Bring back TXCREATE #177
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -103,6 +103,47 @@ On top of the types defined in the table above, the following validity constrain | |
| - the total size of not yet deployed container might be up to `data_size` lower than the above values due to how the data section is rewritten and resized during deployment (see [Data Section Lifecycle](#data-section-lifecycle)) | ||
| - the total size of a container must not exceed `MAX_INITCODE_SIZE` (as defined in EIP-3860) | ||
|
|
||
| ## Transaction Types | ||
|
|
||
| Introduce new transaction type `InitcodeTransaction` which extends EIP-1559 (type 2) transaction by adding a new field `initcodes: List[ByteList[MAX_INITCODE_SIZE], MAX_INITCODE_COUNT]`. | ||
|
|
||
| The `initcodes` can only be accessed via the `TXCREATE` instruction (see below), therefore `InitcodeTransactions` are intended to be sent to contracts including `TXCREATE` in their execution. | ||
|
|
||
| We introduce a standardised Creator Contract (i.e. written in EVM, but existing at a known address, such as precompiles), which eliminates the need to have create transactions with empty `to`. The Creator Contract will be predeployed at the EOF activation block. Note that such introduction of the Creator Contract is needed, because only EOF contracts can create EOF contracts. See the appendix below for Creator Contract code. | ||
|
|
||
| Under transaction validation rules `initcodes` are not validated for conforming to the EOF specification. They are only validated when accessed via `TXCREATE`. This avoids potential DoS attacks of the mempool. If during the execution of an `InitcodeTransaction` no `TXCREATE` instruction is called, such transaction is still valid. | ||
|
|
||
| `initcodes` data is similar to calldata for two reasons: | ||
| 1) It must be fully transmitted in the transaction. | ||
| 2) It is accessible to the EVM, but it can't be fully loaded into EVM memory. | ||
|
|
||
| For these reasons, define cost of each of the `initcodes` items same as calldata (16 gas for non-zero bytes, 4 for zero bytes -- see EIP-2028). The intrinsic gas of an `InitcodeTransaction` is extended by the sum of all those items' costs. | ||
|
|
||
| EIP-3860 and EIP-170 still apply, i.e. `MAX_CODE_SIZE` as 24576, `MAX_INITCODE_SIZE` as `2 * MAX_CODE_SIZE`. Define `MAX_INITCODE_COUNT` as 256. | ||
|
|
||
| `InitcodeTransaction` is invalid if either: | ||
| - there are more than `MAX_INITCODE_COUNT` entries in `initcodes` | ||
| - `initcodes` is an empty array | ||
| - length of any entry in `initcodes` exceeds `MAX_INITCODE_SIZE` | ||
| - any entry in `initcodes` has zero length | ||
| - the `to` is `nil` | ||
|
|
||
| #### RLP and signature | ||
|
|
||
| Given the definitions from [EIP-2718](https://eips.ethereum.org/EIPS/eip-2718) the `TransactionPayload` for an `InitcodeTransaction` is the RLP serialization of: | ||
|
|
||
| ``` | ||
| [chain_id, nonce, max_priority_fee_per_gas, max_fee_per_gas, gas_limit, to, value, data, access_list, initcodes, y_parity, r, s] | ||
| ``` | ||
|
|
||
| `TransactionType` is `INITCODE_TX_TYPE` (`0x05`) and the signature values `y_parity`, `r`, and `s` are calculated by constructing a secp256k1 signature over the following digest: | ||
|
|
||
| ``` | ||
| keccak256(INITCODE_TX_TYPE || rlp([chain_id, nonce, max_priority_fee_per_gas, max_fee_per_gas, gas_limit, to, value, data, access_list, initcodes])) | ||
| ``` | ||
|
|
||
| The [EIP-2718](https://eips.ethereum.org/EIPS/eip-2718) `ReceiptPayload` for this transaction is `rlp([status, cumulative_transaction_gas_used, logs_bloom, logs])`. | ||
|
|
||
| ## Execution Semantics | ||
|
|
||
| Code executing within an EOF environment will behave differently than legacy code. We can break these differences down into i) changes to existing behavior and ii) introduction of new behavior. | ||
|
|
@@ -118,6 +159,7 @@ Code executing within an EOF environment will behave differently than legacy cod | |
| - The instruction `JUMPDEST` is renamed to `NOP` and remains charging 1 gas without any effect. | ||
| - Note: jumpdest-analysis is not performed anymore. | ||
| - EOF contract may not deploy legacy code (it is naturally rejected on the code validation stage) | ||
| - Legacy creation transactions (any tranactions with empty `to`) are invalid in case `data` contains EOF code (starts with `EF00` magic) | ||
| - When executed from a legacy contract, if instructions `CREATE` and `CREATE2` have EOF code as initcode (starting with `EF00` magic) | ||
| - deployment fails (returns 0 on the stack) | ||
| - caller's nonce is not updated and gas for initcode execution is not consumed | ||
|
