core, eth, params: merge geth v1.17.6 (v1.17.6 sync)

[pratikspatil024]

Important

Status, 2026-10-01: #2345, #2346, #2354 and #2455 have all merged into upstream-merge-v1.17.4, and this is the last open PR of the sync. It targets the base directly and is no longer part of a GitHub stack. The base was merged in with 0348dbfcb, which keeps this branch's tree unchanged: GitHub had rebased #2455, so the base's #2455 commits are copies of ones this branch already has, and the only difference (the switch-dispatch generator) must stay in this branch's v1.17.6 form. The diff is v1.17.6 only again. Merge with a merge commit. Never rebase this branch (no "Update branch" with rebase); it carries upstream authors' commits. The notes below describe the earlier stacked layout.

Reviewer guide — stacked PR 13 of 13. Part of the combined go-ethereum v1.17.4 + v1.17.5 + v1.17.6 upstream sync, which ships as one stable release. v1.17.6 joins it because upstream schedules the fork this work builds toward (Amsterdam, on Sepolia) in v1.17.6. Every PR in the stack merges into the base branch upstream-merge-v1.17.4; that base merges into develop once, at the very end — not per-PR.

Merge-commit only — never squash. Squashing rewrites a branch's SHAs and breaks every PR stacked above it.

Review bottom-up: #2308, #2319, #2325, #2328, #2337, #2340, #2341, #2342 and #2343 have merged into the base. The open stack is #2345 → #2346 → #2354 → #2455 → this PR. Start at #2345 — every PR above inherits it, so reviewing top-down means re-reviewing.

Review fixes and cascade, 2026-09-29: three findings from Marcello's review were fixed where the code lives and cascaded up. #2346 f2fde2692: the switch-dispatch interpreter (runSwitch, on by default on mainnet BPs) now charges EIP-8037 state gas like Run, with a new Amsterdam differential test. #2354 0166dcc9d: the pebble v1 path, which every existing bor database opens through, keeps bor's tuning and metrics (new databases use pebble v2; upgrade command tracked in POS-3746). #2354 aac7dd01e: V2's FinaliseFast settles EIP-8246 self-destructs like serial Finalise. Cascade: #2354 256e10c39 → #2455 d96fcc81b → #2461 440ed1ac4; at #2455 and #2461 the generator mirrors that level's renamed Run helpers. All dormant until Amsterdam except the pebble fix, which restores today's behaviour for existing nodes. Every hop builds and passes the dispatch/V2/parity tests, and the witness regeneration tests pass on the top of the chain.

Review follow-up and cascade, 2026-09-30: per Jerry's review on #2346 (49aa12b54), both BlockSTM processors run their pre-execution system calls through the serial PreExecution instead of inline copies (no behaviour change there). Merged up into #2354 (ab0921f62), that closes a dormant gap: #35285 had moved the EIP-7997 factory insert (first Amsterdam block) into PreExecution, which V1/V2 didn't call, so they would have disagreed with serial on the root at activation; TestV2PreExecEIP7997Activation covers it. Cascade: #2354 ab0921f62 → #2455 c27e0f29a → #2461 caeadc89b, where upstream's #35458 drops the insert altogether and the test goes with it. Every hop builds and passes the parity/V2 tests, and the witness regeneration tests pass on the top of the chain.

Expected-red / flaky checks (not code blockers): Quality metrics (diffguard — skipped by team decision; it also mis-scopes across a stacked diff), and codecov/patch (a diff that is mostly upstream code carried in verbatim). The kurtosis stateless leg's producer-rotation threshold flakes on develop too and is re-run by hand. Full per-batch conflict-resolution reasoning is in docs/upstream-merges/.

---

Stacked PR — base is ppatil-upstream-eip2780-8037 (#2455), not develop.
Merge with a merge commit. Never squash. GitHub may flag conflicts from its
rebase-based stack view; this sync is merge-based by design. Do not use
"Resolve conflicts" or gh stack rebase — both would rewrite the branch.

Summary

Merges go-ethereum v1.17.6 into bor, completing the upstream sync. 136
upstream first-parent commits (v1.17.5..v1.17.6), taken in 8 batches, each
a single GPG-signed merge commit whose resolution reasoning is recorded in the
ledger. The upstream v1.17.6 tag (3d84c6b2e) is an ancestor of this branch.

Upstream range	v1.17.5..v1.17.6, 136 first-parent commits
Batches	8 (1, 2, 3a, 3b, 4, 5, 6, 7), all GPG-signed, plus three merges of #2455 to pick up its fixes
Net diff vs base	290 files (+8,657 / −3,656); 60 are byte-identical to geth v1.17.6
Conflicts resolved	284 across the milestone (b1 32, b2 36, b3a 30, b3b 44, b4 47, b5 22, b6 20, b7 53)
New hardforks enabled	none — AmsterdamBlock stays nil on every Bor surface

Documentation lives in docs/upstream-merges/v1.17.6/:

• plan.md — batch boundaries, deviations, ordered follow-ups
• ledger.md — every resolution, per batch, with reasoning
• fork-register.md — per-fork/EIP dormancy verification
• needs-wiring.md — everything intentionally not adopted, and what adopting it would take

Consensus-relevant items, for the reviewer's attention first

StateDB takes params.Rules instead of fork booleans (#35498, batch 3a).
Finalise, IntermediateRoot, Commit and CommitWithUpdate changed
signature, and every Bor call site was translated so behaviour is unchanged:

• sites that know the block build config.Rules(num, false, time), with
  isMerge false because Bor's Difficulty can be nil;
• sites that passed a literal true use the EVM's rules, since EIP-158 is
  active from genesis on every Bor network;
• a literal false becomes params.Rules{}.

