The cairo-evm from zpoken

cairo-evm's People

Watchers

cairo-evm's Issues

EVM Opcodes (tracking issue)

Cairo to EVM Proof of Concept

The first stage of compiler development is to achieve compilation of the simplest "Cairo" contract in evm
Cairo contract:

fn main() -> felt252 {
    1
}

Sierra code:

type felt252 = felt252;

libfunc felt_const<1> = felt_const<1>;
libfunc store_temp<felt252> = store_temp<felt252>;

felt_const<1>() -> ([0]);
store_temp<felt252>([0]) -> ([1]);
return([1]);

program::program::main@0() -> (felt252);

The implementation of Proof of Concept involves, at a minimum, resolving these issues:

necessary to provide data type conversion during compilation for felt252, since it is known that EVM uses 32 byte integers
require implementation of at least a few opcodes from the #1
research regarding whether we can inherit the compiler cairo-lang-sierra-to-casm
We must study well implementation kakarot. Although kakarot solves the reverse issue (solidity to cairo) it is likely to be useful to us, as it has already encountered and successfully solved Cairo and EVM compatibility issues.
It seems RUST will be the priority language for compiler development, but it is worth exploring other options

Proof of Concept Success Criteria:

byte code resulting from cairo compilation is successfully executed in EVM with the expected result

Timeline:
Target 1 month of research and development to demonstrate a working Proof of Concept and then evaluate our estimates of the timeline for full implementation

Recommend Projects