|
@@ -126,29 +168,6 @@ Code executing within an EOF environment will behave differently than legacy cod | |
|
|
||
| **NOTE** Like for legacy targets, the aforementioned behavior of `EXTCODECOPY`, `EXTCODEHASH` and `EXTCODESIZE` does not apply to EOF contract targets mid-creation, i.e. those report same as accounts without code. | ||
|
|
||
| **NOTE** Legacy contract and legacy creation transactions may not deploy EOF code, that is behavior from [EIP-3541](https://eips.ethereum.org/EIPS/eip-3541) is not modified. | ||
|
|
||
| ### New Behavior | ||
|
|
@@ -198,13 +217,12 @@ The following instructions are introduced in EOF code: | |
| - peform (and charge for) memory expansion using `[input_offset, input_size]` | ||
| - load initcode EOF subcontainer at `initcontainer_index` in the container from which `EOFCREATE` is executed | ||
| - let `initcontainer` be that EOF container, and `initcontainer_size` its length in bytes | ||
| - check call depth limit and whether caller balance is enough to transfer `value` | ||
| - in case of failure returns 0 on the stack, caller's nonce is not updated and gas for initcode execution is not consumed. | ||
| - caller's memory slice [`input_offset`:`input_size`] is used as calldata | ||
| - execute the container and deduct gas for execution. The 63/64th rule from EIP-150 applies. | ||
| - increment `sender` account's nonce | ||
| - calculate `new_address` as `keccak256(0xff || sender || salt)[12:]` | ||
| - behavior on `accessed_addresses` and address colission is same as `CREATE2` (rules for `CREATE2` from [EIP-684](https://eips.ethereum.org/EIPS/eip-684) and [EIP-2929](https://eips.ethereum.org/EIPS/eip-2929) apply to `EOFCREATE`) | ||
| - an unsuccesful execution of initcode results in pushing `0` onto the stack | ||
| - can populate returndata if execution `REVERT`ed | ||
|
|
@@ -216,11 +234,26 @@ The following instructions are introduced in EOF code: | |
| - set `state[new_address].code` to the updated deploy container | ||
| - push `new_address` onto the stack | ||
| - deduct `200 * deployed_code_size` gas | ||
| - `TXCREATE (0xed)` instruction | ||
| - Works the same as `EOFCREATE` except: | ||
| - does not have `initcontainer_index` immediate | ||
| - pops one more value from the stack (first argument): `tx_initcode_hash` | ||
| - loads the initcode EOF container from the transaction `initcodes` array which hashes to `tx_initcode_hash` | ||
| - fails (returns 0 on the stack) if such initcode does not exist in the transaction, or if called from a transaction of `TransactionType` other than `INITCODE_TX_TYPE` | ||
| - caller's nonce is not updated and gas for initcode execution is not consumed. Only `TXCREATE` constant gas was consumed | ||
| - let `initcontainer` be that EOF container, and `initcontainer_size` its length in bytes | ||
| - deduct `2 * ((initcontainer_size + 31) // 32)` gas (EIP-3860 charge) | ||
| - just before executing the initcode container: | ||
| - **validates the initcode container and all its subcontainers recursively** | ||
| - validation includes checking that the `initcontainer` does not contain `RETURN` or `STOP` | ||
| - in addition to this, checks if the initcode container has its `len(data_section)` equal to `data_size`, i.e. data section content is exactly as the size declared in the header (see [Data section lifecycle](#data-section-lifecycle)) | ||
| - fails (returns 0 on the stack) if any of those was invalid | ||
| - caller’s nonce is not updated and gas for initcode execution is not consumed. Only `TXCREATE` constant and EIP-3860 gas were consumed | ||
| - `RETURNCONTRACT (0xee)` instruction | ||
| - loads `uint8` immediate `deploy_container_index` | ||
| - pops two values from the stack: `aux_data_offset`, `aux_data_size` referring to memory section that will be appended to deployed container's data | ||
| - cost 0 gas + possible memory expansion for aux data | ||
| - ends initcode frame execution and returns control to `EOFCREATE` or `TXCREATE` caller frame. | ||
| - `deploy_container_index` and `aux_data` are used to construct deployed contract (see above) | ||
| - instruction exceptionally aborts if after the appending, data section size would overflow the maximum data section size or underflow (i.e. be less than data section size declared in the header) | ||
| - `DATALOAD (0xd0)` instruction | ||
|
|
@@ -348,6 +381,36 @@ During scanning, for each instruction: | |
|
|
||
| Annotated examples of EOF formatted containers demonstrating several key features of EOF can be found in [this test file within the `evmone` project repository](https://github.com/ethereum/evmone/blob/master/test/unittests/eof_example_test.cpp). | ||