CommitSnapshot keeps its bool signature. TestV2ForkParity classifies the
new rules field.

Receipt hashing in a pipeline (#35738, batch 7) — declined. Upstream hashes
the receipt trie and bloom in a background pipeline fed from inside the tx
loop, and the validator reads its digest. On Bor, Finalize appends the
post-Madhugiri state-sync receipt after the loop, and BlockSTM assembles its
own results. A pipeline fed only from the loop would miss receipts and fail the
receipt-root check on import. needs-wiring.md records what a safe port needs.

Gas tracing rework (#35646, batch 6) — tracing only, gas results unchanged.
The new UsedExecutionGas -= coldCost in the EIP-2929 CALL variant (live on
Bor) doesn't reach GasBudget.Used() or any receipt, block or refund
computation. PIP-88 twin lockstep: gasSLoadPIP88 and
makeGasSStoreFuncPIP88 are unchanged, their upstream originals are unchanged
in the batch, and every shared function's divergence from upstream is the same
before and after the merge.

A Bor tracing bug was caught and fixed (batch 6). After #35646, muxTracer
exposes only the V2 depth hooks. Bor's WrapStateSyncHooks wrapped the V1
fields only, so debug_traceBlock* with muxTracer on a block with a
state-sync tx failed ("incorrect number of top-level calls"). The wrapper now
exposes V2 and forwards via the Emit* helpers, keeping a per-frame entry-gas
stack for V1 inner tracers.

Fork validation in the state transition (#35588, batch 5) — adopted.
Access-list, blob and set-code messages are rejected before their fork. Bor's
state-sync and system calls never go through preCheck, and canonical history
can't contain those types before their fork.

Amsterdam on Sepolia (#35734, batch 7) — declined. Bor's Sepolia config has
no time-based forks. BorMainnetChainConfig, AmoyChainConfig, the chain
presets and the genesis JSON are untouched.

Every upstream fork and EIP in this range is merged dormant, verified in
fork-register.md. Nothing here changes block production or validation on
mainnet or Amoy.

Declined, each recorded in needs-wiring.md

• BAL-driven parallel execution (#35264 and follow-ups), BAL download and replay
  (#35386, #35423 replay half), engine-API BAL fixes (#35580, #35664).
• The sparse blob pool, blob fetcher and buffer (#35367, #35429, #35572, #35589,
  #35766), eth/71 and eth/72 devp2p tests, snap v2 (#35705), #35533 parallel
  snap/1 processing.
• ExecuteConfig tracer activation (#35512), otel spans in the parallel
  processor (#35693, which adds ctx to ApplyTransaction).
• The async block write in writeBlockWithState (#35595 write half — the
  function also writes witnesses), the downloader scheduling rework (#35680) and
  download/snap progress metrics (#35678, #35689).
• The precompile-cache disable flag (#35753) — the cache (#35388) is adopted but
  not wired.

Ported by hand

• #35299 snap storage-range abort and #35722 zero-origin no-proof account
  range, both into Bor's pre-split eth/protocols/snap Syncer.
  TestSyncFullAccountRangeNoProof fails without #35722's fix.

Executed tests

Per batch (all eight):

• go build ./..., go vet ./... — only the two known lock-copy findings.
• make test contract (-p 1 -short, excluding cmd/) — 144/144 packages
  at every batch after its fixes.
• TestBorHardforkPrecompileContinuity, TestReinforceMultiClientPreCompilesTest,
  TestV2ForkParity, TestPDBMethodParity — pass.
• TestV2WitnessRegeneration{AllBlocks,PipelinedSRCAllBlocks,PipelinedSRCChained},
  non-short — pass; Chained 222/222.

Milestone:

• BOR_BLOCKSTM_TEST=1 TestV2BlockSTMAllBlocks — 238/241 blocks consistent, the
  same as the baseline.
• Full cmd/ tier — only the known pre-existing failures: cmd/geth (5),
  cmd/evm (4) and the cmd/devp2p/internal/ethtest nil-BorConfig panic.

