Update "--dev" to use the Cancun hardfork.

[Amxx]

Currently, calling ethereumjs --dev=true starts a network using the london hardfork. This is quite different from mainnet, and IMO is not representative of what dev would expect.

With DenCun hitting mainnet in the next days, I believe that this should be updated to Cancun.

[holgerd77]

Will circle in @g11tech in here, most experienced with these dev flags: Gajinder, can we update here, or does this have side effects on tests (maybe we'll directly see on CI run, not yet finished while I am writing here)?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 86.91%. Comparing base (891ee51) to head (45f4402).

Additional details and impacted files

Flag	Coverage Δ
block	88.43% <ø> (ø)
blockchain	91.61% <ø> (ø)
client	84.89% <ø> (ø)
common	98.43% <ø> (ø)
devp2p	82.12% <ø> (ø)
ethash	∅ <ø> (∅)
evm	74.05% <ø> (ø)
genesis	99.98% <ø> (ø)
rlp	∅ <ø> (∅)
statemanager	77.00% <ø> (ø)
trie	89.01% <ø> (-0.32%)	⬇️
tx	95.55% <ø> (ø)
util	89.19% <ø> (ø)
vm	79.85% <ø> (ø)
wallet	88.35% <ø> (ø)

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

[g11tech]

LGTM, not approving if we want the unreleased changelog edit removed

[g11tech]

Also as mentioned in the PR, let's just schedule it to dencun

The only reason we may want it to be on London however is to have a runnable chain without a CL

[holgerd77]

Updated this via UI

[holgerd77]

The only reason we may want it to be on London however is to have a runnable chain without a CL

@g11tech can't judge here. So: merge or not? (maybe also for others) 🙂

[holgerd77]

Updated this via UI

[acolytec3]

Unfortunately, I don't think its as simple as changing the starting hardfork to start on Cancun. If you run npx tsx bin/client.ts --dev=true you get this sort of output:

[03-12|20:17:18] INFO ---------------------------
[03-12|20:17:18] INFO Client started successfully
[03-12|20:17:18] INFO ---------------------------
[03-12|20:17:28] INFO txsByPriceAndNonce selected txs=0, skipped byNonce=0 byPrice=0 byBlobsLimit=0 
[03-12|20:17:28] INFO Miner: Assembling block from 0 eligible txs (baseFee: 7)
[03-12|20:17:28] INFO Miner: Sealed block with 0 txs (in turn)
[03-12|20:17:28] WARN Execution of block number=1 hash=0x8c93…dc62 hardfork=cancun failed:
Error: invalid block stateRoot, got: 0x5e2514a6fc32e2f9b31cbe99d78a5caa2e5e8fbd51461faa8d3f0cc59714ab37, want: 0x874ad777665ba3859329672df99b4bfee3ee775b36379d773ae69dc6d8612db2 (vm hf=cancun -> block number=1 hash=0x8c93e4329f71ba3f7fd33c6f001572b1e31bd02b8ba7c7a044d2438e6523dc62 hf=cancun baseFeePerGas=7 txs=0 uncles=0) 
[03-12|20:17:28] INFO Executed blocks count=1 first=0 hash=0xa8e0…b110 td=3 baseFee=7 hardfork=cancun last=0 hash=0xa8e0…b110 txs=0 
[03-12|20:17:36] INFO Stopped synchronization.

It's trying to start a PoA network with post-merge hardforks which is an invalid combination with the current dev genesis config. I agree with the spirit of the PR but we need to think about something a little more fully orbed before we take this in, maybe provide something like the lodestar-quickstart setup so it comes paired with a CL client so it can produce valid blocks.