Write deposit workflow for morpho looped staking strategy + tests #5
Conversation
mikalsande
changed the title
|
Got the test env setup working, so now we can run integration tests locally. All the original tests run green, our Morpho tests fails, but at least now we have a functioning local environment we can use for integration testing. To run Morpho tests. bash testenv-anvil.sh
...
You can now run integration tests with: make test-integration
RPC_URL=http://localhost:9123 CORE_LOCATOR_ADDRESS=0x84eA74d481Ee0A5332c457a4d796187F6Ba67fEB make test-integration
Run Morpho tests:
FOUNDRY_PROFILE=test CORE_LOCATOR_ADDRESS=0x84eA74d481Ee0A5332c457a4d796187F6Ba67fEB \
forge test --match-contract MorphoLoopStrategyTest -vvvvv --fork-url http://localhost:9123
Will sleep for "forever"Then we can run integration tests in another terminal. FOUNDRY_PROFILE=test CORE_LOCATOR_ADDRESS=0x84eA74d481Ee0A5332c457a4d796187F6Ba67fEB \
forge test --match-contract MorphoLoopStrategyTest -vvvvv --fork-url http://localhost:9123
...
├─ [0] VM::deployCode("src/strategy/MorphoLoopStrategy.sol:MorphoLoopStrategy", 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)
│ ├─ [2677836] → new MorphoLoopStrategy@0xa0Cb889707d426A7A386870A03bc70d1b0697598
│ │ ├─ [2530] StvStETHPool::fallback() [staticcall]
│ │ │ ├─ [2019] StvStETHPool::WSTETH() [delegatecall]
│ │ │ │ └─ ← [Return] WstETH: [0x36C02dA8a0983159322a80FFE9F24b1acfF8B570]
│ │ │ └─ ← [Return] WstETH: [0x36C02dA8a0983159322a80FFE9F24b1acfF8B570]
│ │ ├─ [2662] StvStETHPool::fallback() [staticcall]
│ │ │ ├─ [2151] StvStETHPool::STETH() [delegatecall]
│ │ │ │ └─ ← [Return] Lido: [0xf2A08B9C303496f7FF99Ce2d4A6b6efb65E0e752]
│ │ │ └─ ← [Return] Lido: [0xf2A08B9C303496f7FF99Ce2d4A6b6efb65E0e752]
│ │ ├─ emit Initialized(version: 18446744073709551615 [1.844e19])
│ │ ├─ emit FeaturePaused(featureId: 0x43bfa4d224389b1d46b861e73ea11cdb97b1781b56bee2ffd44f313f312fb273, account: MorphoLoopStrategyTest: [0x7FA9385bE102ac3EAc297483Dd6233D62b3e1496])
│ │ └─ ← [Return] 13094 bytes of code
│ └─ ← [Return] MorphoLoopStrategy: [0xa0Cb889707d426A7A386870A03bc70d1b0697598]
├─ [0] VM::prank(0x0000000000000000000000000000000000001004)
│ └─ ← [Return]
├─ [1271] MorphoLoopStrategy::initialize(0x0000000000000000000000000000000000001004)
│ └─ ← [Revert] InvalidInitialization()
└─ ← [Revert] InvalidInitialization()
Suite result: FAILED. 0 passed; 1 failed; 0 skipped; finished in 24.08ms (0.00ns CPU time)
Ran 1 test suite in 266.04ms (24.08ms CPU time): 0 tests passed, 1 failed, 0 skipped (1 total tests)
Failing tests:
Encountered 1 failing test in test/integration/morpho-loop.test.sol:MorphoLoopStrategyTest
[FAIL: InvalidInitialization()] setUp() (gas: 0)
Encountered a total of 1 failing tests, 0 tests succeeded |
|
Tests using a Morpho Mock works now. bash testenv-anvil.sh
FOUNDRY_PROFILE=test \
CORE_LOCATOR_ADDRESS=0x84eA74d481Ee0A5332c457a4d796187F6Ba67fEB \
forge test --match-contract MorphoLoopStrategyTest -vvvvv --fork-url http://localhost:9123Next is to use the feedback from Morpho for how to set up Morpho properly for integration tests so that we use the same level of realism in our integration tests as Morpho uses them selves in their own repo. |
|
Got all but 1 test to run ✅ in CI. The |
| address public immutable MARKET_LOAN_TOKEN; | ||
| address public immutable MARKET_COLLATERAL_TOKEN; | ||
| address public immutable MARKET_ORACLE; | ||
| address public immutable MARKET_IRM; | ||
| uint256 public immutable MARKET_LLTV; |
There was a problem hiding this comment.