Local kurtosis devnet (kurtosis-pos v1.4.2, CI's pipeline-e2e topology:
3 witness-producing validators with pipelined import, 2 stateless validators,
no-witness / witness / stateless RPC nodes, and a released
0xpolygon/bor:latest baseline, plus anvil L1 and bridge/tx spammers), about
1,980 blocks:

• Health probes pass before and after fault injection: all containers running,
  finalized hashes agree on all 9 nodes (including the released baseline),
  milestones, checkpoints and state-sync advancing; lastStateId identical on
  every full-state node.
• CI's stateless suite passes every test: block-hash consensus, post-veBlop
  HF, fast-forward sync (124-block gap caught up in 16 s), milestone latency
  under injected latency, load with producer rotation, extreme-latency
  recovery, the L1→L2 bridge and producer restart recovery. The rotation
  threshold flaked once, as it does on develop, and passed on re-run.
• CI's pipeline suite passes: pipelined SRC on every full-sync node with 0
  root mismatches, ~880 witnesses per witness-producing node, the pipeline
  self-gated off on stateless nodes, and the released baseline agrees.
• debug_traceBlockByNumber with muxTracer and callTracer on a state-sync
  block: no errors, output byte-identical to the released image.

Accepted-red CI checks

• Quality metrics (diffguard) — skipped by team decision.
• codecov — expected red on a merge of this shape.

Rollout notes

• Not consensus-affecting as merged. Every new fork and EIP is dormant; no
  gate is flipped and no activation height is set.
• No coordinated upgrade required for this PR on its own — it is one link in
  a stack that ships as a single stable release.
• Operator-visible RPC changes:
  • eth_getHeaderByNumber("pending") returns null, as does a safe /
    finalized tag that can't be resolved (#35627).
  • A null for a required non-pointer argument returns invalid argument N
    (#35576).
  • debug_traceCall's block parameter is optional and defaults to latest
    (#35583).
  • eth_call / eth_estimateGas reject blob or set-code arguments without
    to (#35695).
  • Calls carrying an access list, blob hashes or authorizations against a
    block before the matching fork now fail (#35588).
  • eth_createAccessList no longer rejects up front a call with more
    authorizations than its gas covers (#35698).
• Dependencies: otel 1.46.0, with go mod tidy raising the indirects it
  requires. The Go toolchain stays at 1.26.8.
• version/version.go now reads 1.17.6 / stable, matching upstream's
  release commit.

Follow-ups owed on top of this milestone

• POS-3738 (Amsterdam blockers): Chicago's/Lisovo's instruction sets shadow
  Amsterdam's on real Bor configs. The EIP-8037 tests now run with Bor = nil;
  core/vm/eip8038_test.go, core/eip7708_test.go and core/eip8246_test.go
  are held back; TestEstimateGasAmsterdam and the Amsterdam subtest of
  TestCreateAccessListAuthorizationGas are skipped ("missing head header").
• POS-3737 (BAL): everything the BAL declines above would need.
• POS-3739: the miner has no state.Error() check before sealing (upstream
  #35427); also on develop.

🤖 Generated with Claude Code

[pratikspatil024]

@claude review

[Copilot]

Copilot review overview

🟡 Changes recommended

Discovery RPC input validation, ambiguous payload-detail matching, and release toolchain inconsistencies remain unresolved.

Review effort: Balanced
Findings: 3 · 2

Open (5)

• Reject negative node counts before ReadNodes · New
• Reject negative node counts before ReadNodes · New
• Match immutable block fields when associating bad payloads · New
• Build RPM sources with the documented Go toolchain · New
• Align release jobs with the documented Go toolchain · New

What changed in this PR

Completes the Bor sync with go-ethereum v1.17.6, incorporating upstream state, RPC, networking, tracing, storage, tooling, and release changes while keeping Amsterdam dormant.

Changes:

• Migrates state finalization APIs to params.Rules and updates Bor execution paths.
• Adds upstream RPC, discovery, database, tracing, and client fixes.
• Updates release metadata, dependencies, fixtures, and deprecated network/configuration surfaces.

