[flcl42]

Changes

• Schedule the Amsterdam fork on Hoodi at timestamp 1793036568 (0x6adf9118, Mon 26 Oct 2026 17:42:48 UTC, Gloas epoch 132352), matching eth-clients/hoodi#31.
• Add amsterdamTime to Chains/hoodi.json and AmsterdamTimestamp to HoodiSpecProvider, mirroring the Sepolia scheduling in Schedule Sepolia Amsterdam fork #13455.
• Extend the Hoodi chainspec-vs-hardcoded spec comparison, transition activations and fork id / fork hash tests with the Amsterdam boundary (0x3d068b59).

Types of changes

What types of changes does your code introduce?

• Bugfix (a non-breaking change that fixes an issue)
• New feature (a non-breaking change that adds functionality)
• Breaking change (a change that causes existing functionality not to work as expected)
• Optimization
• Refactoring
• Documentation update
• Build-related changes
• Other: Description

Testing

Requires testing

• Yes

If yes, did you write tests?

• Yes

Notes on testing

The fork hash after Amsterdam was cross-checked with an independent CRC32 over the Hoodi genesis hash and fork timestamps.

Documentation

Requires documentation update

• Yes
• No

Requires explanation in Release Notes

• Yes
• No

Remarks

The 200M gas limit in the GAS_LIMIT_SCHEDULE of the Hoodi PR is a consensus-layer setting delivered to the EL as the payload target gas limit, so Blocks.TargetBlockGasLimit in the Hoodi configs is left unchanged.

[claude]

Claude finished @flcl42's task in 55s —— View job

---

PR review: Schedule Hoodi Amsterdam fork

• Read the diff and repo rules
• Compare against the Sepolia Amsterdam scheduling and the geth-genesis loader
• Compare against upstream feat: schedule Hoodi Glamsterdam activation eth-clients/hoodi#31
• Post review

Result: no Critical, High, Medium or Low findings. Mergeable.

What I checked

• Timestamp: 0x6adf9118 is 1793036568, the same value as amsterdamTime in hoodi.json. (1793036568 − 1742213400) / 384 = 132352 matches the Gloas epoch. It also matches amsterdamTime in the upstream metadata/genesis.json.
• EIP set: upstream chainspec.json turns on 13 EIPs at this timestamp: 2780, 7708, 7778, 7843, 7928, 7954, 7976, 7981, 8024, 8037, 8038, 8246 and 8282. These are exactly the EIPs Forks/25_Amsterdam.cs enables. In geth-style genesis, GethGenesisLoader expands them from amsterdamTime through HardforkLabels.ExpandAll, and AmsterdamTimestamp is taken from Eip7928TransitionTimestamp. So nothing else in hoodi.json needs to change.
• Blob schedule: upstream adds no Amsterdam blobSchedule entry, and Amsterdam inherits from BPO2. Hoodi therefore keeps the BPO2 blob params (14/21) after the fork, which is correct.
• Genesis: Amsterdam is not active at genesis, so the genesis header and hash stay the same (no BAL hash or slot number is added).
• HoodiSpecProvider: follows the same pattern as SepoliaSpecProvider. ToTransitionActivations picks up the new entry automatically.
• Tests: they cover the fork-id next change at BPO2, the boundary just before Amsterdam, and the new hash 0x3d068b59, plus the hardcoded-vs-chainspec comparison and ForkScheduleSpecProviderTests. GethGenesisLoaderTests.HoodiEip7949Activations builds its cases from TransitionActivations, so it covers the new boundary without changes.
• Gas limit (from Remarks): I agree. GAS_LIMIT_SCHEDULE is a CL setting, so leaving TargetBlockGasLimit unchanged is right.

I didn't build the code or run the tests here, and I didn't recompute the fork hash myself. CI's ForkInfoTests will check it.

[github-actions]

Stateless guest cost

block	steps	Δ steps	prover cost	Δ cost
25526356	114,858,260	+0.000%	18,023,701,141	+0.000%
25532382	122,275,303	+0.000%	20,188,051,719	+0.000%
25532442	160,502,247	+0.000%	23,073,742,490	+0.000%
25532454	144,718,716	+0.000%	22,678,597,008	+0.000%
25532471	142,171,422	+0.000%	22,153,301,588	+0.000%
25532480	94,229,716	+0.000%	15,642,191,964	+0.000%
25532553	134,967,615	+0.000%	20,613,897,566	+0.000%
25532642	133,261,770	+0.000%	20,831,100,476	+0.000%
25532711	130,987,613	+0.000%	19,614,554,743	+0.000%
all 9 blocks	1,177,972,662	+0.000%	182,819,138,695	+0.000%

Variable cost by bucket, summed over shared blocks (excludes BASE)

bucket	before	after	Δ
MAIN	80,102,035,140	80,102,141,016	+0.000%
OPCODES	16,498,366,188	16,498,385,898	+0.000%
PRECOMPILES	63,844,059,139	63,844,059,139	+0.000%
MEMORY	19,788,723,996	19,788,764,226	+0.000%

60f49147dee9 (merge commit) · 9 pinned blocks · ziskemu is deterministic, so these figures are exact and any non-zero delta is real.

Compared against 0213c60141ee.