feat(test-execute): Implement `pre.deterministic_deploy_contract` #1934

marioevz · 2025-12-16T22:33:29Z

🗒️ Description

Implements a new function in the pre object called deterministic_deploy_contract, which uses https://github.com/Arachnid/deterministic-deployment-proxy to deploy contracts at deterministic addresses using CREATE2.

The function takes the following parameters:

deploy_code: The bytecode to be deployed. Will be deployed at keccak256( 0xff ++ 0x4E59B44847B379578588920CA78FBF26C0B4956C ++ salt ++ keccak256(init_code))[12:]
salt: Salt to be passed to the CREATE2 opcode. Defaults to zero.
initcode: Initcode that deploys the code. It will not be checked in any way against the value passed in deploy_code until the transactions is sent to the network using execute. Can be skipped and the initcode will be auto-generated.
setup_calldata: If specified, a transaction will be sent to the contract after deployment with this calldata. Can be used to reset storage values before the test starts.

🔗 Related Issues or PRs

Closes #1933

✅ Checklist

All: Ran fast tox checks to avoid unnecessary CI fails, see also Code Standards and Enabling Pre-commit Checks:
```
uvx tox -e static
```
All: PR title adheres to the repo standard - it will be used as the squash commit message and should start type(scope):.
All: Considered adding an entry to CHANGELOG.md.
All: Considered updating the online docs in the ./docs/ directory.
All: Set appropriate labels for the changes (only maintainers can apply labels).
Tests: Ran mkdocs serve locally and verified the auto-generated docs for new tests in the Test Case Reference are correctly formatted.
Tests: For PRs implementing a missed test case, update the post-mortem document to add an entry the list.
Ported Tests: All converted JSON/YML tests from ethereum/tests or tests/static have been assigned @ported_from marker.

Cute Animal Picture

marioevz · 2025-12-16T22:35:41Z

One open question is that, in execute, we will not be able to run two different tests that use the same contract without causing issues.
One potential solution could be to use a lock for these contracts, but it could lead to tests dead-locking themselves.

marioevz · 2025-12-16T22:37:23Z

Example test: https://github.com/ethereum/execution-specs/blob/dd68f48bd0400d35b0bbaf00fd8be016f523edf2/tests/constantinople/eip1014_create2/test_deterministic_deployment.py

I've verified this locally using fill and kurtosis, the test can be run over and over and the deployment only happens once.

codecov · 2025-12-16T23:09:12Z

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 86.33%. Comparing base (50ef7f0) to head (7566504).

Additional details and impacted files

@@               Coverage Diff                @@
##           forks/amsterdam    #1934   +/-   ##
================================================
  Coverage            86.33%   86.33%           
================================================
  Files                  538      538           
  Lines                34557    34557           
  Branches              3222     3222           
================================================
  Hits                 29835    29835           
  Misses                4148     4148           
  Partials               574      574

Flag	Coverage Δ
unittests	`86.33% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

CPerezz

Important things also:

Missing max code size validation in execute mode. The filler implementation checks max_code_size (line 176), but execute mode doesn't. This should be patched.
DETERMINISTIC_DEPLOYMENT_CONTRACT_ADDRESS is defined multiple times.

CPerezz · 2025-12-18T14:40:33Z

packages/testing/src/execution_testing/cli/pytest_commands/plugins/execute/pre_alloc.py

+DETERMINISTIC_DEPLOYMENT_CONTRACT_ADDRESS = Address(
+    0x4E59B44847B379578588920CA78FBF26C0B4956C
+)
+DETERMINISTIC_DEPLOYMENT_SIGNER_ADDRESS = Address(
+    0x3FAB184622DC19B6109349B94811493BF2A45362
+)
+DETERMINISTIC_DEPLOYMENT_TRANSACTION = Bytes(
+    "0xf8a58085174876e800830186a08080b853604580600e600039806000f350fe7fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffe03601600081602082378035828234f58015156039578182fd5b8082525050506014600cf31ba02222222222222222222222222222222222222222222222222222222222222222a02222222222222222222222222222222222222222222222222222222222222222"
+)
+


What are these hardcoded things? Why do we have a DETERMINISTIC_DEPLOYMENT_CONTRACT_ADDRESS?

