Gas Optimizations

[code423n4]

State variables that could be set immutable

In the following files there are state variables that could be set immutable to save gas.

Code instances:

    wrapped in ConnextPriceOracle.sol
    remoteDomain in Replica.sol
    tokenBeacon in TokenRegistry.sol

Unused state variables

Unused state variables are gas consuming at deployment (since they are located in storage) and are
a bad code practice. Removing those variables will decrease deployment gas cost and improve code quality.
This is a full list of all the unused storage variables we found in your code base.

Code instances:

    NomadBase.sol, committedRoot
    PriceOracle.sol, isPriceOracle

Unnecessary array boundaries check when loading an array element twice

There are places in the code (especially in for-each loops) that loads the same array element more than once.
In such cases, only one array boundaries check should take place, and the rest are unnecessary.
Therefore, this array element should be cached in a local variable and then be loaded
again using this local variable, skipping the redundant second array boundaries check:

Code instances:

    SwapUtils.sol.removeLiquidity - double load of minAmounts[i]
    Multicall.sol.aggregate - double load of calls[i]

Caching array length can save gas

Caching the array length is more gas efficient.
This is because access to a local variable in solidity is more efficient than query storage / calldata / memory.
We recommend to change from:

for (uint256 i=0; i<array.length; i++) { ... }

to:

uint len = array.length
for (uint256 i=0; i<len; i++) { ... }

Code instances:

    StableSwap.sol, _pooledTokens, 81
    GovernanceRouter.sol, _calls, 419
    SwapUtils.sol, balances, 558
    GovernanceRouter.sol, _localCalls, 244
    GovernanceRouter.sol, domains, 443
    TypedMemView.sol, memViews, 813
    GovernanceRouter.sol, _remoteCalls, 248
    RelayerFacet.sol, _transferIds, 164
    SwapUtils.sol, xp, 205
    ConnextPriceOracle.sol, tokenAddresses, 176
    SwapUtils.sol, pooledTokens, 844
    SwapUtils.sol, pooledTokens, 1039
    GovernanceRouter.sol, domains, 326
    SwapUtils.sol, pooledTokens, 1055

Prefix increments are cheaper than postfix increments

Prefix increments are cheaper than postfix increments.
Further more, using unchecked {++x} is even more gas efficient, and the gas saving accumulates every iteration and can make a real change
There is no risk of overflow caused by increamenting the iteration index in for loops (the ++i in for (uint256 i = 0; i < numIterations; ++i)).
But increments perform overflow checks that are not necessary in this case.
These functions use not using prefix increments (++x) or not using the unchecked keyword:

Code instances:

    change to prefix increment and unchecked: Queue.sol, i, 95
    change to prefix increment and unchecked: Merkle.sol, i, 34
    change to prefix increment and unchecked: GovernanceRouter.sol, i, 248
    change to prefix increment and unchecked: Queue.sol, i, 111
    change to prefix increment and unchecked: ConnextPriceOracle.sol, i, 176
    change to prefix increment and unchecked: Merkle.sol, i, 128
    change to prefix increment and unchecked: SwapUtils.sol, i, 591
    change to prefix increment and unchecked: SwapUtils.sol, i, 300
    change to prefix increment and unchecked: StableSwapFacet.sol, i, 415
    change to prefix increment and unchecked: SwapUtils.sol, i, 425
    change to prefix increment and unchecked: SwapUtils.sol, i, 924
    change to prefix increment and unchecked: GovernanceRouter.sol, i, 443
    change to prefix increment and unchecked: GovernanceRouter.sol, i, 244
    change to prefix increment and unchecked: Merkle.sol, i, 60
    change to prefix increment and unchecked: SwapUtils.sol, i, 254
    change to prefix increment and unchecked: SwapUtils.sol, i, 558
    change to prefix increment and unchecked: StableSwap.sol, i, 81

Unnecessary index init

In for loops you initialize the index to start from 0, but it already initialized to 0 in default and this assignment cost gas.
It is more clear and gas efficient to declare without assigning 0 and will have the same meaning:

Code instances:

    SwapUtils.sol, 1039
    StableSwapFacet.sol, 415
    Queue.sol, 71
    SwapUtils.sol, 924
    Multicall.sol, 16
    ConnextPriceOracle.sol, 176
    SwapUtils.sol, 405
    GovernanceMessage.sol, 97
    SwapUtils.sol, 425

Storage double reading. Could save SLOAD

Reading a storage variable is gas costly (SLOAD). In cases of multiple read of a storage variable in the same scope, caching the first read (i.e saving as a local variable) can save gas and decrease the
overall gas uses. The following is a list of functions and the storage variables that you read twice:

Code instances:

    SponsorVault.sol: relayerFeeCap is read twice in reimburseRelayerFees

Unnecessary default assignment

Unnecessary default assignments, you can just declare and it will save gas and have the same meaning.

