Bal devnet 6 by benaadams · Pull Request #11436 · NethermindEth/nethermind

benaadams · 2026-04-30T11:21:18Z

Brings Nethermind in line with the BAL devnet-6 spec: Block Access Lists (EIP-7928), EIP-8037 two-dimensional gas metering (with halt vs revert child-frame restoration split), and the additional Amsterdam EIPs.

Syncs bal-devnet-6 at 5Ggas/s - 10Ggas/s

Changes

Protocol

EIP-7928 (BAL)
- Widen BlockAccessIndex and indexed-change types (BalanceChange, NonceChange, CodeChange, StorageChange) to the spec uint32.
- Replace the legacy -1 "prestate" sentinel with Eip7928Constants.PrestateIndex = uint.MaxValue and a dedicated PrestateAwareIndexComparer so prestate entries grafted by LoadPreStateToSuggestedBlockAccessList keep iteration order.
- Tighten RLP encoding/decoding in AccountChangesDecoder, SlotChangesDecoder, and BlockAccessListDecoder (reject empty storage_changes, reject empty AccountChanges, remap IndexOutOfRangeException/ArgumentOutOfRangeException to RlpException, dispose pooled lists on every decode failure).
- New ValidateBlockLevelAccessListIndexBounds enforces the per-EIP-7928 [0, txCount + 1] index range; ValidateBlockLevelAccessListSize enforces the block_gas_limit / ItemCost cap.
- BlockAccessList.Merge / mutators now correctly invalidate the cached _itemCount. (Align EIP-7928 BlockAccessIndex with uint32 spec, add validation #11362)
- AccountChanges.GetNonce / GetBalance switched from linear scan to binary search; GetCode / GetCodeHash hardened with explicit prestate-loaded invariant guards.
EIP-8037 (2D gas)
- Static cost_per_state_byte = 1174 retained from devnet-3; the value is now threaded through EthereumGasPolicy rather than inlined as a constant. Every state-gas computation goes through GasCostOf.Calculate* helpers (SSetState, CreateState, NewAccountState, CodeDepositState, PerAuthBaseState, PerEmptyAccountState, SSetReversalRefund); matching helper added to RefundOf.
- The parameter is propagated through EvmInstructions.Storage/.Create, CodeDepositHandler, IntrinsicGasCalculator, EthereumGasPolicy, TransactionProcessor, SystemTransactionProcessor, plus the OP / Taiko / Xdc transaction processors.
EIP-8037 child-frame restore on halt vs revert
- RestoreChildStateGas(parent, child, initialReservoir, childStateRefund) keeps the existing semantics on explicit REVERT (state-gas refunds burned at the revert boundary, propagate StateGasSpillBurned for top-level halt accounting).
- New RestoreChildStateGasOnHalt returns the child's full reservoir + accumulated StateGasUsed to the parent and preserves any inline state-gas refunds, matching the spec rule that the failing call chain resets state gas back to the top-most failing call.
- VirtualMachine.UnwindOnError and RevertRefundToHalt updated; new EthereumGasPolicy.ResetForHalt snaps state-gas back to its tx-start shape on top-level halt.
- Regression test Eip8037_exceptional_halt_must_restore_child_inline_state_refund covers the divergence.
- ConsumeStateGas no longer zeroes the reservoir before the regular-dimension UpdateGas fails - on OOG the reservoir is now preserved for the parent to recover.
EIP-8037 per-tx 2D inclusion check - new Eip8037BlockGasInclusionCheck.Validate (mirrors execution-specs PR 2703); applied in the parallel validator before gas accounting so the worker's underlying ParallelExecutionException is not masked by the block-level check. The pre-execution helper is also gated against below-intrinsic state-gas via a Math.Max(0, ...) clamp on worstCaseRegular.
State: introduce IBlockAccessListSource and propagate it through VirtualMachine, VmState, TransactionProcessor, and SystemTransactionProcessor so BAL data is available at the right layer for both prewarm and validation. BlockAccessListBasedWorldState.Restore now correctly forwards transient-storage snapshots (was previously a no-op). TracedAccessWorldState.Caches delegates through _innerWorldState.ScopeProvider as IPreBlockCaches and throws a clear exception when wired against an incompatible inner state.
Amsterdam fork - enable EIP-7976 (calldata floor 16 gas/token) and EIP-7981 in 25_Amsterdam.cs. These two are the "Amsterdam floor pricing" bundle: in ChainSpecBasedSpecProvider they fall back to chainSpec.AmsterdamTimestamp when their per-EIP transition is omitted, so they auto-enable on the Amsterdam timestamp without requiring per-EIP transitions in the chainspec. Behaviour is pinned by Amsterdam_timestamp_enables_bundled_floor_pricing_eips_when_individual_transitions_are_missing and called out inline at the call site so the divergence from every other Amsterdam EIP is obvious.

Post-review feedback

All 21 items from the linked review (high / medium / low priority + Copilot summary) have been addressed, plus a follow-up inline-review pass (12 additional comments) covering naming, comments, and small ergonomic tightenings. Highlights:

Hot-path allocation removal: replaced LINQ SequenceEqual with indexed loops in SlotChanges.Equals / AccountChanges.Equals; dropped [.. item.AccountChanges] in BlockAccessListDecoder.GetContentLength.
Decode robustness: Rlp.DecodeArrayPool<T> disposes pooled results on every failure path (not only range exceptions); BlockAccessListDecoder rejects empty AccountChanges entries up front so they cannot inflate ItemCount.
Defensive correctness: BlockReceiptsTracer.SetReceipt / EndBlockTrace no longer NREs on partial harvests under worker-exception conditions; the size and index-bounds BAL validators now agree on the block.BlockAccessList! precondition.
Spec anchors: added direct EIP-7702 / EIP-8037 / exec-specs PR-2703 references for the StateGasSpillBurned / Reclassified / Refunded accounting split, the inclusion-check asymmetry between regular and state dimensions, and the CREATE init-code-too-large SetOutOfGas / CreditStateGasRefund pairing.
Cleanup: deduplicated GasCostOf.BlockAccessListItem and Eip7928Constants.ItemCost; updated stale bal-devnet-4 references in GasCostOf.cs to bal-devnet-6; reordered [MethodImpl] attributes so they immediately precede their methods on RestoreChildStateGas and RestoreChildStateGasOnHalt.
Follow-up inline review (12 items): restored the inner-exception message in ParallelExecutionException's detail string so the underlying tx rejection reason is preserved; added class-level doc on BlockAccessListValidationIndex and ParentReaderLease; added an inline block comment over ValidateBlockAccessList distinguishing the index-based vs streaming fast-paths from the slow diagnostic path; switched the parent-state-header build in ParallelTxProcessorWithWorldStateManager to object-initializer syntax; renamed [ThreadStatic] t_currentTxTracers to _currentTxTracers in OpcodeBlockTracer to match the codebase's single-underscore field convention; dropped devnet/PR references from EIP-8037 comments; documented why the StorageLane insertion / scratch sort is bespoke (no Array.Sort overload covers three parallel arrays without boxing); restored the Amsterdam floor pricing bundle fallback in ChainSpecBasedSpecProvider with an inline comment naming the test that pins the contract.

