feat!: Add Keccak constraints #150
Conversation
Indeed, it is not needed to be done here, the support is done on ACVM. I will also remove the other functions (except pedersen) and put them in ACVM. |
|
@TomAFrench I have renamed the Since the pwg function for Keccak256 was put in after the fallible traits, PR 151 would be a blocker for the Keccak256 opcode unless we just implemented the Keccak256 functionality in aztec_backend until we can land acvm >0.11. I left a comment here on 151 and spoke to @phated on slack. I went with re-implementing it in here so we don't rush 151 and have it come out in a less than ideal state, coupled with the fact that the change is isolated and mild. Let me know your thoughts on this |
|
Ah, we haven't updated to ACVM 0.11.0 yet. |
|
We should add a simple testcase for a keccak constraint. I'm ok with delaying this as we want to get this out for nouns though. |
* acvm-0.12.0: fix bad rebase chore: Release 0.1.2 (#183) fix: Remove star dependencies to allow publishing (#182) chore: Release 0.1.1 (#181) fix: Add description so crate can be published (#180) chore: update readme to new name and add contract note (#177) feat!: update to acvm with non-homogeneous bb calls (#169) update to latest changes chore: Release 0.1.0 (#173) use patch syntax chore: update to use new black box solver interface feat!: update to target acvm-84b5d18d chore(ci): Update tokens (#174) chore: Add release-please and publish workflows to project (#172) feat!: Update to ACVM v0.11.0 (#151) chore: remove usage of `std::mem::forget` (#164) chore: Enforce proper conversion of memory into fixed length array (#163) chore: Add test for Keccak256 constraint (#158) chore: use `WASMValue` for wasm arguments as well as return values (#157) feat!: Add Keccak constraints (#150)
This PR includes Keccak constraints when it serialise an ACIR circuit having Keccak opcodes.
It must be used with a Barretenberg version which handle them (cf. AztecProtocol/barretenberg#393)