common: handle forkHash on timestamp == genesis timestamp

[jochem-brouwer]

Fixes:

./hive --sim "ethereum/engine" --client ethereumjs --sim.parallelism 16 --sim.limit="engine-cancun/Request Blob Pooled Transactions"

If fork is at timestamp genesisTimestamp then this fork is supposed to be active at timestamp = 0 and not participate in the forkHash calculation of the devp2p/ETH STATUS message.

See also:

hyperledger/besu#5744
ethereum/go-ethereum#27895
https://eips.ethereum.org/EIPS/eip-6122

[codecov]

Codecov Report

Merging #2959 (74c4559) into master (b54a05c) will increase coverage by 0.03%.
Report is 1 commits behind head on master.
The diff coverage is 50.00%.

❗ Current head 74c4559 differs from pull request most recent head 72466f6. Consider uploading reports for the commit 72466f6 to get more accurate results

Additional details and impacted files

Flag	Coverage Δ
block	88.66% <ø> (ø)
blockchain	92.58% <ø> (ø)
client	87.59% <ø> (+0.05%)	⬆️
common	98.56% <50.00%> (-0.10%)	⬇️
ethash	∅ <ø> (∅)
evm	70.43% <ø> (ø)
rlp	?
statemanager	82.98% <ø> (ø)
trie	89.89% <ø> (+0.25%)	⬆️
tx	95.88% <ø> (ø)
util	86.78% <ø> (ø)
vm	79.20% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

[g11tech]

lgtm

[holgerd77]

Great find! This is a relatively sensitive change and can have side effects on other girl hash calculations, I would feel better with 1-2 dedicated test cases for it.

[jochem-brouwer]

Yes, you are right, especially these "small" changes can be tested quickly with small tests. I will add tests.

For a general comment though, some hive tests are very complex and would take long to integrate - what we could do (mid-term) is to create a hive CI runner and then run these specific tests to cover these test cases (and for a bonus: also include this into codecov, but that would be complex since we need to hook into hive's docker image of our client)