This could be simplified as a MarketParams. We'd need to drop the immutable, but it's easy to check if the args are mutable, they'll need a way to be replaced.
There was a problem hiding this comment.
As I understood it we want to use immutable where we can to save gas for storage and structs cannot be immutable for some reason. Happy to change this this to store the struct, gas optimization is not something we need to really think about yet.
| _pauseFeature(SUPPLY_FEATURE); | ||
| } | ||
|
|
||
| function initialize(address _admin) external initializer { |
There was a problem hiding this comment.
Note: this is to be used as an upgradeable contract I suppose, so it's called by a proxy atomically with the constructor or with some logic so this is never left to be called in different tx than constructor.
| _checkFeatureNotPaused(SUPPLY_FEATURE); | ||
|
|
||
| LoopSupplyParams memory params = abi.decode(_params, (LoopSupplyParams)); | ||
| if (params.targetLeverageBp < 10000) revert InvalidLeverage(); |
There was a problem hiding this comment.
Doesn't make sense to have targetLeverageBp 10_000, in that case should it be params.targetLeverageBp <= 10000?!
There was a problem hiding this comment.
I agree, this was just a placeholder. We can set it some other minimum level, maybe 1.5x is a reasonable minimum?
There was a problem hiding this comment.
Actually, I don't know what is a good minimum. It must be > 1x, but it must also be high enough that there is a tangible benefit to using leverage. That requires some financial analysis that I don't really know how to do.
| console.log("Additional minting capacity from borrow (stETH):", additionalMintingCapacitySteth); | ||
|
|
||
| // The actual wstETH we can mint is the minimum of what we want and what's available | ||
| uint256 additionalWstethFromBorrow = additionalMintingCapacitySteth < targetBorrowAmount |
There was a problem hiding this comment.
Here we are comparing wstEth with Eth, right? Maybe add a _token as a suffix so we know what we are comparing. In Rust we could create a type per token to ensure only dealing with the same tokens in operations.
| console.log("Leverage loop completed"); | ||
|
|
||
| // Clear callback context | ||
| delete _callbackContext; |
There was a problem hiding this comment.
Why is this done? It doesn't free storage, is it a limit in the stack?
There was a problem hiding this comment.
It deletes the callback context so to make sure that no further callbacks can come from the same source.
src/strategy/MorphoLoopStrategy.sol
Outdated
| actualLeverageBp = (position.collateral * 10000) / initialWsteth; | ||
| } | ||
|
|
||
| emit LoopExecuted(msg.sender, initialWsteth, position.collateral, position.borrowShares, actualLeverageBp); |
There was a problem hiding this comment.
Great event. We can maybe add some constraints in the end to verify things worked as expected
mikalsande
force-pushed
the
hoodi_morpho_strategy
branch
from
ad30def to
a463869
Compare
mikalsande
force-pushed
the
hoodi_morpho_strategy
branch
from
05931d5 to
d6d7b6e
Compare
2 participants
We have deployed Vault + DeFi wrapper to Hoodi: #2
Commit we deployed from: https://github.com/ChorusONe/vaults-wrapper/blob/9b3512886386f5991832db51a085633bf640f099/README.md#deployment
To make sure things will be testable and correctly versioned we start building our Looped Staking strategy with Morpho lending from commit 9b35128
This PR supersedes #4