refactor: rewrite revert and precompile decoding #6222
Conversation
| @@ -29,7 +29,7 @@ use proptest::{ | |||
| test_runner::{TestCaseError, TestRunner}, | |||
| }; | |||
| use revm::{ | |||
| primitives::{Address as rAddress, HashMap, U256 as rU256}, | |||
| primitives::{Address as aAddress, HashMap, U256 as rU256}, | |||
There was a problem hiding this comment.
I think we can just use alloy_primitives::Address now
There was a problem hiding this comment.
Anvil stuff, I don't want to get into that
| _ => unreachable!(), | ||
| }; | ||
|
|
||
| // TODO: Other chain precompiles |
There was a problem hiding this comment.
I'm a bit unsure if we want to implement precompiles for other chains—we had discussed this is a thing we probably don't want to do (albeit we could probably do Optimism's easily?) cc @mattsse
| node.decode_precompile(precompile_fn, &self.labels); | ||
| } else if let RawOrDecodedCall::Raw(ref bytes) = node.trace.data { | ||
| if bytes.len() >= 4 { | ||
| // TODO: chain ID argument |
There was a problem hiding this comment.
This is to decode different precompiles per chain id?
There was a problem hiding this comment.
Yes that's what I was thinking, but maybe I'm thinking ahead too much. Maybe we should just try to decode whatever revm has set as precompiles
There was a problem hiding this comment.
Yeah I think might be best to stick to what revm has—I think soft consensus is to not get into maintaining precompiles for other chains—they can become a pain. IIRC arbitrum, which is the biggest offender, has a repo that simulates their custom precompile that ppl can use
|
A thing to note is that now |
Motivation
Closes #6102
Fixes #5337
Solution