Block processing stats

Fixed parallel block-validation ProcessingStats accounting (29e728b398): the per-block trace harvest now reaches EndBlockTrace even when the executor runs transactions in parallel, so the per-block Processed | Blocks xN | Block throughput metric line is accurate for both sequential and parallel paths. The IBlockTracer / CompositeBlockTracer / OpcodeBlockTracer / OpcodeCountingTxTracer surfaces were updated together to keep the trace-collection seam consistent.

Performance optimizations

A six-commit sweep on the parallel BAL hot path. Each lands behind the existing test suite (state, decoder, processor, validator, journal) plus dedicated regressions; functional behaviour is unchanged.

d9916eba Optimize - debox the BAL change journal. Replaced the polymorphic Stack<Change> (struct with object PreviousValue / object PreTxBalance etc.) with parallel typed lanes (List<ChangeType> discriminator + List<BalanceChangeDetails> / NonceChangeDetails / CodeChangeDetails / StorageChangeDetails). Renames Pop{Balance,Nonce,Code}Change to non-nullable TryPop* returning out T so the rollback path no longer round-trips through Nullable<T>. Generic PopChange<T> now constrained where T : struct, IIndexedChange. New NoPreviousBlockAccessIndex sentinel disambiguates "no in-block predecessor" from prestate. New BlockAccessListJournalTests cover the lane invariants.
41bc16ad Super-optimize - drop SortedDictionary from the hot path. BlockAccessList._accountChanges is now Dictionary<Address, AccountChanges> with a lazily-built sorted snapshot (_sortedAccountChanges, invalidated on insert) exposed as ReadOnlySpan<AccountChanges> AccountChangesByAddress. Adds UnorderedAccountChanges for paths that do not need ordering, EnsureCapacity on Merge, a BlockAccessList.FromSortedAccountChanges factory for the decoder fast-path (avoids the dictionary -> sort round-trip), and GenericEqualityComparer.GetOptimized<Address>() for hash-friendly comparison.
e289dfeb Ultra-optimize - replace SortedList<uint, T> with IndexedChanges<T>. Custom lane stores prestate in a dedicated _hasPrestate / _prestate slot and real changes in a dense uint-sorted array. Eliminates PrestateAwareIndexComparer from the read hot path and removes the per-iteration prestate-skip branch from every foreach over balance / nonce / code changes. HasRealBefore and TryGetLastBeforeOrPrestate are O(log n) binary searches over the dense lane (was O(n) linear scan with prestate filter). New IndexedChangeValues<T> typed accessor replaces IList<T> interface dispatch in BAL surface APIs; IndexedChangesJsonConverter keeps wire JSON shape stable. Decoders (IndexedChangeDecoder, simplified AccountChangesDecoder / SlotChangesDecoder / BlockAccessListDecoder) emit straight into the new lane.
c598327c Turbo-optimize - reusable gas-validation handoff. New GasValidationResultSlot (lock + Monitor.PulseAll + ExceptionDispatchInfo) replaces the per-tx TaskCompletionSource<GasValidationResult> worker -> validator handoff. Pre-allocated GasValidationResultSlot[] pool reused across blocks, zero per-block TCS / Task / continuation allocations on the parallel-validation path. New readonly struct GasValidationResult carries (BlockGasUsed, BlockStateGasUsed, IntrinsicGas, Exception) so the validator's EIP-8037 inclusion check uses the worker's pre-computed intrinsic gas instead of recomputing on the main thread. Tightens BlockReceiptsTracer.SetReceipt and CompositeBlockTracer to handle partial-harvest under worker exceptions.
67848bf9 Eldritch-optimize - tx execution reordering. Parallel worker now consumes _txExecutionOrder[] instead of the canonical tx index. The first max(8, ProcessorCount * 2) slots stay in canonical order so the lead window starts executing immediately; the tail is sorted by TxExecutionSortKey to reduce inter-worker contention on shared Address / StorageCell lanes. _perTxBal[] is still keyed on canonical index (txIndex + 1) so the validator sees results in spec order. New CancelIncompleteGasResults ensures a worker-side throw cancels every still-pending GasValidationResultSlot so the validator unblocks and surfaces the original exception.
cbb1ab1b Forbidden optimizations - lookup-time prestate. Stop eagerly mutating the wire BAL with PrestateIndex sentinels read from the live mutable state. BlockAccessListBasedWorldState reads now go through (1) BAL membership, (2) last real change before _blockAccessIndex, (3) a parent-rooted readonly IWorldState fallback. The fallback is built per-tx via PrewarmerEnvFactory + IReadOnlyTxProcessorSource, sharing PreBlockCaches with BlockCachePreWarmer.WarmupFromBal so parallel-IO warmup values feed validation reads and worker misses populate the cache for downstream workers. LoadPreStateToSuggestedBlockAccessList and the _lastLoadedBal guard are deleted. ApplyStateChanges now reads the prior balance directly from stateProvider at the correct tx-body prestate ordering. Pooled ParentReaderLease retires the per-tx readonly scope back to its env pool on processor return. Verified end-to-end against a kurtosis devnet-6 enclave (block 792 -> 3513 monotonic, zero Exception | Invalid Block | Unhandled | Fatal | ERROR | InvalidBlockLevelAccessList signals).

Tests / CI

Bump pyspec fixtures to snobal-devnet-6@v1.1.0 (fixtures_snobal-devnet-6.tar.gz).
Replace per-fork Amsterdam test files with a uniform fixture pattern in Tests.cs; add ArchiveFixtureOverrides hook for archive-version-specific patches.
Expand the Pyspec sharding from 4 -> 16 chunks on ubuntu-latest; the larger fixture set blew the previous 60-min job budget.
Drop the SkipIfNotLinuxX64Ci / SkipIfNotLinuxX64 [SetUp] guards from the abstract Pyspec fixtures (they turned macOS/ARM jobs into "Zero tests ran -> exit 1" noise) and add a single CiSentinelTests.AlwaysPasses fixture so any all-skipped shard still has one passing test.
New regression and unit coverage in Eip8037Tests, Eip8037RegressionTests, Eip8037BlockGasIntegrationTests, Eip8037BlockGasInclusionCheckTests, Eip7928Tests, BlockAccessListItemCountTests, PrestateAwareIndexComparerTests, AccountChangesPrestateTests, BlockAccessListDecoderTests, TransactionProcessorEip4844Tests, BlockValidatorTests, BlockAccessListBasedWorldStateTests, TracedAccessWorldStateTests, BlockReceiptsTracerTests, BlockProcessorTests, BlockhashProviderTests, ChainSpecBasedSpecProviderTests, BlockAccessListJournalTests, BlockCachePreWarmerTests, CompositeBlockTracerTests, plus engine ExecutionPayloadV4Tests.

