[autoopt] 20260414-008-sparse-prune-retained-slices

[mediocregopher]

Reuse sorted retained leaves during sparse-trie pruning

Evidence

• The baseline samply profile shows reth_trie_sparse::arena::ArenaSparseSubtrie::prune with 6,150 exclusive samples on the sparse-trie maintenance path.
• SparseStateTrie::prune already computes one LFU-selected retained set for the whole prune pass, but the arena and parallel trie implementations each cloned and re-sorted retained_leaves again inside every prune call.
• The hot internal caller already supplies sorted per-address slot lists from retained_slots_by_address, so most of that clone/sort work was avoidable bookkeeping rather than trie logic.

Hypothesis

If we sort retained account leaves once in SparseStateTrie::prune and let sparse-trie prune implementations borrow already sorted slices instead of cloning them, gas throughput improves by ~0.2-0.6% because sparse-trie pruning does less repeated leaf-list allocation and sorting on the hot cache-maintenance path.

Success Metric

• gas_per_second (mgas_s.pct in summary.json) improves by >0.2%

Plan

• Sort account LFU retention once in crates/trie/sparse/src/state.rs.
• Update crates/trie/sparse/src/arena/mod.rs and crates/trie/sparse/src/parallel.rs to borrow sorted retained_leaves when possible and only allocate on unsorted fallback inputs.
• Verify with cargo check -p reth-trie-sparse and cargo test -p reth-trie-sparse prune --lib.

[mediocregopher]

derek bench

[github-actions]

Your PR title doesn't follow the Conventional Commit guidelines.

Example of valid titles:

• feat: add new user login
• fix: correct button size
• docs: update README

Usage:

• feat: Introduces a new feature
• fix: Patches a bug
• chore: General maintenance tasks or updates
• test: Adding new tests or modifying existing tests
• bench: Adding new benchmarks or modifying existing benchmarks
• perf: Performance improvements
• refactor: Changes to improve code structure
• docs: Documentation updates
• ci: Changes to CI/CD configurations
• revert: Reverts a previously merged PR
• deps: Updates dependencies

Breaking Changes

Breaking changes are noted by using an exclamation mark. For example:

• feat!: changed the API
• chore(node)!: Removed unused public function

Help

For more information, follow the guidelines here: https://www.conventionalcommits.org/en/v1.0.0/

[decofe]

cc @mediocregopher

✅ Benchmark complete! View job

Benchmark Results

⚠️ Feature is 1 commit behind main. Consider rebasing for accurate results.

Metric	main	auto-opt/20260414-sparse-prune-retained-slices	Change
Mean	34.09ms	34.37ms	+0.81% ⚪ (±3.18%)
StdDev	30.01ms	31.22ms
P50	28.05ms	28.33ms	+1.02% ⚪ (±3.29%)
P90	51.94ms	52.01ms	+0.13% ⚪ (±9.48%)
P99	200.57ms	204.56ms	+1.99% ⚪ (±17.66%)
Mgas/s	969.22	980.48	+1.16% ⚪ (±2.03%)
Wall Clock	14.04s	14.14s	+0.69% ⚪ (±3.13%)

200 blocks, 100 warmup

Wait Time Breakdown

Persistence Wait

Metric	main	auto-opt/20260414-sparse-prune-retained-slices
Mean	27.18ms	28.22ms
P50	0.03ms	0.03ms
P95	307.96ms	305.78ms

Trie Cache Update Wait

Metric	main	auto-opt/20260414-sparse-prune-retained-slices
Mean	0.44ms	0.24ms
P50	0.00ms	0.00ms
P95	2.07ms	1.08ms

Execution Cache Update Wait

Metric	main	auto-opt/20260414-sparse-prune-retained-slices
Mean	0.01ms	0.00ms
P50	0.00ms	0.00ms
P95	0.00ms	0.00ms

Grafana Dashboard

Charts

Latency, Throughput & Diff

Wait Time Breakdown

Gas vs Latency

Grafana Dashboard

View real-time metrics