Add common boilerplate for Berlin hardfork

[cgewecke]

Part of #736

• Berlin Meta: EIP 2070
• reference PR ethereumjs-common 51

[codecov]

Codecov Report

Merging #755 into master will decrease coverage by 0.01%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #755      +/-   ##
==========================================
- Coverage   92.02%   92.00%   -0.02%     
==========================================
  Files          46       43       -3     
  Lines        2998     2991       -7     
  Branches      469      469              
==========================================
- Hits         2759     2752       -7     
  Misses        144      144              
  Partials       95       95              

Flag	Coverage Δ
#account	94.11% <ø> (ø)
#block	88.09% <ø> (ø)
#blockchain	88.74% <ø> (ø)
#common	94.11% <ø> (-0.26%)	⬇️
#tx	94.02% <ø> (ø)
#vm	93.09% <ø> (ø)

Impacted Files	Coverage Δ
packages/common/src/genesisStates/index.ts
packages/common/src/chains/index.ts

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 56d02e7...e7847df. Read the comment docs.

[holgerd77]

Just some side note, in case anyone is wondering: these consensus and finality fields are from the old Casper/sharding days when it was still planned to do the consensus/finality switches on-chain, modeled this this way to be able to do the respective switches cleanly along the Casper HFs.

So these can very well removed, for the moment this is I guess rather distracting and confusing than anything else.

[cgewecke]

@holgerd77 Should we remove those fields in this PR or in a separate one?

[holgerd77]

Think I will merge here now since this is ready-for-merge here anyhow (thanks!). If you want you can open a separate PR on this, not too pressing though.

[holgerd77]

Great! 😄 I would guess the Kovan network will do this HF as well so I think you can also add the boilerplate code to the kovan.json chain file? Otherwise this already looks good.

The base fees for the opcodes never made it to the Common library and are still collected in opcodes.ts. This should be changed I guess at some point and the base fees also moved over, with a start in the respective HF file on the "birth" of the opcode. For now it should be ok to add these to the VM as you've already done.

These are the only fees from the EIP to take into account I guess?

On the VM side you can already add berlin to the list of supportedHardforks (from my side also in this PR), then it get's allowed to use this HF switch in the VM and it should be easier to integrate on tests.

[holgerd77]

LGTM