EELS passing

Devnet smoke verification

The full bundle (protocol + post-review feedback + metrics fix + perf sweep) was exercised against a kurtosis devnet-6 enclave (lighthouse CL + Nethermind EL, fresh genesis, catch-up sync). Latest run reaches block 20491+ (current head) with zero AGENTS.md mandatory error signals (Exception | Invalid Block | Unhandled | Fatal | ERROR); per-block stats line shows ~9000 MGas/s, ~14,900 tps, ~250 Blk/s during catch-up.

Types of changes

New feature (a non-breaking change that adds functionality)
Bugfix (a non-breaking change that fixes an issue)
Build-related changes

Testing

Requires testing

Yes

If yes, did you write tests?

Yes

Notes on testing

Pyspec fixture suite (fixtures_snobal-devnet-6.tar.gz, snobal-devnet-6@v1.1.0) green on Linux x64; macOS/ARM shards now report at least one pass per shard via the sentinel fixture so they no longer fail with "Zero tests ran".
New focused tests cover EIP-8037 gas accounting (including the halt vs revert split), EIP-7928 BAL encoding/decoding/validation, the uint32 index widening, the new per-tx 2D inclusion check, the ConsumeStateGas OOG reservoir preservation invariant, BlockReceiptsTracer partial-harvest robustness, IndexedChanges<T> lane invariants, the BlockAccessList change-journal lanes, the GasValidationResultSlot worker-cancel path, the tx-execution-order sort key, the lookup-time prestate fallback through BlockAccessListBasedWorldState._parentReader, and the Amsterdam floor pricing bundle fallback in ChainSpecBasedSpecProvider.
Devnet-6 kurtosis cold-start (lighthouse + Nethermind) reaches 2700+ blocks with zero exceptions on the live BAL/EIP-8037 hot paths.

Documentation

Requires documentation update

No

) * fix(bal): align EIP-7928 BlockAccessIndex with uint32 spec, add validation Widens BlockAccessIndex from uint16 to uint32 per EIP-7928 (commit 645099785a) and geth bal-devnet-4. Hardens the BAL decoder so truncated/malformed wire bytes produce a clean RlpException at engine_newPayloadV5 instead of crashing. Adds the missing validation rules geth bal-devnet-4 enforces: empty storage_changes per slot is rejected, and BlockAccessIndex is bounded by [0, txCount + 1]. Decoder primitives: - New Rlp.ValueDecoderContext.DecodeUInt() / RlpStream.Encode(uint) / Rlp.LengthOf(uint) helpers. - Rlp.Decode<T> and Rlp.DecodeArrayPool<T> wrap IndexOutOfRangeException / ArgumentOutOfRangeException as RlpException so any truncated-RLP primitive read surfaces consistently. - BlockAccessListDecoder rejects an empty AccountChanges entry (0xc0 inside the outer list) and SlotChangesDecoder rejects an empty StorageChange list for a slot — geth bal-devnet-4 "empty storage writes" parity. Type widening: - IIndexedChange.Index, BalanceChange/NonceChange/CodeChange/StorageChange.Index, BlockAccessList.Index and BlockAccessIndex on the Change journal record all become uint. SortedList<int, T> keys flip to SortedList<uint, T>; ushort query parameters on AccountChanges become uint. Prestate sentinel preservation: - Replaces the legacy -1 int sentinel with Eip7928Constants.PrestateIndex (uint.MaxValue). Adds PrestateAwareIndexComparer that orders the sentinel before all real indices, restoring the iteration semantics that AccountChanges.GetBalance/GetNonce/GetCode/AccountExists and ApplyStateChanges' [^1].Index check rely on. - Decoder-built SortedLists also use the prestate-aware comparer so that LoadPreStateToSuggestedBlockAccessList grafting prestate onto the suggested BAL after decode keeps it sorted first. - Loop predicates that compare change.Key directly against blockAccessIndex explicitly skip PrestateIndex so the raw uint comparison doesn't trigger on the huge sentinel value. Validation: - BlockValidator gains ValidateBlockLevelAccessListIndexBounds enforcing index <= txCount + 1 (mirrors geth's index < txCount + 2 check) with a new BlockLevelAccessListIndexOutOfRange error message. Tests: - New PrestateAwareIndexComparerTests, AccountChangesPrestateTests covering the comparer and prestate-fallback iteration semantics. - BlockAccessListDecoderTests adds: empty-bytes / truncated-outer-list / inner-empty-list throw RlpException; empty storage_changes per slot throw; decoded SlotChanges accepts a later prestate graft as first; BalanceChange round-trips for indices 0x10_0000 and uint.MaxValue. - BlockValidatorTests adds tx-index bound cases (0, 1 valid; 2, uint.MaxValue rejected for a 0-tx block). - ExecutionPayloadV4Tests covers the engine-API decoding-error path for malformed BAL bytes. * style: drop unused usings flagged by IDE0005 CI surfaced four IDE0005 warnings (treated as errors): - BlockAccessListManager: drop `using Nethermind.Crypto;` — Keccak.OfAnEmptySequenceRlp comes from Nethermind.Core.Crypto, already imported. - PrestateAwareIndexComparerTests / AccountChangesPrestateTests: drop `using Nethermind.Core;` — Eip7928Constants resolves via the test's parent namespace Nethermind.Core.Test.BlockAccessLists. - Eip8037Tests: drop `using System;` — no System.* type referenced directly. * fix(bal): record system pre/post-block SSTOREs as reads per EIP-7928 v5.7.0 EIP-7928 v5.7.0 specifies that SSTOREs performed during system pre-block calls (EIP-2935 BlockHashHistory, EIP-4788 BeaconRootContract) and post-block calls (EIP-7002 withdrawal requests, EIP-7251 consolidation requests) are recorded in the BAL as storage reads on the touched slot — not as storage changes with post-values. Same-value writes are skipped entirely. Without this, nethermind generated a BAL whose Keccak256 differs from what eels-built fixtures expect, so the BlockAccessListHash check fails for every Amsterdam block that touches a system contract slot (most pyspec tests). IWorldState gains an opt-in scope: IDisposable? BeginSystemPreBlockScope() TracedAccessWorldState implements it via an int depth counter. While depth > 0, Set(storageCell, value) reclassifies the recording: AddStorageRead when the slot value would change, no-op when unchanged. The state mutation still applies via the inner world state. BlockAccessListManager wraps the three system contract entry points with the scope: StoreBeaconRoot (EIP-4788 system tx), ApplyBlockhashStateChanges (EIP-2935 fast-path Set), and ProcessExecutionRequests (EIP-7002 / EIP-7251 post-block system txs). Parallel-mode state application: In parallel processing, non-system slots wrap stateProvider with BlockAccessListBasedWorldState whose Set is a no-op — actual state mutation relies on ApplyStateChanges replaying the suggested BAL's storage_changes. With the spec-correct BAL containing reads instead of changes for the system slots, ApplyStateChanges has nothing to replay for them, so the canonical state would diverge. TxProcessorWithWorldState gains an `isSystemSlot` parameter. The ParallelTxProcessorWithWorldStateManager passes `isSystemSlot: i == 0 || i == _len - 1` (pre-execution and post-execution slots). For those slots, the TracedAccessWorldState wraps stateProvider directly, so system pre/post-block SSTOREs mutate the canonical state regardless of BAL recording. Tx slots (1..n) keep the BAL-backed wrapping unchanged. Sequential pyspec tests (which is what the Ethereum.Blockchain.Pyspec.Test suite runs) are unaffected by the parallel slot change but benefit from the BAL recording fix; the BlockAccessListHash check now passes for blocks that previously failed solely on system pre-block storage encoding. Note: a residual InvalidStateRoot mismatch remains on a subset of Amsterdam pyspec tests (~70/360 ecrecover_weird_v + a similar slice of stMemoryStress). These were previously masked by the BAL hash error firing first. The remaining state divergence appears unrelated to BAL recording and is left for follow-up. * test(jsonrpc): update Eth_get_block_access_list_by_* fixtures for EIP-7928 v5.7.0 The Eth_get_block_access_list_by_hash and _by_number tests had hardcoded the pre-v5.7.0 shape, recording the EIP-2935 BlockHashHistory system pre-block SSTORE as a storageChanges entry with the post-value. v5.7.0 records system pre-block SSTOREs as storageReads (slot key only). Updated both expected JSON strings to match the new spec-compliant output. * Revert "fix(bal): record system pre/post-block SSTOREs as reads per EIP-7928 v5.7.0" This reverts commit 364f294. * Revert "test(jsonrpc): update Eth_get_block_access_list_by_* fixtures for EIP-7928 v5.7.0" This reverts commit 937ca5d. * review: address PR #11362 feedback - Rlp.DecodeArrayPool<T>: dispose partially-allocated ArrayPoolList<T> before wrapping IndexOutOfRangeException/ArgumentOutOfRangeException as RlpException so the rented buffer is returned to the pool. - Rlp.ValueDecoderContext.DecodeUInt(): collapse case 0 to a single `return RlpHelpers.ThrowNonCanonicalInteger(Position)` (DoesNotReturn, uint) to match DecodeInt and drop the dead `return default`. - BlockAccessListManager.GetPostExecution(): use uint.MaxValue literal to match the uint? balIndex parameter. - AccountChanges.SlotChangesAtIndex: build the returned SortedList with PrestateAwareIndexComparer.Instance so a later prestate graft sorts first, matching the rest of the codebase. - PrestateAwareIndexComparer xmldoc: clarify that decoded BALs also use this comparer (so later prestate grafting preserves order). * test(pyspec): skip EELS bal@v5.7.0 ported_static fixtures with legacy state-test conversion bug EELS's `from_state_test` conversion for ported_static tests omits the EIP-2935 / EIP-4788 system pre-block SSTORE entries from the suggested BAL, while a real client (and Nethermind) executes them — so every such fixture's BlockAccessListHash diverges from what we compute. 91 such tests were the entire residual pyspec failure set on the bal-devnet-4a branch. Detect the conversion via the legacy difficulty value 0x20000 baked into the post-merge mixHash field — real prevRandao would never be exactly 0x...020000 — and Assert.Ignore those tests. Track upstream EELS fix; remove the guard once bal@>v5.7.0 ships with the system pre-block SSTOREs included in the BAL. * fix(pyspec test): drop ?-annotation in non-nullable file CS8632: PyspecTestFixture.cs is not under `#nullable enable`, so the `string?` introduced in 6fb6527 broke the build of every pyspec job. `string` works the same here — the value already gets a null check on the next line. * test(pyspec): also skip blockchain_test_engine_from_state_test variant The first guard only walked test.Blocks, which is null for engine fixtures. Engine fixtures keep the same legacy 0x...020000 sentinel, just on the JSON `prevRandao` field of the engine_newPayload params. Walk EngineNewPayloads too. * Update tests * fix(eip-8037): pin cost_per_state_byte at static 1174 for bal-devnet-4 bal-devnet-4 keeps cost_per_state_byte static at 1174 (carried over from bal-devnet-3), removing the per-block-gas-limit scaling formula that an earlier draft of EIP-8037 specified. snøbal-devnet-4 fixtures encode the same gas usage (63574) at 1M / 5M / 10M / 30M block gas limits, confirming the value is now invariant across blockGasLimit. Reduce CalculateCostPerStateByte to a direct return of CostPerStateByte and drop the dynamic-quantization helpers and BitOperations dependency. The blockGasLimit parameter is retained on call sites in case a future devnet revisits per-block scaling. Update the EIP-8037 unit test to pin the static behaviour rather than asserting quantized values.