Code instances:

    Version.sol (L#9) : uint8 public constant VERSION = 0;
    VersionFacet.sol (L#16) : uint8 internal immutable _version = 0;
    Version0.sol (L#9) : uint8 public constant VERSION = 0;
    BridgeFacet.sol (L#68) : uint16 public constant AAVE_REFERRAL_CODE = 0;

Rearrange state variables

You can change the order of the storage variables to decrease memory uses.

Code instances:

In TypedMemView.sol,rearranging the storage fields can optimize to: 2 slots from: 3 slots.
The new order of types (you choose the actual variables):
1. uint256
2. bytes29
3. uint8

In FacetHelper.sol,rearranging the storage fields can optimize to: 11 slots from: 12 slots.
The new order of types (you choose the actual variables):
1. bytes32
2. bytes32
3. uint256
4. uint256
5. uint256
6. address
7. uint32
8. uint32
9. uint32
10. address
11. address
12. address
13. address
14. address

In GovernanceRouter.sol,rearranging the storage fields can optimize to: 5 slots from: 6 slots.
The new order of types (you choose the actual variables):
1. uint256
2. uint256
3. XAppConnectionManager
4. address
5. uint32
6. uint32
7. address

In Replica.sol,rearranging the storage fields can optimize to: 4 slots from: 5 slots.
The new order of types (you choose the actual variables):
1. uint256
2. uint256
3. uint256
4. uint32
5. uint8

In ProposedOwnableUpgradeable.sol,rearranging the storage fields can optimize to: 6 slots from: 8 slots.
The new order of types (you choose the actual variables):
1. uint256
2. uint256
3. uint256
4. uint256
5. address
6. bool
7. bool
8. address

In Executor.sol,rearranging the storage fields can optimize to: 4 slots from: 5 slots.
The new order of types (you choose the actual variables):
1. bytes
2. uint256
3. uint256
4. address
5. uint16

In RelayerFeeMessage.sol,rearranging the storage fields can optimize to: 5 slots from: 6 slots.
The new order of types (you choose the actual variables):
1. uint256
2. uint256
3. uint256
4. uint256
5. uint8
6. uint8

Use bytes32 instead of string to save gas whenever possible

Use bytes32 instead of string to save gas whenever possible.
String is a dynamic data structure and therefore is more gas consuming then bytes32.

Code instance:

    TypedMemView.sol (L287), string memory err = string( abi.encodePacked("Type assertion failed. Got 0x", uint80(g), ". Expected 0x", uint80(e)) );

Short the following require messages

The following require messages are of length more than 32 and we think are short enough to short
them into exactly 32 characters such that it will be placed in one slot of memory and the require
function will cost less gas.
The list:

Code instances:

    Solidity file: SwapUtils.sol, In line 784, Require message length to shorten: 33, The message: Cannot get more than pool balance
    Solidity file: OZERC20.sol, In line 254, Require message length to shorten: 34, The message: ERC20: approve to the zero address
    Solidity file: OZERC20.sol, In line 253, Require message length to shorten: 36, The message: ERC20: approve from the zero address
    Solidity file: OZERC20.sol, In line 186, Require message length to shorten: 35, The message: ERC20: transfer to the zero address
    Solidity file: SwapUtils.sol, In line 697, Require message length to shorten: 33, The message: Cannot get more than pool balance
    Solidity file: GovernanceRouter.sol, In line 240, Require message length to shorten: 36, The message: !domains length matches calls length
    Solidity file: OZERC20.sol, In line 226, Require message length to shorten: 33, The message: ERC20: burn from the zero address
    Solidity file: TypedMemView.sol, In line 745, Require message length to shorten: 40, The message: TypedMemView/copyTo - Null pointer deref
    Solidity file: OZERC20.sol, In line 185, Require message length to shorten: 37, The message: ERC20: transfer from the zero address

Use != 0 instead of > 0

Using != 0 is slightly cheaper than > 0. (see code-423n4/2021-12-maple-findings#75 for similar issue)

Code instances:

    ProposedOwnableUpgradeable.sol, 276: change '_proposedOwnershipTimestamp > 0' to '_proposedOwnershipTimestamp != 0'
    SwapUtils.sol, 1007: change 'balance > 0' to 'balance != 0'
    StableSwapFacet.sol, 132: change 'balance > 0' to 'balance != 0'
    BridgeFacet.sol, 499: change '_amount > 0' to '_amount != 0'
    SponsorVault.sol, 258: change 'balance > 0' to 'balance != 0'
    AmplificationUtils.sol, 86: change 'futureA_ > 0' to 'futureA_ != 0'
    SwapUtils.sol, 954: change 'balance > 0' to 'balance != 0'
    SwapUtils.sol, 916: change 'balance > 0' to 'balance != 0'

