left({\color{gray}3.28 \mathrm{\%}}\right) $$	Flame Graph

read_scaling_linkless

Function	Value	Mean	Flame graphs
entity_by_id	1 entities	$$15.9 \mathrm{ms} \pm 68.6 \mathrm{μs}\left({\color{gray}0.783 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	10 entities	$$16.5 \mathrm{ms} \pm 80.9 \mathrm{μs}\left({\color{gray}3.05 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	100 entities	$$16.3 \mathrm{ms} \pm 61.1 \mathrm{μs}\left({\color{gray}1.66 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	1000 entities	$$16.8 \mathrm{ms} \pm 80.1 \mathrm{μs}\left({\color{gray}1.92 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	10000 entities	$$24.6 \mathrm{ms} \pm 172 \mathrm{μs}\left({\color{gray}2.73 \mathrm{\%}}\right) $$	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	$$31.0 \mathrm{ms} \pm 275 \mathrm{μs}\left({\color{gray}0.004 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/book/v/1	$$31.4 \mathrm{ms} \pm 249 \mathrm{μs}\left({\color{gray}1.80 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/building/v/1	$$31.1 \mathrm{ms} \pm 295 \mathrm{μs}\left({\color{gray}2.64 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/organization/v/1	$$31.2 \mathrm{ms} \pm 310 \mathrm{μs}\left({\color{gray}-3.398 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/page/v/2	$$31.5 \mathrm{ms} \pm 265 \mathrm{μs}\left({\color{gray}1.06 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/person/v/1	$$31.6 \mathrm{ms} \pm 257 \mathrm{μs}\left({\color{gray}-0.674 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/playlist/v/1	$$31.1 \mathrm{ms} \pm 273 \mathrm{μs}\left({\color{gray}0.559 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/song/v/1	$$30.6 \mathrm{ms} \pm 221 \mathrm{μs}\left({\color{gray}-3.060 \mathrm{\%}}\right) $$	Flame Graph
entity_by_id	entity type ID: https://blockprotocol.org/@alice/types/entity-type/uk-address/v/1	$$30.9 \mathrm{ms} \pm 298 \mathrm{μs}\left({\color{gray}-2.496 \mathrm{\%}}\right) $$	Flame Graph

representative_read_entity_type

Function	Value	Mean	Flame graphs
get_entity_type_by_id	Account ID: bf5a9ef5-dc3b-43cf-a291-6210c0321eba	$$8.79 \mathrm{ms} \pm 45.3 \mathrm{μs}\left({\color{gray}-0.640 \mathrm{\%}}\right) $$	Flame Graph

representative_read_multiple_entities

Function	Value	Mean	Flame graphs
entity_by_property	traversal_paths=0	0	$$90.7 \mathrm{ms} \pm 381 \mathrm{μs}\left({\color{gray}-0.045 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=255	1,resolve_depths=inherit:1;values:255;properties:255;links:127;link_dests:126;type:true	$$144 \mathrm{ms} \pm 624 \mathrm{μs}\left({\color{gray}0.187 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:0;link_dests:0;type:false	$$97.5 \mathrm{ms} \pm 408 \mathrm{μs}\left({\color{gray}-0.599 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:1;link_dests:0;type:true	$$106 \mathrm{ms} \pm 436 \mathrm{μs}\left({\color{gray}-0.362 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:2;links:1;link_dests:0;type:true	$$116 \mathrm{ms} \pm 468 \mathrm{μs}\left({\color{gray}0.496 \mathrm{\%}}\right) $$
entity_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:2;properties:2;links:1;link_dests:0;type:true	$$123 \mathrm{ms} \pm 533 \mathrm{μs}\left({\color{gray}0.180 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=0	0	$$88.8 \mathrm{ms} \pm 415 \mathrm{μs}\left({\color{gray}0.271 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=255	1,resolve_depths=inherit:1;values:255;properties:255;links:127;link_dests:126;type:true	$$118 \mathrm{ms} \pm 541 \mathrm{μs}\left({\color{gray}0.015 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:0;link_dests:0;type:false	$$95.9 \mathrm{ms} \pm 461 \mathrm{μs}\left({\color{gray}-0.056 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:0;links:1;link_dests:0;type:true	$$104 \mathrm{ms} \pm 582 \mathrm{μs}\left({\color{gray}-0.215 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:0;properties:2;links:1;link_dests:0;type:true	$$107 \mathrm{ms} \pm 675 \mathrm{μs}\left({\color{gray}0.829 \mathrm{\%}}\right) $$
link_by_source_by_property	traversal_paths=2	1,resolve_depths=inherit:0;values:2;properties:2;links:1;link_dests:0;type:true	$$107 \mathrm{ms} \pm 556 \mathrm{μs}\left({\color{gray}0.675 \mathrm{\%}}\right) $$

scenarios

Function	Value	Mean	Flame graphs
full_test	query-limited	$$136 \mathrm{ms} \pm 514 \mathrm{μs}\left({\color{gray}1.99 \mathrm{\%}}\right) $$	Flame Graph
full_test	query-unlimited	$$135 \mathrm{ms} \pm 505 \mathrm{μs}\left({\color{gray}-0.318 \mathrm{\%}}\right) $$	Flame Graph
linked_queries	query-limited	$$67.9 \mathrm{ms} \pm 3.29 \mathrm{ms}\left({\color{lightgreen}-36.440 \mathrm{\%}}\right) $$	Flame Graph
linked_queries	query-unlimited	$$581 \mathrm{ms} \pm 1.07 \mathrm{ms}\left({\color{gray}-3.046 \mathrm{\%}}\right) $$	Flame Graph