commit 3a3078f Author: Marc Harvey-Hill <10379486+Marchhill@users.noreply.github.com> Date: Tue Apr 28 18:23:03 2026 +0200 delete old tests commit 95de888 Merge: 244c2c6 31a673a Author: MarekM25 <marekm2504@gmail.com> Date: Tue Apr 28 18:08:12 2026 +0200 Merge branch 'master' into engine-api-glamsterdam-cleanup commit 244c2c6 Author: MarekM25 <marekm2504@gmail.com> Date: Tue Apr 28 18:05:32 2026 +0200 fix lint commit 9194b8f Author: MarekM25 <marekm2504@gmail.com> Date: Tue Apr 28 18:04:19 2026 +0200 remove tests commit eedfbb7 Author: MarekM25 <marekm2504@gmail.com> Date: Tue Apr 28 17:57:46 2026 +0200 revert formatting changes and zero hash stuff commit dc71aa5 Author: MarekM25 <marekm2504@gmail.com> Date: Tue Apr 28 17:38:42 2026 +0200 cleanups commit ec2fce3 Author: MarekM25 <marekm2504@gmail.com> Date: Tue Apr 28 17:33:11 2026 +0200 fixing commit 90a1da8 Author: MarekM25 <marekm2504@gmail.com> Date: Tue Apr 28 17:15:58 2026 +0200 remove test commit 4451656 Author: MarekM25 <marekm2504@gmail.com> Date: Tue Apr 28 17:12:32 2026 +0200 Cleanup mess commit 2425748 Merge: 4a90460 a36154c Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Mon Apr 27 14:33:36 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit 4a90460 Merge: 3549768 18d60a4 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Mon Apr 27 14:08:25 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit 3549768 Merge: 30fcc36 b71c352 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Mon Apr 27 14:07:39 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit 30fcc36 Merge: 540e4a2 ed6a354 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Mon Apr 27 13:09:48 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit 540e4a2 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sun Apr 26 04:52:49 2026 +0300 minor fixes commit 010547b Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sun Apr 26 03:22:39 2026 +0300 test fixes commit 6fcc136 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sun Apr 26 02:11:12 2026 +0300 fixes commit 0bccb06 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sun Apr 26 01:52:02 2026 +0300 test fix commit 830c578 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sun Apr 26 01:38:55 2026 +0300 fix tests commit 1f5fd2c Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sun Apr 26 01:35:17 2026 +0300 claude review again commit d87491f Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sun Apr 26 00:57:57 2026 +0300 fixes commit dd431e2 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sun Apr 26 00:51:52 2026 +0300 fix tests commit 71dc99f Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sat Apr 25 22:40:50 2026 +0300 fixes commit 9c03d12 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sat Apr 25 22:32:58 2026 +0300 update per 786 commit a4c4f3c Merge: fc49e7e 9cba44c Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Sat Apr 25 21:38:25 2026 +0300 Merge remote-tracking branch 'origin/master' into engine-api-glamsterdam commit fc49e7e Merge: bca0c63 42c551f Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Mon Apr 20 16:14:33 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit bca0c63 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Mon Apr 20 16:14:10 2026 +0300 remove unused import commit 3d2c973 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Mon Apr 20 16:09:17 2026 +0300 fix taiko tests commit b30ef5c Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Mon Apr 20 15:34:35 2026 +0300 fix tests commit 6f131cd Merge: a8c884a 0fe41f4 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Mon Apr 20 14:55:46 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit a8c884a Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Mon Apr 20 14:54:40 2026 +0300 conflicts fix commit 96eb912 Merge: 82a1f70 425a2a3 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Mon Apr 20 14:52:33 2026 +0300 Merge remote-tracking branch 'origin/master' into engine-api-glamsterdam # Conflicts: # src/Nethermind/Nethermind.Merge.Plugin/BlockTreeExtensions.cs # src/Nethermind/Nethermind.Merge.Plugin/Handlers/ForkchoiceUpdatedHandler.cs # src/Nethermind/Nethermind.Taiko/Rpc/TaikoForkchoiceUpdatedHandler.cs commit 82a1f70 Merge: 7b6133f 436c65b Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Mon Apr 20 13:16:55 2026 +0300 Merge remote-tracking branch 'origin/master' into engine-api-glamsterdam commit 7b6133f Merge: 2f068ec 02c2026 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Thu Apr 16 14:49:46 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit 2f068ec Merge: bdd7ee5 5464c8b Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Thu Apr 16 14:10:02 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit bdd7ee5 Merge: f09cb41 bfaaeb0 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Wed Apr 15 16:45:19 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit f09cb41 Merge: 370acdf cf03d18 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Wed Apr 15 16:24:17 2026 +0300 Merge remote-tracking branch 'origin/master' into engine-api-glamsterdam commit 370acdf Merge: d1d0370 6cd0ea4 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Mon Apr 13 14:28:21 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit d1d0370 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Fri Apr 10 17:30:54 2026 +0300 more fixes commit c99d6cc Merge: 4d0f215 a52cb90 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Fri Apr 10 17:24:06 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit 4d0f215 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Fri Apr 10 17:23:42 2026 +0300 claude review commit cb6defe Merge: af63142 0057bd8 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Fri Apr 10 12:34:37 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit af63142 Merge: 1b338f3 2f24891 Author: Stavros Vlachakis <89769224+svlachakis@users.noreply.github.com> Date: Fri Apr 10 12:08:36 2026 +0300 Merge branch 'master' into engine-api-glamsterdam commit 1b338f3 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Fri Apr 10 12:08:13 2026 +0300 claude review commit 619259b Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Thu Apr 9 18:32:59 2026 +0300 taiko fix commit d067793 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Thu Apr 9 18:19:14 2026 +0300 cleaner design commit 1e27be1 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Thu Apr 9 18:12:04 2026 +0300 improve tests matching the specs commit a77e182 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Thu Apr 9 18:04:58 2026 +0300 improve test commit 4ba069c Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Thu Apr 9 18:01:59 2026 +0300 fix comment commit fe6f9b0 Author: stavrosvl7 <stavrosvl7@gmail.com> Date: Thu Apr 9 17:53:57 2026 +0300 engine api changes as per ethereum/execution-apis#770 ethereum/execution-apis#760