File	Description
version/​version.go	Bumps version to 1.17.6.
triedb/​pathdb/​metrics.go	Corrects a metric namespace.
tests/​state_test.go	Adapts state-root calls to rules.
tests/​state_test_util.go	Propagates fork rules through state tests.
tests/​block_test_util.go	Adds slot handling and numeric base-fee comparison.
rpc/​websocket.go	Reads WebSocket frames atomically.
rpc/​http.go	Adds framed HTTP body reading.
rpc/​client_test.go	Delays fault injection until handshake completion.
params/​bootnodes.go	Removes Holesky discovery configuration.
p2p/​server.go	Uses the shared discovery UDP connection.
p2p/​pipes/​pipe.go	Closes listeners after dial failure.
p2p/​discover/​v5_udp_test.go	Uses fallback UDP configuration.
p2p/​discover/​common.go	Exposes shared UDP connection support.
miner/​worker.go	Adds payload-reversion metadata fields.
miner/​pipeline.go	Supplies fork rules when deriving roots.
miner/​payload_building.go	Retains receipts and construction details.
metrics/​runtimehistogram.go	Fixes negative histogram maximum handling.
metrics/​metered_lru.go	Adds a metered LRU cache.
internal/​web3ext/​web3ext.go	Updates debug RPC extensions.
internal/​ethapi/​transaction_args.go	Validates recipient-less blob/set-code calls.
internal/​ethapi/​testdata/​eth_getTransactionReceipt-with-logs.json	Refreshes RPC fixture hashes.
internal/​ethapi/​testdata/​eth_getTransactionReceipt-state-sync-tx.json	Refreshes state-sync receipt fixture.
internal/​ethapi/​testdata/​eth_getTransactionReceipt-normal-transfer-tx.json	Refreshes transfer receipt fixture.
internal/​ethapi/​testdata/​eth_getTransactionReceipt-dynamic-tx-with-logs.json	Refreshes dynamic-fee fixture.
internal/​ethapi/​testdata/​eth_getTransactionReceipt-create-contract-with-access-list.json	Refreshes contract receipt fixture.
internal/​ethapi/​testdata/​eth_getTransactionReceipt-create-contract-tx.json	Refreshes creation receipt fixture.
internal/​ethapi/​testdata/​eth_getTransactionByHash-with-logs.json	Refreshes transaction lookup fixture.
internal/​ethapi/​testdata/​eth_getTransactionByHash-state-sync-tx.json	Refreshes state-sync lookup fixture.
internal/​ethapi/​testdata/​eth_getTransactionByHash-normal-transfer-tx.json	Refreshes transfer lookup fixture.
internal/​ethapi/​testdata/​eth_getTransactionByHash-dynamic-tx-with-logs.json	Refreshes dynamic transaction fixture.
internal/​ethapi/​testdata/​eth_getTransactionByHash-create-contract-with-access-list.json	Refreshes access-list lookup fixture.
internal/​ethapi/​testdata/​eth_getTransactionByHash-create-contract-tx.json	Refreshes creation lookup fixture.
internal/​ethapi/​testdata/​eth_getTransactionByBlockHashAndIndex-state-sync-tx.json	Refreshes indexed state-sync fixture.
internal/​ethapi/​testdata/​eth_getTransactionByBlockHashAndIndex-create-contract-with-access-list.json	Refreshes indexed creation fixture.
internal/​ethapi/​testdata/​eth_getHeaderByNumber-tag-pending.json	Expects null pending headers.
internal/​ethapi/​testdata/​eth_getHeaderByNumber-tag-latest.json	Refreshes latest-header fixture.
internal/​ethapi/​testdata/​eth_getHeaderByNumber-number-latest-1.json	Refreshes previous-header fixture.
internal/​ethapi/​testdata/​eth_getHeaderByNumber-number-1.json	Refreshes numbered-header fixture.
internal/​ethapi/​testdata/​eth_getHeaderByNumber-number-0.json	Refreshes genesis-header fixture.
internal/​ethapi/​testdata/​eth_getHeaderByHash-hash-latest.json	Refreshes hash-header fixture.
internal/​ethapi/​testdata/​eth_getHeaderByHash-hash-latest-1.json	Refreshes previous hash-header fixture.
internal/​ethapi/​testdata/​eth_getHeaderByHash-hash-1.json	Refreshes numbered hash-header fixture.
internal/​ethapi/​testdata/​eth_getHeaderByHash-hash-0.json	Refreshes genesis hash-header fixture.
internal/​ethapi/​testdata/​eth_getBlockReceipts-tag-pending.json	Refreshes pending-receipts fixture.
internal/​ethapi/​testdata/​eth_getBlockReceipts-tag-latest.json	Refreshes latest-receipts fixture.
internal/​ethapi/​testdata/​eth_getBlockReceipts-block-with-legacy-transfer-tx.json	Refreshes legacy receipt fixture.
internal/​ethapi/​testdata/​eth_getBlockReceipts-block-with-legacy-contract-call-tx.json	Refreshes contract-call receipt fixture.
internal/​ethapi/​testdata/​eth_getBlockReceipts-block-with-dynamic-fee-tx.json	Refreshes dynamic-fee receipt fixture.
internal/​ethapi/​testdata/​eth_getBlockReceipts-block-with-contract-create-tx.json	Refreshes creation receipt fixture.
internal/​ethapi/​testdata/​eth_getBlockReceipts-block-with-blob-tx.json	Refreshes blob receipt fixture.
internal/​ethapi/​testdata/​eth_getBlockByNumber-tag-pending.json	Refreshes pending-block fixture.
internal/​ethapi/​testdata/​eth_getBlockByNumber-tag-pending-fullTx.json	Refreshes full pending-block fixture.
internal/​ethapi/​testdata/​eth_getBlockByNumber-tag-latest.json	Refreshes latest-block fixture.
internal/​ethapi/​testdata/​eth_getBlockByNumber-number-1.json	Refreshes numbered-block fixture.
internal/​ethapi/​testdata/​eth_getBlockByNumber-number-0.json	Refreshes genesis-block fixture.