Nevermind. I assume we will set Nick's deployer unless we provide another one? If that is the case, we can close this.

IIUC, these info comes from Nick’s deterministic deployer:

DETERMINISTIC_DEPLOYMENT_CONTRACT_ADDRESS: the contract address of Nick’s deployer (the factory address).

DETERMINISTIC_DEPLOYMENT_SIGNER_ADDRESS: the address recovered via ecrecover from the randomly generated (v, r, s) values.

DETERMINISTIC_DEPLOYMENT_TRANSACTION: the transaction used to deploy Nick’s factory contract.

The information documented here.

CPerezz · 2025-12-18T14:58:13Z

packages/testing/src/execution_testing/cli/pytest_commands/plugins/execute/pre_alloc.py

+                    protected=False,
+                    nonce=0,
+                    gas_price=0x174876E800,
+                    gas_limit=0x0186A0,
+                    to=None,
+                    value=0,
+                    data=Bytes(


Why do we hardcode gas_price here to 0x174876E800 (100 Gwei) ?

The transaction that deploys the proxy must be exactly like this in order to deploy to the expected address: https://github.com/Arachnid/deterministic-deployment-proxy?tab=readme-ov-file#proxy-address. This is just how Nick's method works.

Could you confirm that your miner uses 0x4e59b44847b379578588920ca78fbf26c0b4956c please? I used this as it was the one that was linked in @jochem-brouwer message in the original issue: #1933 (comment)

Yes yes. That's Nick's deployer. I was asking about gas price and gas_limit hardcoded here rather than anything else.

The address won';t be affected by that right?

LouisTsai-Csie

Leave some comments! I now have a better understanding of the big picture, will revisit CPerezz' PR to see how to integrate.

LouisTsai-Csie · 2025-12-23T08:22:38Z

packages/testing/src/execution_testing/cli/pytest_commands/plugins/filler/pre_alloc.py

+        contract_address = compute_deterministic_create2_address(
+            salt=salt, initcode=initcode
+        )
+        max_code_size = self._fork.max_code_size()
+        assert len(deploy_code) <= max_code_size, (
+            f"code too large: {len(deploy_code)} > {max_code_size}"
+        )
+
+        super().__setitem__(
+            contract_address,
+            Account(
+                nonce=1,
+                balance=minimum_balance,
+                code=deploy_code,
+                storage={},
+            ),
+        )


Some feedback:

In execute remote, we will first check if the contract exists or not, i believe we should do the similar verification here, or we might override the existing contract if we do not have proper salt management. Moreover, in post-state object, it would not raise error when there is duplicate entry.

def test_collision_detection( blockchain_test: BlockchainTestFiller, pre: Alloc, ) -> None: """ Test collision detection for deterministic deployments. """ deploy_code = Op.SSTORE(1, Op.CALLDATALOAD(0)) addr_1 = pre.deterministic_deploy_contract( deploy_code=deploy_code, setup_calldata=Hash(0), label="contract_1", ) # This would override the addr_1 account addr_2 = pre.deterministic_deploy_contract( deploy_code=deploy_code, setup_calldata=Hash(1), label="contract_2", ) # If we are not aware the `addr_1` and `addr_2` are the same, the test might not behave as intended. assert addr_1 == addr_2, "Contract address is the same" sender = pre.fund_eoa() tx_1 = Transaction( sender=sender, to=addr_1, data=Hash(1), gas_limit=100_000, ) tx_2 = Transaction( sender=sender, to=addr_2, data=Hash(2), gas_limit=100_000, ) post = { # this is overriden by the `addr_2` account setting, as `addr_1` is the same as `addr_2`. addr_1: Account( deploy_code=deploy_code, storage={ 1: 1, }, ), addr_2: Account( deploy_code=deploy_code, storage={ 1: 2, }, ), } blockchain_test(pre=pre, post=post, blocks=[Block(txs=[tx_1, tx_2])])