benaadams · 2026-05-08T08:16:51Z

@claude re-review

claude · 2026-05-08T08:17:13Z

Claude finished @benaadams's task in 4m 27s —— View job

PR Re-Review: BAL devnet-6 (Round 10)

Read rule files and gather context
Check new commits since Round 9 (46922f8, 7763e42, 376ea15, 688ae94)
Check outstanding findings from Round 9 (L1, L2)
Review new code changes
Post final findings

Summary

Four commits since Round 9. All are refactoring/dead-code removal with no new correctness issues. The "Kamil fix" revert restores the two-call pattern in EvmInstructions.Call.cs (separate TryGetDelegation + GetCachedCodeInfo rather than a combined call), which is safer given the different repository instances involved. The "Reduce code" commits eliminate dead methods, consolidate duplicated logic, and simplify internals without changing semantics.

New Commits

46922f8 fix — Tiny: changes string? _testChunk to string _testChunk in TestChunkFilterTests.cs. ✅

7763e42 Revert "Kamil fix" — Reverts 7954eaa. The combined GetCachedCodeInfo(codeSource, spec, out Address? delegated) overload exists, but the revert restores the two-call pattern (TryGetDelegation via vm.TxExecutionContext.CodeInfoRepository then GetCachedCodeInfo via vm.CodeInfoRepository). The two calls use the same underlying repository instance in practice but the original shape is more explicit about intent. ✅

376ea15 Reduce code changes — Removes verbose XML doc comments from IGasPolicy<TSelf> (replacing <summary> blocks with brief inline comments), trims dead code from AccountChanges (CheckWasChanged no-op, AccountChanged alias for HasStateChanges, ToDictionary, SeedSortedStorageCaches), and condenses Eip8037BlockGasInclusionCheckTests. Pure reduction, no semantic change. ✅

688ae94 Reduce code — Larger refactor across 17 files:

CompleteEip8037Halt extracted (TransactionProcessor.cs): The duplicated EIP-8037 halt sequence in FailContractCreate and Refund is now a shared [AggressiveInlining] private method. The comments about using postIntrinsicStateReservoir are now in one place. Semantically identical to the prior inline form. ✅
BuildUp snapshot drop (TransactionProcessor.cs:39-40): The IntrinsicGas<TGasPolicy> overload of Process no longer calls WorldState.TakeSnapshot(true) on BuildUp. The non-generic Process(Transaction, ITxTracer, ExecutionOptions) retains the snapshot (line 125-128). The generic overload is only called by the BAL parallel path which passes Commit, not BuildUp, so this is correct. ✅
AddressIndex.GetOrAdd via CollectionsMarshal (BlockAccessListValidationIndex.cs): CollectionsMarshal.GetValueRefOrAddDefault is a single-pass lookup+insert that avoids a double hash. Address implicitly converts to AddressAsKey (line 292 of Address.cs). The old _addresses reverse-lookup array and GetAddress were removed (no production callers). ✅
TryGetRow simplification (BlockAccessListValidationIndex.cs): Always sets row = (int)index, returns index <= lastIndex. All call sites use if (TryGetRow(..., out row)) { use_row; } so the unconditional set is never observed on false return. ✅
BlockAccessListValidationIndex constructor now requires non-null suggested; CreateMutableEmpty factory removed. The generated index is always constructed from a suggested index (only call site: BlockAccessListManager.PrepareForProcessing:103). ✅
HasRequiredReadAccountMissing inlined into RegisterGeneratedDelta: The static method is replaced by an inline loop with _hasGeneratedRequiredReadAccountMismatch = true; break;, gated by !_hasGeneratedRequiredReadAccountMismatch. Early-exit semantics preserved. ✅
Dead-code removal: AccountChanges.GetAllSlots (no callers), BlockAccessList.RemoveAccountChanges (no callers), CheckPerTxInclusion without pre-computed intrinsic (all callers now use the pre-computed overload). ✅
*ChangeDetails → readonly record struct: Private journal structs now use synthesized Equals/GetHashCode/ToString. They are stored in List<T> and accessed via CollectionsMarshal.AsSpan by ref — the synthesized members are never called, so this is cosmetic-only overhead. ✅

Prior Findings — Status

Finding	Status
L1 `_completed.Wait()` no cancellation	Still present — `finally _completed.Set()` guarantees forward progress; catastrophic-fault path only. Not blocking.
L2 `ParallelExecutionException` message duplication	Still present — cosmetic only.

All Medium and High findings from all prior rounds remain resolved.

Critical: 0 | High: 0 | Medium: 0 | Low: 2 (both acknowledged, non-blocking)

The PR is in the same mergeable state as Round 9. The "Reduce code" sweep is purely additive in terms of quality — cleaner, shorter, no regressions introduced. Ready to merge.

LukaszRozmej · 2026-05-08T08:46:48Z

I think ideas that should be easy wins: 1, 2, 3, 4, 5, 6, 9, 12, 14, 15,
I think ideas that need profiling to be considered: 8 (Frozen Dictionary constructed in background), 11, 17, 18, 20, 21, 22, 24, 25, 30
I think ideas that are probably too much of a hassle: 7, 10, 13, 16, 19, 26, 27, 28, 29,

codecov · 2026-05-08T09:50:20Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 0.00%. Comparing base (db8b6e9) to head (644c447).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@             Coverage Diff             @@
##           master   #11436       +/-   ##
===========================================
- Coverage   74.43%        0   -74.44%     
===========================================
  Files        2831        0     -2831     
  Lines      114624        0   -114624     
  Branches    16572        0    -16572     
===========================================
- Hits        85324        0    -85324     
+ Misses      26377        0    -26377     
+ Partials     2923        0     -2923

☔ 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.

github-actions · 2026-05-08T10:10:41Z

EXPB Benchmark Comparison

Run: View workflow run

superblocks

Scenario: nethermind-flat-superblocks-bal-devnet-6-delay0s

Client Processing (SSE)

Metric	PR	Master (cached)	Delta
AVG (ms)	1256.77	1257.63	-0.07%
MEDIAN (ms)	1129.0	1164.9	-3.08%
P90 (ms)	1722.9	1736.8	-0.80%
P95 (ms)	2251.4	2133.2	+5.54%
P99 (ms)	3969.8	3433.0	+15.64%
MIN (ms)	804.6	783.9	+2.64%
MAX (ms)	3969.8	3433.0	+15.64%

K6 TTFB

Metric	PR	Master (cached)	Delta
AVG (ms)	1484.92	1484.13	+0.05%
MEDIAN (ms)	1320.47	1367.54	-3.44%
P90 (ms)	1955.12	1990.19	-1.76%
P95 (ms)	2714.95	2532.57	+7.20%
P99 (ms)	3578.91	3437.59	+4.11%
MIN (ms)	929.36	920.70	+0.94%
MAX (ms)	4108.63	3602.27	+14.06%

realblocks

Scenario: nethermind-flat-realblocks-bal-devnet-6-delay0s

Client Processing (SSE)

Metric	PR	Master (cached)	Delta
AVG (ms)	35.16	35.87	-1.98%
MEDIAN (ms)	28.7	29.5	-2.71%
P90 (ms)	64.3	65.7	-2.13%
P95 (ms)	76.4	76.7	-0.39%
P99 (ms)	124.0	124.1	-0.08%
MIN (ms)	1.6	1.5	+6.67%
MAX (ms)	252.6	229.4	+10.11%

K6 TTFB

Metric	PR	Master (cached)	Delta
AVG (ms)	41.08	41.44	-0.87%
MEDIAN (ms)	34.23	35.55	-3.71%
P90 (ms)	73.23	73.44	-0.29%
P95 (ms)	83.30	83.73	-0.51%
P99 (ms)	133.84	128.79	+3.92%
MIN (ms)	2.59	2.26	+14.60%
MAX (ms)	257.57	259.92	-0.90%

benaadams · 2026-05-08T14:18:03Z

EELS passing

Marchhill · 2026-05-08T14:02:10Z

        }

-        private TxReceipt[] ProcessTransactionsParallel(Block block, ProcessingOptions processingOptions, CancellationToken token)
+        private TxReceipt[] ProcessTransactionsParallel(Block block, ProcessingOptions processingOptions, BlockReceiptsTracer outerReceiptsTracer, CancellationToken token)


I agree, could postpone such refactoring changes to a later PR though to prevent even more conflicts

- BlockAccessListManager: restore inner-exception message in ParallelExecutionException detail; drop devnet/PR refs from EIP-8037 comment; add fast/slow path explanation; switch BlockHeader build to object initializer; add ParentReaderLease doc comment. - BlockAccessListValidationIndex: add class-level summary; document why the multi-array sort is bespoke. - OpcodeBlockTracer: rename t_currentTxTracers to _currentTxTracers to match Nethermind's underscore-prefix field convention. - ChainSpecBasedSpecProvider: drop inconsistent AmsterdamTimestamp fallback for Eip7976/Eip7981 (other Amsterdam EIPs don't have it). - GethGenesisLoader: revert accidental ordering swap.

EIP-7976 and EIP-7981 are the "Amsterdam floor pricing" bundle: they auto-enable on the Amsterdam timestamp without requiring per-EIP transitions in the chainspec. Behaviour is covered by Amsterdam_timestamp_enables_bundled_floor_pricing_eips_when_individual_transitions_are_missing. My earlier removal in 79a1678 treated the fallback as inconsistent with the other Amsterdam EIPs and broke that test. Restoring the fallback with a code comment so the rationale is visible inline.

benaadams · 2026-05-08T15:57:17Z

@claude re-review

claude · 2026-05-08T15:57:37Z

Claude finished @benaadams's task in 3m 22s —— View job

PR Re-Review: BAL devnet-6 (Round 11)

Read rule files and gather context
Check new commits since Round 10 (644c447, 3bd79ab, 79a1678, 2b8b649)
Review Marchhill's approval comments and benaadams responses
Check benchmark results
Post final findings

Summary

Four commits since Round 10. All are clean.