internal/​ethapi/​testdata/​eth_getBlockByHash-hash-latest.json	Refreshes latest block-by-hash fixture.
internal/​ethapi/​testdata/​eth_getBlockByHash-hash-genesis.json	Refreshes genesis block-by-hash fixture.
internal/​ethapi/​testdata/​eth_getBlockByHash-hash-1.json	Refreshes numbered block-by-hash fixture.
internal/​ethapi/​simulate.go	Uses EVM rules during simulation finalization.
internal/​ethapi/​override/​override.go	Updates override finalization and difficulty semantics.
internal/​ethapi/​api.go	Updates header tags, access lists, and debug finalization.
internal/​cli/​server/​config.go	Retains ignored state-size configuration compatibility.
internal/​build/​util.go	Adds nonfatal command execution support.
graphql/​schema.go	Exposes slot and block-access-list fields.
graphql/​graphql.go	Resolves block access-list hashes.
ethdb/​memorydb/​memorydb.go	Distinguishes empty-key and range deletion.
ethdb/​dbtest/​testsuite.go	Tests empty-key batch deletion.
ethclient/​simulated/​backend.go	Installs required system contracts.
ethclient/​ethclient.go	Validates withdrawals and missing transaction counts.
ethclient/​ethclient_test.go	Adds system contracts to test genesis.
eth/​tracers/​logger/​logger_json.go	Enforces trace limits for call frames.
eth/​tracers/​js/​tracer_test.go	Updates execution-gas assertions.
eth/​tracers/​internal/​tracetest/​supply_test.go	Adds system contracts to supply tests.
eth/​tracers/​internal/​tracetest/​selfdestruct_state_test.go	Adds system contracts to tracer fixtures.
eth/​state_accessor.go	Commits reconstructed state using rules.
eth/​protocols/​snap/​sync.go	Supports proofless zero-origin ranges.
eth/​protocols/​snap/​handler.go	Correctly aborts bounded storage queries.
eth/​protocols/​eth/​handler.go	Updates network documentation.
eth/​gasestimator/​gasestimator.go	Returns exact reduced transfer gas.
eth/​filters/​api.go	Validates ranges beginning at genesis.
eth/​fetcher/​tx_fetcher.go	Clarifies transaction-size enforcement.
eth/​ethconfig/​gen_config.go	Removes state-size TOML wiring.
eth/​ethconfig/​config.go	Removes state-size runtime configuration.
eth/​catalyst/​simulated_beacon.go	Applies Bor’s block-based Amsterdam gate.
eth/​catalyst/​queue.go	Looks up local payload construction details.
eth/​catalyst/​api_test.go	Deploys required system contracts.
eth/​backend.go	Removes state-size trie configuration.
crypto/​kzg4844/​kzg4844_gokzg.go	Adds Go KZG cell recovery.
crypto/​kzg4844/​kzg4844_ckzg_nocgo.go	Adds the unsupported no-CGO stub.
crypto/​kzg4844/​kzg4844_ckzg_cgo.go	Adds C KZG cell recovery.
crypto/​bn256/​cloudflare/​twist.go	Preserves projective state during negation.
crypto/​bn256/​cloudflare/​bn256_test.go	Tests negative G2 pairing behavior.
core/​vm/​jump_table.go	Renames regular gas to execution gas.
core/​vm/​interpreter_test.go	Supplies rules during finalization.
core/​vm/​interface.go	Changes StateDB finalization to accept rules.
core/​vm/​instructions_test.go	Updates execution-gas assertions.
core/​vm/​gas_table_test.go	Updates finalization rules in gas tests.
core/​vm/​evm_precompile_cache_test.go	Updates precompile gas assertions.
core/​vm/​eips.go	Uses execution-gas accounting.
core/​vm/​dispatch_test.go	Updates dispatch finalization and gas fields.
core/​vm/​dispatch_bench_test.go	Updates benchmark finalization rules.
core/​vm/​contracts_fuzz_test.go	Adapts precompile invocation signature.
core/​v2_witness_regen_test.go	Propagates rules through witness regeneration.
core/​v2_serial_parity_gen_test.go	Updates generated-state commits.
core/​v2_serial_parity_fuzz_test.go	Updates parity root derivation.
core/​v2_selfdestruct_self_beneficiary_test.go	Updates self-destruct test finalization.
core/​v2_pre_exec_system_call_test.go	Updates pre-execution state commits.
core/​v2_metamorphic_parity_test.go	Updates metamorphic parity finalization.
core/​v2_exist_nonce_test.go	Updates nonce parity commits.
core/​v2_blockstm_test.go	Updates BlockSTM test commits.
core/​v1_differential_test.go	Updates V1 differential root handling.
core/​types/​state_account.go	Adds buffered slim-account RLP encoding.
core/​txpool/​locals/​journal.go	Reports journal close failures.
core/​txpool/​legacypool/​list_test.go	Updates txpool state finalization.
core/​tracing/​gen_gas_change_reason_stringer.go	Adds new gas-change reason strings.
core/​stateless.go	Derives stateless roots using fork rules.
core/​state/​witness_readset_test.go	Updates witness fixture commits.
core/​state/​witness_prewalk_test.go	Updates prewalk fixture commits.
core/​state/​v2_executor_differential_test.go	Updates executor parity finalization.
core/​state/​v2_differential_test.go	Updates differential state commits.
core/​state/​trie_prefetcher_test.go	Updates prefetcher state commits.
core/​state/​sync_test.go	Updates state-sync fixture commits.
core/​state/​statedb_hooked.go	Passes rules through hooked finalization.
core/​state/​statedb_hooked_test.go	Updates hooked-state tests.
