Implement EIP-7907 #1231
Implement EIP-7907 #1231
Conversation
|
Tracking a PR that is looking to make small updates to the EIP |
|
Nice writeup! One missing piece here is the warm/cold code state as explicitly explained in this PR. |
Pushed a commit e2d1c46 that incorporates changes from the following PRs to the specs. Please have a look
Also, the EIP still needs to explicitly clarify how to handle code that is delegated via 7702. In this implementation, I have assumed that the delegated code is treated with the same warm/cold behaviour but it is best to explicitly spec that out. |
gurukamath
force-pushed
the
eips/osaka/eip-7907
branch
from
e2d1c46 to
690b3b4
Compare
| STACK_DEPTH_LIMIT, | ||
| process_create_message, | ||
| ) | ||
|
|
||
| call_data = memory_read_bytes( | ||
| evm.memory, memory_start_position, memory_size | ||
| ) | ||
| if len(call_data) > 2 * MAX_CODE_SIZE: | ||
| if len(call_data) > MAX_INIT_CODE_SIZE: | ||
| raise OutOfGasError | ||
|
|
||
| create_message_gas = max_message_call_gas(Uint(evm.gas_left)) |
There was a problem hiding this comment.
L107: wondering if we should add evm.message.warm_code_addresses.add(contract_address) to account for CREATE* invocations
| @@ -86,5 +86,6 @@ def prepare_message( | |||
| accessed_addresses=accessed_addresses, | |||
There was a problem hiding this comment.
L71: wondering if we should add warm_code_addresses.add(current_target) to account for CREATE* addresses and tx.recipient given its considered warm automatically (?)
(closes #1220 )
What was wrong?
EIP-7907 is not implemented on the Osaka branch.
Related to Issue #1220
How was it fixed?
Implement EIP-7907