New Commits

644c447 dedupe tests — PyspecTestFixture.cs refactored: introduces PyspecBlockchainFixtureBase(bool parallel, bool batchRead, bool heavy) as the shared root; PyspecEngineBlockchainTestFixture / PyspecSyncBlockchainTestFixture / PyspecAmsterdamBlockchainTestFixture all collapse to one-line () primary-constructor delegates — removes ~44 duplicate lines of [SetUp], protected override bool?, and attribute boilerplate. Clean reduction. ✅

3bd79ab Fix bad merge — HeadersSyncFeed.BuildNewBatch / ProcessPersistedPortion rework to handle the race where HeadersDestinationNumber advances above _lowestRequestedHeaderNumber mid-sync: StartNumber = Math.Max(HeadersDestinationNumber, ...) and RequestSize = Math.Min(headersLeft, requestSize) clamp correctly; the null return is eliminated (always returns a batch whose RequestSize may be 0 when the destination is already reached). ProcessPersistedPortion is widened to non-nullable. This is sync infrastructure, not BAL-related; the fix is conservative and ShouldBuildANewBatch() in PrepareRequest's outer guard prevents entering the build path when the destination is already reached. ✅

79a1678 Address PR review comments — Addresses all of Marchhill's open items:

ParallelExecutionException detail string restored to $"Parallel execution failure: {innerException.Message}" — inner rejection reason is now preserved. ✅
Devnet/PR refs dropped from EIP-8037 comment; replaced with plain "EIP-8037" reference. ✅
Fast/slow path block comment added over ValidateBlockAccessList explaining the index-based vs streaming fast-paths vs slow diagnostic path. ✅
BlockHeader build switched to object-initializer syntax in CreateParentStateHeader. ✅
ParentReaderLease class doc added (RAII wrapper, pooled source, returns on Dispose). ✅
BlockAccessListValidationIndex class-level summary added covering Build vs mutable ctor, column-oriented lane layout, and _hasOutOfRangeChange semantics. ✅
SortRow bespoke sort comment added explaining why Array.Sort is not used (no 3-parallel-array overload without boxing). ✅
OpcodeBlockTracer: t_currentTxTracers → _currentTxTracers at all 5 sites. Matches the codebase convention (_threadStaticBuffer in Hash256.cs, _reusableWriteBatch in DbOnTheRocks.cs). ✅
ChainSpecBasedSpecProvider: Amsterdam fallback removed (then re-added in next commit — see below). ✅
GethGenesisLoader: accidental ordering swap reverted. ✅

2b8b649 Restore Amsterdam floor pricing EIP fallback with explanation — 79a1678 inadvertently broke Amsterdam_timestamp_enables_bundled_floor_pricing_eips_when_individual_transitions_are_missing by removing the ?? chainSpec.AmsterdamTimestamp fallback for EIP-7976/7981. This commit restores it with a three-line inline comment explaining the "Amsterdam floor pricing bundle" special case (auto-enables on Amsterdam timestamp without per-EIP transitions). The commit message and comment are precise about why this diverges from every other Amsterdam EIP. ✅

Prior Findings — Status

Finding	Status
L2 (Round 10) `ParallelExecutionException` message duplication	✅ Fixed — `WithDetail($"Parallel execution failure: {innerException.Message}")`
L1 (Round 10) `_completed.Wait()` no cancellation	Still present — `finally _completed.Set()` guarantees forward progress; catastrophic-fault path only. Non-blocking.

All Medium and High findings from all prior rounds remain resolved.

Benchmark Results

Metric	Superblocks	Realblocks
AVG delta	-0.07%	-1.98%
Median delta	-3.08%	-2.71%
P90 delta	-0.80%	-2.13%
P99 delta	+15.64%	-0.08%

Superblocks P99 shows +15.64% (tail variance) but median is -3.08% — consistent with single-run statistical noise at tail percentiles. Realblocks are uniformly flat or slightly improved. No performance regression.

Critical: 0 | High: 0 | Medium: 0 | Low: 1 (acknowledged, non-blocking)

All high and medium findings across all eleven review rounds are resolved or explicitly acknowledged with rationale. Devnet-6 smoke run (20k+ blocks, zero exception/invalid-block signals), EELS passing, extensive unit/regression/integration coverage, and flat-to-slightly-improved benchmarks. The PR is ready to merge.

…hmarks bal-devnet-6 (#11436) is now in master, so this brings everything from master including the BAL devnet-6 PR plus subsequent master commits. Conflict resolution: - Took ours for every BAL-touching file (decoders, manager, validator, BlockAccessListBased/TracedAccess world states, parallel-loop executor, cache pre-warmer, BAL tests). Our branch's ReadOnly*/Generated*/*AtIndex split diverges substantially from upstream's unified BlockAccessList, and upstream merging bal-devnet-6 into master pulled in the unified version. - Removed (again) BlockAccessListValidationIndex + tests, BlockAccessList, AccountChanges, SlotChanges, BlockAccessListItemCountTests, and BlockAccessListJournalTests: all reference the removed unified type. - Removed the now-broken Rlp.Encode(BlockAccessList) overload in Rlp.cs; callers use the generic Rlp.Encode<T> path for ReadOnly/Generated BALs. - Took theirs for TxReceiptConverter (upstream's boxing-free ForcedNumberConversion.Value refactor, already includes our EIP-7778/8037 diagnostic fields) and ChainSpecBasedSpecProvider (comment-only addition). - Took ours for OpcodeBlockTracer (kept t_ prefix on the [ThreadStatic] field, our existing convention). - For .github/workflows/nethermind-tests-flat.yml took upstream's unconditional pre-build + 'dotnet build-server shutdown || true' approach (matches their CI peak-memory PR) while keeping our extra Cache NuGet packages step. - Updated EngineModuleTests.V6 ResultWrapper generic args from IEnumerable to IReadOnlyList to match upstream's API change.

## Summary - The `nethermindeth/nethermind#master master-bad-iszero` image build has been failing because `nevermind/bad-iszero.patch` no longer applies cleanly to upstream master ([failing run](https://github.com/ethpandaops/eth-client-docker-image-builder/actions/runs/25737190830/job/75577551462)). - Upstream Nethermind PR [NethermindEth/nethermind#11436](NethermindEth/nethermind#11436) (commit `481b6b17cc`) replaced the per-file `using Word = Vector256<byte>;` aliases with a project-wide MSBuild `<Using ... Alias=\"EvmWord\" />` in `src/Nethermind/Directory.Build.props`, renaming all `Word` symbols to `EvmWord`. The underlying type (`Vector256<byte>`) is unchanged, so `value.AsUInt64()` keeps working. - This bumps the patch's context/signature from `Word Operation(Word value)` to `EvmWord Operation(EvmWord value)`. ## Test plan - [x] Dry-run `patch -p 1` against current upstream `nethermindeth/nethermind` master — applies cleanly. - [x] Applied patch and confirmed `OpIsZero.Operation` now reads with `EvmWord` types and the bad-iszero hack body intact. - [ ] Confirm the next scheduled `master-bad-iszero` build (linux-amd64 + linux-arm64) succeeds end-to-end.