core/​state/​statedb_fuzz_test.go	Updates fuzzed finalization calls.
core/​state/​state_test.go	Updates state test commits.
core/​state/​safe_base_test.go	Updates flat-diff commits.
core/​state/​reader_test.go	Updates cache-reader fixture commits.
core/​state/​parallel_statedb.go	Adapts parallel StateDB finalization.
core/​state/​parallel_statedb_coverage_test.go	Updates parallel finalization coverage.
core/​state/​parallel_read_equivalence_test.go	Updates read-equivalence finalization.
core/​state/​amsterdam_witness_skew_test.go	Updates Amsterdam witness fixture commits.
core/​state_transition_test.go	Updates Amsterdam gas expectations.
core/​state_processor_test.go	Adds system contracts to processor tests.
core/​state_prefetcher.go	Uses EVM rules for prefetched roots.
core/​state_prefetcher_intermediate_root_test.go	Updates prefetch commit rules.
core/​rawdb/​freezer.go	Handles independent freezer group tails.
core/​rawdb/​freezer_table.go	Resets tables truncated below their tail.
core/​rawdb/​database.go	Propagates iterator errors.
core/​rawdb/​chain_freezer.go	Closes freezer after era DB failure.
core/​rawdb/​accessors_chain_test.go	Tests canonical ancient access-list lookup.
core/​parallel_state_processor.go	Propagates rules through parallel execution.
core/​parallel_state_processor_hooks_parity_test.go	Extends V2 hook parity coverage.
core/​parallel_state_processor_fork_parity_test.go	Updates rules-based fork expectations.
core/​mainnet_witness_benchmark_test.go	Updates benchmark root finalization.
core/​genesis_test.go	Removes Holesky genesis coverage.
core/​gen_genesis.go	Hex-encodes genesis slot numbers.
core/​forkid/​forkid_test.go	Removes Holesky fork-ID cases.
core/​filtermaps/​map_renderer.go	Handles chain-shortening reorgs.
core/​filtermaps/​checkpoints.go	Removes Holesky checkpoints.
core/​eth_transfer_logs_test.go	Adds system contracts to transfer tests.
core/​chain_makers_test.go	Adds system contracts to PoS chains.
core/​blockchain_test.go	Updates commits and shared genesis allocation.
core/​block_validator.go	Validates state roots using rules.
consensus/​misc/​eip7997.go	Removes the superseded factory transition.
consensus/​ethash/​consensus.go	Finalizes Ethash state with rules.
consensus/​clique/​clique.go	Finalizes Clique state with rules.
consensus/​bor/​statesync_gas_budget_test.go	Updates state-sync root calls.
consensus/​bor/​statesync_execution_parity_test.go	Updates state-sync parity commits.
consensus/​bor/​statefull/​processor.go	Finalizes Bor calls with EVM rules.
consensus/​bor/​bor.go	Uses rules for Bor root derivation.
consensus/​bor/​bor_test.go	Updates Cancun commit rules.
consensus/​beacon/​consensus.go	Finalizes beacon state with rules.
common/​mclock/​simclock.go	Renames timer callback storage.
common/​lru/​blob_lru_test.go	Tests key-size accounting.
cmd/​utils/​flags_legacy.go	Retains a deprecated state-size flag.
cmd/​keeper/​go.mod	Updates indirect dependencies.
cmd/​geth/​main.go	Marks state-size flag deprecated.
cmd/​geth/​consolecmd_test.go	Moves console tests from Holesky to Hoodi.
cmd/​geth/​config.go	Accepts the deprecated config field.
cmd/​geth/​chaincmd.go	Removes Holesky history import selection.
cmd/​evm/​testdata/​evmrun/​9.out.2.txt	Relaxes timing output matching.
cmd/​evm/​testdata/​35/​txs.json	Adds an empty transaction fixture.
cmd/​evm/​testdata/​35/​env.json	Adds a missing-base-fee environment.
cmd/​evm/​testdata/​35/​alloc.json	Adds an empty allocation fixture.
cmd/​evm/​t8n_test.go	Tests missing post-London base fee.
cmd/​evm/​staterunner.go	Updates diagnostic root derivation.
cmd/​evm/​runner.go	Updates state dump commits.
cmd/​evm/​main.go	Updates dump root derivation.
cmd/​evm/​internal/​t8ntool/​transition.go	Updates transition allocation roots.
cmd/​devp2p/​README.md	Documents Hoodi and combined discovery.
cmd/​devp2p/​nodesetcmd.go	Removes Holesky node filtering.
cmd/​devp2p/​main.go	Adds shared discovery RPC serving.
cmd/​devp2p/​discv5cmd.go	Adds discovery-v5 RPC methods.
cmd/​devp2p/​discv4cmd.go	Refactors discovery-v4 RPC serving.
cmd/​blsync/​main.go	Removes Holesky CLI support.
build/​deb/​ethereum/​deb.install	Installs relocated completion files.
build/​completions/​zsh/​_geth	Adds zsh completion support.
build/​completions/​bash/​geth	Adds bash completion support.
beacon/​params/​networks.go	Removes Holesky light-client configuration.
beacon/​params/​checkpoint_holesky.hex	Removes the Holesky checkpoint.
accounts/​abi/​type.go	Normalizes contract ABI types as addresses.
accounts/​abi/​bind/​v2/​lib.go	Ignores mismatched watched events.
accounts/​abi/​bind/​v2/​lib_test.go	Tests mismatched event handling.
accounts/​abi/​abigen/​source2.go.tpl	Uses original ABI error names.
accounts/​abi/​abigen/​source.go.tpl	Skips mismatched generated events.
accounts/​abi/​abigen/​bindv2_test.go	Tests normalized error names.
.gitea/​workflows/​release.yml	Changes release build toolchains.
.gitea/​workflows/​release-copr.yml	Adds COPR release publishing.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[claude]

