feat(geth): update go-ethereum version to v1.16.9, enable Osaka hard fork

[XinyuCRO]

• Upgrade go-ethereum from v1.15.11 to v1.16.9
  • go-ethereum branch: https://github.com/crypto-org-chain/go-ethereum/tree/release/1.16
  • rebase all commits of 1.15 branch on 1.16 branch

[codecov]

Codecov Report

❌ Patch coverage is 61.70213% with 18 lines in your changes missing coverage. Please review.
✅ Project coverage is 44.21%. Comparing base (f959eb9) to head (3853b84).
⚠️ Report is 4 commits behind head on develop.

Files with missing lines	Patch %	Lines
x/evm/statedb/statedb.go	16.66%	5 Missing ⚠️
rpc/backend/utils.go	0.00%	2 Missing ⚠️
x/evm/keeper/migrations.go	0.00%	2 Missing ⚠️
x/evm/keeper/set_code_authorizations.go	0.00%	2 Missing ⚠️
x/evm/statedb/statedb_hooked.go	84.61%	2 Missing ⚠️
x/evm/types/chain_config.go	60.00%	1 Missing and 1 partial ⚠️
rpc/types/simulate.go	0.00%	1 Missing ⚠️
rpc/types/types.go	0.00%	1 Missing ⚠️
x/evm/types/msg.go	0.00%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #957      +/-   ##
===========================================
+ Coverage    44.15%   44.21%   +0.06%     
===========================================
  Files          196      197       +1     
  Lines        16896    16927      +31     
===========================================
+ Hits          7461     7485      +24     
- Misses        8645     8652       +7     
  Partials       790      790              

Files with missing lines	Coverage Δ
x/evm/migrations/v8/migrate.go	100.00% <100.00%> (ø)
x/evm/types/tx_args.go	97.87% <100.00%> (ø)
rpc/types/simulate.go	82.78% <0.00%> (ø)
rpc/types/types.go	7.89% <0.00%> (ø)
x/evm/types/msg.go	57.73% <0.00%> (ø)
rpc/backend/utils.go	49.71% <0.00%> (ø)
x/evm/keeper/migrations.go	28.57% <0.00%> (ø)
x/evm/keeper/set_code_authorizations.go	29.62% <0.00%> (ø)
x/evm/statedb/statedb_hooked.go	42.28% <84.61%> (+3.71%)	⬆️
x/evm/types/chain_config.go	58.04% <60.00%> (+0.07%)	⬆️
... and 1 more

🚀 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.

[JayT106]

@claude review PR

[github-actions]

Claude finished @JayT106's task in 7m 35s —— View job

---

PR Review: feat(geth): update go-ethereum version to v1.16.9, enable Osaka hard fork

Tasks:

• Read repository CLAUDE.md for guidelines
• Analyze PR diff and changed files
• Review core changes (go-ethereum upgrade, Osaka hard fork)
• Check for state-breaking changes and consensus impact
• Review new files and migrations
• Review test coverage
• Post final review feedback

---

Summary

This PR upgrades go-ethereum from v1.15.11 → v1.16.9 and enables the Osaka hard fork. Overall the approach is consistent with prior fork integrations (Shanghai → Cancun → Prague → Osaka). Several issues are worth addressing before merging.

---

⚠️ State-Breaking / Consensus-Breaking Changes

This PR is a consensus-breaking upgrade and must be coordinated with an upgrade plan.

In-Place Migration (v7 → v8):

• x/evm/module.go:65 correctly bumps ConsensusVersion() to 8
• x/evm/migrations/v8/migrate.go sets OsakaTime = 0, immediately activating Osaka for all chains on upgrade
• Migration is registered at x/evm/module.go:150 as Migrate7to8
• The upgrade handler at evmd/upgrades.go:26-30 calls RunMigrations, which will trigger the v8 migration

Impact: All validators must upgrade to this version before the sdk53 upgrade height or the chain will fork. This follows the same pattern as Prague/Cancun activation. ✓

---

Issues Found

1. P256VERIFY Test Vector Byte Misalignment — tests/integration_tests/test_call.py

The P256VERIFY_VALID_VECTOR dict has hex strings of incorrect lengths for three fields:

• signature_s: 63 chars (should be 64 = 32 bytes)
• public_key_x: 63 chars (should be 64 = 32 bytes)
• public_key_y: 66 chars (should be 64 = 32 bytes)

The combined total (64+64+63+63+66 = 320 chars = 160 bytes) happens to be the correct total for P256VERIFY_INPUT_LENGTH, so bytes.fromhex("".join(vector.values())) won't raise an error and the length assertion at line 172 will pass. However, after concatenation, the field boundaries are misaligned — the precompile will receive mathematically wrong values for sig_s, pk_x, and pk_y, and signature verification is expected to fail.

The correct hex-encoded 32-byte values should have leading zeros where needed. Fix this →

2. Silently Discarded Error — rpc/backend/utils.go:257

reward, _ := tx.EffectiveGasTip(blockBaseFee)

The error from EffectiveGasTip is silently discarded. While the reward == nil guard below handles the nil case, the error carries information about why the tip is uncalculable (e.g., max fee cap below base fee). This matches the old EffectiveGasTipValue behavior, but consider whether log-level error surfacing would be useful in debug scenarios.

3. GetStateAndCommittedState — Missing HookedStateDB delegation test

statedb_hooked.go:78-80 adds GetStateAndCommittedState but statedb_hooked_test.go doesn't appear to test this new delegation. Given the coverage report flags statedb_hooked.go at 40%, adding a test case for this would be low-effort and improves confidence.

---

Design Observations (Non-blocking)

SetCode hook refactor (statedb_hooked.go:184-198):
The new implementation:

1. Adds tracing.CodeChangeReason parameter — correct propagation of the upstream geth change
2. Uses crypto.Keccak256Hash(prev) instead of ethtypes.EmptyCodeHash for empty prev code — functionally equivalent since keccak("") == EmptyCodeHash
3. Guards with prevHash != codeHash to avoid spurious hook calls on unchanged code — this is a behavior improvement
4. Supports OnCodeChangeV2 hook with reason argument — good forward compatibility

SkipFromEOACheck → SkipTransactionChecks (msg.go:325, tx_args.go:232,327):
This is a go-ethereum v1.16 rename. The new field is semantically broader (covers more than just EOA sender checks) but the boolean usage is identical. ✓

Migration consistency:
v8 migration correctly follows the v7 pattern: read params, set new field to zero (immediate activation), validate, write back. The MigrateStoreEmptyStore test will actually panic (not return an error) because cdc.MustUnmarshal(nil, ...) panics on nil input. The test require.Error(t, err) won't catch a panic — this is a pre-existing pattern from v7, so not introduced here, but worth noting.

---

Test Coverage (from Codecov)

Files below 30% patch coverage for new code:

• x/evm/keeper/migrations.go — 0% (2 lines)
• x/evm/keeper/set_code_authorizations.go — 0% (2 lines, just the SetCode call-site changes)
• rpc/types/simulate.go — 0% (1 line)
• rpc/types/types.go — 0% (1 line)

These are mostly one-liner API-compatibility changes, so the low coverage is understandable. The migration itself has 100% coverage via migrate_test.go. ✓

---

Verdict

The PR is well-structured and follows established patterns. The P256VERIFY test vector alignment bug (point 1) is the most significant issue — the integration test will likely fail or produce incorrect results once Osaka is active. This should be fixed before merge.