Adopts master's parallel-execution architecture (PR #11436) for parent-state reads while keeping our split BAL-type design: * Parallel workers now rent an IReadOnlyTxProcessingScope from a pooled IReadOnlyTxProcessingEnvFactory, scoped against the parent state root captured at PrepareForProcessing. BlockAccessListBasedWorldState falls through to that snapshot reader for any (address, slot) the suggested BAL doesn't carry at the current block-access index. * BAL completeness is preserved: reads for an account that isn't declared in the suggested BAL at all throw InvalidBlockLevelAccessListException with the same message format the sequential validator produces from ValidateBlockAccessList. * Parallel execution now requires stateProvider.IsInScope so a parent state root is always capturable; sequential path takes over otherwise. Drops prestate-loading machinery in favour of the parent reader: * LoadPreStateToSuggestedBlockAccessList + per-account TaskCompletionSource gate (EnablePrestateGate / SignalPrestateLoaded / WaitForPrestate) are gone — workers no longer block on a per-account gate before reading prestate-dependent state. * ReadOnlyAccountChanges drops LoadPreStateBalance/Nonce/Code/Storage, GetSlotsForPreStateLoad, RecordWasChanged, ExistedBeforeBlock, and EmptyBeforeBlock; ReadOnlySlotChanges drops LoadPreStateChange. * Adds HasStateChanges, IsStorageRead, and TryGetLast{Balance,Nonce,Code} ChangeBefore convenience APIs the new world-state read paths need. * IndexedChanges keeps its prestate-slot storage and the wire-level PrestateIndex sentinel guard in IndexedChangeDecoder — both are dormant at runtime now but still defend against a malicious peer. The slot-0 loader iteration in BlockProcessor.ParallelBlockValidationTransactionsExecutor disappears (only ApplyStateChanges remains in iteration 0); slot 1 keeps the pre-execution StoreBeaconRoot / ApplyBlockhashStateChanges step and continues to synchronize with IncrementalValidation via preExecutionDoneTcs. Test updates: * BlockAccessListBasedWorldStateTests + the relevant TracedAccessWorldState test now wire up a parent reader (the inner state itself, scoped against genesis) and stop populating BAL prestate sentinels — covered values live in the parent state instead. * Removes AccountChangesPrestateTests and ReadOnlyAccountChangesTests entirely (every test exercised a deleted API). * Restores the parametrized PrepareForProcessing_keeps_parallel_bal_execution_for_validated_eip8037_blocks test and adds back the PrepareForProcessing_disables_parallel_bal_execution_when_state_provider_is_not_scoped test now that the IsInScope gate is in place. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

github-actions Bot added configuration evm optimism state+storage xdc labels Apr 30, 2026

benaadams force-pushed the bal-devnet-6 branch from 5483434 to 0e57cb6 Compare April 30, 2026 11:44

flcl42 and others added 9 commits April 30, 2026 13:49

Add EIP-8037 and Amsterdam fixes

a42f63b

Run all tests

c3b5089

fix(test): match TCS Exception type to IncrementalValidation signature

c025025

fix test

f85bc70

fixes

73720bd

fix

f55b9e8

Fixes

711ad58

lint

78a7e24

benaadams force-pushed the bal-devnet-6 branch from 0e57cb6 to 78a7e24 Compare April 30, 2026 11:49

benaadams added 3 commits April 30, 2026 14:32

update tests

10b7db0

fix tests

a86fc07

fix ci

f4f4e2c

benaadams added the skip-evm-benchmarks label Apr 30, 2026

benaadams added 2 commits April 30, 2026 16:26

remove skips

47bacb8

More pyspec test chunks

c9bd335

github-actions Bot added the devops label Apr 30, 2026

github-actions Bot added the taiko related to the taiko alethia rollup label Apr 30, 2026

benaadams added 3 commits April 30, 2026 17:38

update tests

306e63e

remove skips

dd84988

halt changes

2c6329f

github-actions Bot added bug fix + reliability build changes labels Apr 30, 2026

benaadams added 2 commits May 8, 2026 08:15

Reduce code changes

376ea15

Reduce code

688ae94

dedupe tests

644c447

kamilchodola added reproducible-benchmark labels May 8, 2026

kamilchodola added reproducible-benchmark and removed reproducible-benchmark labels May 8, 2026

Fix bad merge

3bd79ab

kamilchodola mentioned this pull request May 8, 2026

Bal devnet 6 benchmarks #11511

Closed

16 tasks

Merge branch 'master' into bal-devnet-6

39cffc2

Marchhill approved these changes May 8, 2026

View reviewed changes

benaadams added 2 commits May 8, 2026 16:29

LukaszRozmej approved these changes May 8, 2026

View reviewed changes

benaadams merged commit 481b6b1 into master May 8, 2026
970 of 976 checks passed

benaadams deleted the bal-devnet-6 branch May 8, 2026 19:16

benaadams restored the bal-devnet-6 branch May 8, 2026 19:16

benaadams mentioned this pull request May 8, 2026

Fix parallel execution metrics and opcode tracing #11445

Closed

7 tasks

barnabasbusa mentioned this pull request May 12, 2026

Fix nevermind bad-iszero patch for EvmWord rename ethpandaops/eth-client-docker-image-builder#378

Merged

3 tasks

Conversation

benaadams commented Apr 30, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes

Protocol

Post-review feedback

Block processing stats

Performance optimizations

Tests / CI

Devnet smoke verification

Types of changes

Testing

Requires testing

If yes, did you write tests?

Notes on testing

Documentation

Requires documentation update

Uh oh!

benaadams commented May 8, 2026

Uh oh!

claude Bot commented May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Re-Review: BAL devnet-6 (Round 10)

Summary

New Commits

Prior Findings — Status

Uh oh!

LukaszRozmej commented May 8, 2026

Uh oh!

codecov Bot commented May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

github-actions Bot commented May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

EXPB Benchmark Comparison

superblocks

realblocks

Uh oh!

benaadams commented May 8, 2026

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Marchhill May 8, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

benaadams commented May 8, 2026

Uh oh!

claude Bot commented May 8, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Re-Review: BAL devnet-6 (Round 11)

Summary

New Commits

Prior Findings — Status

Benchmark Results

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

benaadams commented Apr 30, 2026 •

edited

Loading

claude Bot commented May 8, 2026 •

edited

Loading

codecov Bot commented May 8, 2026 •

edited

Loading

github-actions Bot commented May 8, 2026 •

edited

Loading

claude Bot commented May 8, 2026 •

edited

Loading