Beyond the inline findings, I also checked eth/catalyst/api.go's bad-block-recording path for NewPayload import failures (around lines 912-933): the locally-built block looked up by state root shares the same hash as the payload that failed to import in this flow, so recording it under that hash is not a distinct bug.

Extended reasoning...

This is the tail of a large (290-file, +8657/-3656) upstream go-ethereum v1.17.6 merge into Bor, touching EVM/VM internals, state processing, consensus, RPC, and P2P. Three confirmed bugs were already surfaced as inline findings (two of them compile-breaking: a stale StateDB.Commit call signature in tests/bor/helper.go, and a broken core/vm/gen_dispatch generator), which alone means a human still needs to look at this PR. I additionally examined eth/catalyst/api.go's bad-block-recording path on NewPayload failure for a hash-mismatch concern and confirmed the looked-up local block and the failing payload share the same hash in that flow, so it is not a separate issue worth raising.

Additional findings (outside the current diff — GitHub can't attach inline comments there):

• 🔴 tests/bor/helper.go — tests/bor/helper.go:310 still calls the old state.Commit(block, deleteEmptyObjects bool, noStorageWiping bool) signature, but this diff changes StateDB.Commit to Commit(rules params.Rules, block uint64) everywhere else. Building with the integration tag now fails to compile since the args (uint64, bool, bool) no longer match. Every test using buildNextBlock (32 call sites in tests/bor/*_test.go) is unbuildable. Fix: update this call to state.Commit(chain.Config().Rules(b.header.Number, false, b.header.Time), block.NumberU64()), matching the rest of the merge's translation of Commit call sites.

  Why this was flagged

  tests/bor/helper.go:310 has state.Commit(block.NumberU64(), chain.Config().IsEIP158(b.header.Number), false), calling *state.StateDB.Commit with (uint64, bool, bool). core/state/statedb.go changed StateDB.Commit's signature to Commit(rules params.Rules, block uint64) (common.Hash, error) (core/state/statedb.go:188 area per diff) as part of this v1.17.6 merge. helper.go carries the go:build integration tag and was not in this diff's changed-files list, so it was missed during the signature migration. Every other Commit call site in the repo (core/genesis.go, core/blockchain.go, eth/state_accessor.go, etc.) was updated to the new signature; this one was not. Anyone building/running the bor integration test suite (go build -tags integration ./tests/bor/... or go test -tags integration ./tests/bor/...) gets a compile error, whereas on the base branch this compiled and ran fine.

  Verification: normal (test build regression). The diff changes *state.StateDB.Commit to a 2-argument signature: core/state/statedb.go:2522 func (s *StateDB) Commit(rules params.Rules, block uint64) (common.Hash, error). tests/bor/helper.go:310 still invokes the old 3-argument form: root, err := state.Commit(block.NumberU64(), chain.Config().IsEIP158(b.header.Number), false), passing (uint64, bool,…

• 🔴 core/vm/gen_dispatch/main.go — Maintainers who run go generate ./core/vm/ (per the //go:generate go run ./gen_dispatch/ in core/vm/interpreter.go:19) get a regenerated interpreter_dispatch.go that fails to compile. This diff renames GasBudget/GasCosts fields RegularGas->ExecutionGas (and UsedRegularGas->UsedExecutionGas) everywhere, including the committed core/vm/interpreter_dispatch.go, but the code generator that produces that exact file, core/vm/gen_dispatch/main.go, still emits contract.Gas.RegularGas and dynamicCost.RegularGas (lines 256, 259, 417, 420, 442, 445), fields that no longer exist. Fix: rename these template strings to ExecutionGas too, covering all 6 occurrences. build/ci.go's check_generate step (go generate ./... …

  Why this was flagged

  …then diff) will also fail CI once run, since it wasn't part of this PR's per-batch go build/go vet testing.

  The trigger is any go generate ./core/vm/ invocation, e.g. by a contributor adding/editing an opcode, or by build/ci.go's doCheckGenerate (build/ci.go:493-533) which runs go generate ./... and fails the build if generated output differs from committed files. core/vm/gen_dispatch/main.go emits Go source referencing contract.Gas.RegularGas and dynamicCost.RegularGas (lines 256,259,417,420,442,445), but GasBudget/GasCosts (core/vm/gascosts.go) were renamed to ExecutionGas by this diff, and the committed core/vm/interpreter_dispatch.go already uses ExecutionGas. Regenerating overwrites it with code that won't compile, and check_generate flags the mismatch either way. The PR's own testing notes only ran go build ./.../go vet ./..., not go generate, so this wasn't caught.