Unnecessary cast

Code instance:

    IUpdaterManager Home.sol._setUpdaterManager - unnecessary casting IUpdaterManager(_updaterManager)

Use unchecked to save gas for certain additive calculations that cannot overflow

You can use unchecked in the following calculations since there is no risk to overflow:

Code instances:

    Replica.sol (L#143) - confirmAt[_newRoot] = block.timestamp + optimisticSeconds;

uint8 index

Due to how the EVM natively works on 256 numbers, using a 8 bit number here introduces additional costs as the EVM has to properly enforce the limits of this smaller type.
See the warning at this link: https://docs.soliditylang.org/en/v0.8.0/internals/layout_in_storage.html#layout-of-state-variables-in-storage
We recommend to use uint256 for the index in every for loop instead using uint8:

Code instances:

    StableSwapFacet.sol, uint8 i, 415
    Encoding.sol, uint8 i, 22
    Encoding.sol, uint8 i, 36
    StableSwap.sol, uint8 i, 81
    TypedMemView.sol, uint8 i, 158
    TypedMemView.sol, uint8 i, 146

Consider inline the following functions to save gas

You can inline the following functions instead of writing a specific function to save gas.
(see https://github.com/code-423n4/2021-11-nested-findings/issues/167 for a similar issue.)

Code instances

    TypedMemView.sol, nullView, { return NULL; }
    TypedMemView.sol, isNull, { return memView == NULL; }
    TypedMemView.sol, notNull, { return !isNull(memView); }
    Queue.sol, lastItem, { return _q.queue[_q.last]; }
    Queue.sol, isEmpty, { return _q.last < _q.first; }
    AmplificationUtils.sol, getAPrecise, { return _getAPrecise(self); }
    GovernanceMessage.sol, governor, { return _view.index(5, 32); }

Inline one time use functions

The following functions are used exactly once. Therefore you can inline them and save gas and improve code clearness.

Code instances:

    Message.sol, destination
    NomadBase.sol, _setUpdater
    GovernanceMessage.sol, serializeCall
    XAppConnectionManager.sol, _recoverWatcherFromSig
    XAppConnectionClient.sol, _isReplica

Change if -> revert pattern to require

Change if -> revert pattern to 'require' to save gas and improve code quality,
if (some_condition) {
revert(revert_message)
}

to: require(!some_condition, revert_message)

In the following locations:

Code instances:

    UpgradeBeaconProxy.sol, 85
    TypedMemView.sol, 568
    UpgradeBeaconController.sol, 39

Upgrade pragma to at least 0.8.4

Using newer compiler versions and the optimizer gives gas optimizations
and additional safety checks are available for free.

The advantages of versions 0.8.* over <0.8.0 are:

    1. Safemath by default from 0.8.0 (can be more gas efficient than library based safemath.)
    2. Low level inliner : from 0.8.2, leads to cheaper runtime gas. Especially relevant when the contract has small functions. For example, OpenZeppelin libraries typically have a lot of small helper functions and if they are not inlined, they cost an additional 20 to 40 gas because of 2 extra jump instructions and additional stack operations needed for function calls.
    3. Optimizer improvements in packed structs: Before 0.8.3, storing packed structs, in some cases used an additional storage read operation. After EIP-2929, if the slot was already cold, this means unnecessary stack operations and extra deploy time costs. However, if the slot was already warm, this means additional cost of 100 gas alongside the same unnecessary stack operations and extra deploy time costs.
    4. Custom errors from 0.8.4, leads to cheaper deploy time cost and run time cost. Note: the run time cost is only relevant when the revert condition is met. In short, replace revert strings by custom errors.

Code instances:

    XAppConnectionManager.sol
    NomadBase.sol
    Version0.sol
    XAppConnectionClient.sol
    UpgradeBeaconProxy.sol
    Home.sol
    TypeCasts.sol
    GovernanceMessage.sol

Do not cache msg.sender

We recommend not to cache msg.sender since calling it is 2 gas while reading a variable is more.

Code instances:

    https://github.com/code-423n4/2022-06-connext/tree/main/contracts/contracts/nomad-core/contracts/governance/GovernanceRouter.sol#L182
    https://github.com/code-423n4/2022-06-connext/tree/main/contracts/contracts/core/connext/helpers/ConnextPriceOracle.sol#L78
    https://github.com/code-423n4/2022-06-connext/tree/main/contracts/contracts/core/connext/facets/BridgeFacet.sol#L416