Fjord: secp256r1 curve support (RIP-7212)

[mdehoog]

Description

Same as ethereum/go-ethereum#27540, for early rollout for op-stack chains. Currently targeted to be enabled in the delta fjord hardfork.

Tests

Brought tests over from the go-ethereum PR.

Additional context

See https://github.com/ethereum/RIPs/blob/master/RIPS/rip-7212.md (and https://eips.ethereum.org/EIPS/eip-7212).

[tynes]

Currently targeted to be enabled in the delta hardfork.

Where is the discussion for the delta hardfork happening?

[mdehoog]

@tynes I'm using "delta" as a reference to the next hardfork post-canyon, but I don't believe discussions have started about it yet. I opened this PR to begin to build alignment to get this out in the next hardfork.

[tynes]

@tynes I'm using "delta" as a reference to the next hardfork post-canyon, but I don't believe discussions have started about it yet. I opened this PR to begin to build alignment to get this out in the next hardfork.

Got it, thanks! Just want to make sure twitter doesn't think this has already been scheduled for inclusion. I want to get this in but I also want to make sure that we do not act in a way that breaks assumptions about the EVM. We should discuss this at RollCall and get something into the RIPs so that we can get consensus between the L2s on what address we should include the predeploy at

[ulerdogan]

Hello! I am happy to share that the specification of the proposal has been finalized by moving into the RIPs repository and the last changes will be applied with this PR.

[mdehoog]

Thank you for the ping @ulerdogan, I've pulled in your implementation address modifications from ethereum/go-ethereum@master...ulerdogan:go-ethereum:ulerdogan-secp256r1 and targeted the fjord hardfork (one after the 4844 ecotone hardfork)

[ulerdogan]

Thanks for the update @mdehoog! Also, I recommend to include more tests and review for my implementation. I would love to collaborate on this.

[xiaoxiaff]

Hey @ulerdogan! Nice to e-meet you! I am a software engineer from Coinbase and I would love to work with you on more test cases to push this forward.

For the test cases, do you have anything top of your mind? I am thinking to bring some secp256r1 test vectors but not seeing that mentioned in the EIP7212. Let me know if you have any ideas or you are fine with some secp256r1 test vectors.

cc @mdehoog @angel-ding-cb

[ulerdogan]

Hi @xiaoxiaff, nice to meet you too! I agree with you that it'd be good to apply the secp256r1 test vectors as the proposal specs are fully compatible with the original standard. How should we collaborate? Also, this repo by @daimo-eth team is a great example for the test vectors, applying for the Solidity verifier.

[xiaoxiaff]

@ulerdogan let me prepare some tests for this PR and feel free to left comments. If it looks good we can add those to your original code as well.

Double check does the EIP-7212 finalized? It looks like RIP-7212 is already finalized and if we need to keep it consistent with EIP then we probably won't be able to add the test vector to the RIP for reviews? 👀

[ulerdogan]

Cool, thanks!

EIP-7212 will be removed from the EIPs repo and the precompile proposals will remain as only the RIP-7212. So, we don't need to consider the EIP or any other change.

Additionally, today in the RollCall Breakout #3, standardized tests for the RIPs, like Ethereum test repos, have been discussed to make sure that every rollup implementation behaves the same. Looks like this topic will be consulted with the L1 team and there will be another testing repo.

[xiaoxiaff]

@ulerdogan I have the PR ready which includes the test vectors from Google's wycheproof. Let me know if that looks good to you!

cc @mdehoog @angel-ding-cb

[ulerdogan]

Thanks @xiaoxiaff, also commited to my branch.

[sebastianst]

This looks good to go! Only a few open questions for my understanding.

Would be ideal if we could add the Fjord override and superchain config change from #249 to this PR, as we merge it first.

In the future, we should try to have a small separate PR just for adding the new fork fields, overrides etc...

We should also add an e2e test to the monorepo to confirm this is working as expected after Fjord activation.

[sebastianst]

This looks great, thanks @danyalprout @mdehoog @ulerdogan @yukaitu-cb!