|
|
||
| ## Appendix: Creator Contract | ||
|
|
||
| ```solidity | ||
| { | ||
| /// Takes [tx_initcode_hash][salt][init_data] as input, | ||
| /// creates contract and returns the address or failure otherwise | ||
|
|
||
| // init_data.length can be 0, but the first 2 words are mandatory | ||
| let size := calldatasize() | ||
| if lt(size, 64) { revert(0, 0) } | ||
|
|
||
| let tx_initcode_hash := calldataload(0) | ||
| let salt := calldataload(32) | ||
|
There was a problem hiding this comment. I think we should keccak the tx_initcode_hash with the salt. Having salt only deployment as the principal toehold feels like opening up to a race, with no connection to the deployed code. Subsequent ERC contracts can be deployed that do keep this feature, but if the salt only contract goes first the ERC contract set could get front run by literally anyone. Notionally There was a problem hiding this comment. yup, same realization in the "big thread". initcode_hash missing would actually not only open a race but break the feature of cross-chain deployments to same address, with no option of recovery! Will fix in a second. is the addition of There was a problem hiding this comment. I'm not a huge fan of the particular choice of magic value coinciding with EOF magic value. For now I'll leave it out for brevity, but I'll add a ticket to track figuring this out properly when time comes There was a problem hiding this comment. It can be another magic value, but I would like the address mining to be in it's own "namespace" and adding a fixed unique value does that. Any mini-salt is fine. There was a problem hiding this comment. ah, good point. How do we pick such "namespace" though... There was a problem hiding this comment. ah, good point. How do we pick such "namespace" though... There was a problem hiding this comment. My idea with For the ERCs I plan on So it could be the EIP number in hex as an option too. There was a problem hiding this comment.
The same can be said about init_data, right? Seems like the salt should be the hash of all of calldata for this to be secure. There was a problem hiding this comment. If the contract is being called via EXTDELEGATECALL, the caller address is also hashed into deploy address automatically, doesn't this alleviate front-running risk? There was a problem hiding this comment. Yes, is that the expected usage? EOAs are not able to use EXTDELEGATECALL (they can with EIP-7702 but we can't assume that of every EOA). I wouldn't make this contract insecure if used via normal EXTCALL. |
||
|
|
||
| let init_data_size := sub(size, 64) | ||
| calldatacopy(0, 64, init_data_size) | ||
|
|
||
| let ret := txcreate(tx_initcode_hash, callvalue(), salt, 0, init_data_size) | ||
| if iszero(ret) { revert(0, 0) } | ||
|
|
||
| mstore(0, ret) | ||
| return(0, 32) | ||
|
|
||
| // Helper to compile this with existing Solidity (with --strict-assembly mode) | ||
| function txcreate(a, b, c, d, e) -> f { | ||
| f := verbatim_5i_1o(hex"ed", a, b, c, d, e) | ||
| } | ||
| } | ||
| ``` | ||
|
|
||
pdobacz marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| ## Appendix: Original EIPs | ||
|
|
||
| These are the individual EIPs which evolved into this spec. | ||
|
|
@@ -362,4 +425,3 @@ These are the individual EIPs which evolved into this spec. | |
| - 📃[EIP-663](https://eips.ethereum.org/EIPS/eip-663): Unlimited SWAP and DUP instructions [_history_](https://github.com/ethereum/EIPs/commits/master/EIPS/eip-663.md) | ||
| - 📃[EIP-7069](https://eips.ethereum.org/EIPS/eip-7069): Revamped CALL instructions (*does not require EOF*) [_history_](https://github.com/ethereum/EIPs/commits/master/EIPS/eip-7069.md) | ||
| - 📃[EIP-7620](https://eips.ethereum.org/EIPS/eip-7620): EOF - Contract Creation Instructions [_history_](https://github.com/ethereum/EIPs/commits/master/EIPS/eip-7620.md) | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I wonder how this interacts with EIP-7623, probably in its terms
tokens_in_calldatashould be extended with counting the tokens ofinitcodes(which in the end is the same as extending intrinsic gas withinitcodescost and applying calldata cost floor)There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
yea, good catch. I think the intent is that initcodes is charged for just like regular calldata. I propose we delay speccing this, just in case 7623 undergoes some last minute change. I'll leave an issue to handle this
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It's contract data. Perhaps we just outright charge the floor rate for the txcreate initcode, and say that encompasses the INITCODE_WORD_COST as well as all EOF code validation costs.