Moreover, the setup_calldata mechanism is not working for fill but execute remote:

Below is an example:

def test_missing_setup_phase( blockchain_test: BlockchainTestFiller, pre: Alloc, ) -> None: """ Test collision detection for deterministic deployments. """ deploy_code = Op.SSTORE(1, Op.CALLDATALOAD(0)) addr_1 = pre.deterministic_deploy_contract( deploy_code=deploy_code, salt=Hash(1), setup_calldata=Hash(0), label="contract_1", ) addr_2 = pre.deterministic_deploy_contract( deploy_code=deploy_code, salt=Hash(2), setup_calldata=Hash(1), label="contract_2", ) assert addr_1 != addr_2, "Contract addresses should be different" sender = pre.fund_eoa() tx_1 = Transaction( sender=sender, to=addr_1, data=Hash(2), gas_limit=100_000, ) post = { addr_1: Account( deploy_code=deploy_code, storage={ 1: 2, }, ), addr_2: Account( deploy_code=deploy_code, storage={ 1: 1, }, ), } blockchain_test(pre=pre, post=post, blocks=[Block(txs=[tx_1])])

expected behavior (Works under execute remote)

slot 1 in addr_1 increases from 1 to 2

slot 1 in addr_2 increases stays 1

Current behavior (In fill mode)

slot 1 in addr_1 from 0 to 2

slot 1 in addr_2 stays 0

LouisTsai-Csie · 2025-12-23T08:33:11Z

packages/testing/src/execution_testing/cli/pytest_commands/plugins/execute/pre_alloc.py

+                new_bytes=len(initcode)
+            )
+            deploy_gas_limit += calldata_gas_calculator(data=initcode)
+            deploy_gas_limit = min(deploy_gas_limit * 2, 30_000_000)


This 30M hardcoded limit does not work for Osaka (eip-7825).

LouisTsai-Csie · 2025-12-23T08:35:00Z

packages/testing/src/execution_testing/cli/pytest_commands/plugins/execute/pre_alloc.py

        # Limit the gas limit
        deploy_gas_limit = min(deploy_gas_limit * 2, 30_000_000)


Same argument

marioevz requested a review from fselmo December 16, 2025 22:33

marioevz mentioned this pull request Dec 16, 2025

feat(test-cli-execute): Allow use of address during pre.deploy_contract #1933

Open

CPerezz suggested changes Dec 18, 2025

View reviewed changes

LouisTsai-Csie requested changes Dec 23, 2025

View reviewed changes

CPerezz mentioned this pull request Dec 23, 2025

Heavy-setup benchmark test handling ethpandaops/gas-lighting-tracker#23

Open

marioevz added 3 commits December 28, 2025 20:13

feat(test-all): Implement deterministic_deploy_contract

44522fb

feat(tests): deterministic_deploy_contract example

b29deaf

docs: update

7566504

marioevz force-pushed the deterministic-deploy-contract branch from f968266 to 7566504 Compare December 28, 2025 20:13

		# Limit the gas limit
		deploy_gas_limit = min(deploy_gas_limit * 2, 30_000_000)

feat(test-execute): Implement pre.deterministic_deploy_contract #1934

Are you sure you want to change the base?

feat(test-execute): Implement pre.deterministic_deploy_contract #1934

Uh oh!

Conversation

marioevz commented Dec 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🗒️ Description

🔗 Related Issues or PRs

✅ Checklist

Cute Animal Picture

Uh oh!

marioevz commented Dec 16, 2025

Uh oh!

marioevz commented Dec 16, 2025

Uh oh!

codecov bot commented Dec 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

CPerezz left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

LouisTsai-Csie left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

feat(test-execute): Implement `pre.deterministic_deploy_contract` #1934

feat(test-execute): Implement `pre.deterministic_deploy_contract` #1934

marioevz commented Dec 16, 2025 •

edited

Loading

codecov bot commented Dec 16, 2025 •

edited

Loading