  Verification: normal. This diff renamed GasCosts/GasBudget fields RegularGas->ExecutionGas (core/vm/gascosts.go) and updated the committed generated file core/vm/interpreter_dispatch.go (now uses contract.Gas.ExecutionGas, e.g. lines 62-65, 320-386), but did NOT touch the generator core/vm/gen_dispatch/main.go (empty git diff --stat for that path). gen_dispatch/main.go still emits the old field names verbatim…

[pratikspatil024]

Addressed in c11cdaa:

• tests/bor/helper.go (integration build tag): the last StateDB.Commit call is converted to the params.Rules form. I used params.Rules{IsEIP158: chain.Config().IsEIP158(b.header.Number)} rather than config.Rules(...), because the old call passed a literal false for the Cancun flag, and this keeps behaviour identical. All integration-tagged packages build and vet clean.
• core/vm/gen_dispatch/main.go: the RegularGas → ExecutionGas rename is applied to Bor's generator. go generate ./core/vm/ now reproduces the committed interpreter_dispatch.go byte for byte.
• CI lint (golangci-lint v2.11.4, 0 issues locally): restored the "Updated payload" log in miner/payload_building.go, which the merge had dropped along with upstream's BAL-only balhash attribute. Also removed the miner environment's unused revertedTxs/revertedIdx fields and fixed the params import grouping in 15 test files.

The remaining inline findings are upstream verbatim and are answered on their threads.

[codecov]

Codecov Report

❌ Patch coverage is 61.65933% with 610 lines in your changes missing coverage. Please review.
✅ Project coverage is 56.62%. Comparing base (ada1fc7) to head (0348dbf).
⚠️ Report is 1 commits behind head on upstream-merge-v1.17.4.

Files with missing lines	Patch %	Lines
core/tracing/hooks.go	0.00%	52 Missing ⚠️
core/rawdb/accessors_chain.go	36.11%	42 Missing and 4 partials ⚠️
ethdb/dbtest/testsuite.go	0.00%	45 Missing ⚠️
core/vm/operations_verkle.go	0.00%	37 Missing ⚠️
eth/api_debug.go	7.89%	33 Missing and 2 partials ⚠️
core/vm/evm.go	27.50%	27 Missing and 2 partials ⚠️
eth/catalyst/api.go	0.00%	26 Missing ⚠️
core/vm/operations_acl.go	54.71%	20 Missing and 4 partials ⚠️
metrics/metered_lru.go	0.00%	24 Missing ⚠️
core/types/state_account.go	0.00%	21 Missing ⚠️
... and 48 more

❌ Your patch check has failed because the patch coverage (61.65%) is below the target coverage (90.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@                    Coverage Diff                     @@
##           upstream-merge-v1.17.4    #2461      +/-   ##
==========================================================
+ Coverage                   56.42%   56.62%   +0.20%     
==========================================================
  Files                         944      944              
  Lines                      174836   175181     +345     
==========================================================
+ Hits                        98645    99194     +549     
+ Misses                      70158    69997     -161     
+ Partials                     6033     5990      -43     

Files with missing lines	Coverage Δ
accounts/abi/bind/v2/lib.go	57.98% <100.00%> (+2.42%)	⬆️
accounts/abi/type.go	86.48% <100.00%> (+0.04%)	⬆️
common/mclock/simclock.go	92.37% <100.00%> (ø)
consensus/beacon/consensus.go	10.03% <100.00%> (ø)
consensus/bor/bor.go	86.62% <100.00%> (ø)
consensus/clique/clique.go	41.05% <100.00%> (ø)
consensus/ethash/consensus.go	36.95% <100.00%> (ø)
core/block_validator.go	59.73% <100.00%> (ø)
core/chain_makers.go	70.60% <100.00%> (+0.78%)	⬆️
core/filtermaps/checkpoints.go	50.00% <ø> (ø)
... and 92 more

... and 23 files with indirect coverage changes

Files with missing lines	Coverage Δ
accounts/abi/bind/v2/lib.go	57.98% <100.00%> (+2.42%)	⬆️
accounts/abi/type.go	86.48% <100.00%> (+0.04%)	⬆️
common/mclock/simclock.go	92.37% <100.00%> (ø)
consensus/beacon/consensus.go	10.03% <100.00%> (ø)
consensus/bor/bor.go	86.62% <100.00%> (ø)
consensus/clique/clique.go	41.05% <100.00%> (ø)
consensus/ethash/consensus.go	36.95% <100.00%> (ø)
core/block_validator.go	59.73% <100.00%> (ø)
core/chain_makers.go	70.60% <100.00%> (+0.78%)	⬆️
core/filtermaps/checkpoints.go	50.00% <ø> (ø)
... and 92 more

... and 23 files with indirect coverage changes

🚀 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.

[claude]

Claude Code Review

This repository is configured for manual code reviews. Comment @claude review for a one-time review, or @claude review always to subscribe this PR to a review on every future push.

_{Tip: disable this comment in your organization's Code Review settings.}