enha: make save_pending_execution cheaper

[carneiro-cw]

PR Type

Enhancement

---

Description

• Implemented PartialEq trait for EvmInput struct in evm_input.rs, enabling efficient comparison with transaction input and block header.
• Optimized save_pending_execution method in inmemory.rs:
  • Replaced write lock with upgradable read lock for better concurrency.
  • Utilized new PartialEq implementation for simplified conflict checking.
  • Upgraded to write lock only when necessary, reducing lock contention.
• These changes aim to improve performance and reduce overhead in the transaction execution process.

---

Changes walkthrough 📝

Relevant files

Enhancement

evm_input.rs Add PartialEq implementation for EvmInput                                src/eth/executor/evm_input.rs Implemented PartialEq trait for EvmInput struct, comparing it with a tuple of (&TransactionInput, &PendingBlockHeader) The new implementation allows for efficient comparison of EvmInput with transaction input and block header +13/-0    inmemory.rs Optimize save_pending_execution with upgradable read lock src/eth/storage/temporary/inmemory.rs Added RwLockUpgradableReadGuard import from parking_lot Modified save_pending_execution method to use upgradable read lock Simplified conflict checking logic using the new PartialEq implementation Upgraded read lock to write lock only when necessary +7/-4

---

💡 PR-Agent usage: Comment /help "your question" on any pull request to receive relevant information

[github-actions]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

⏱️ Estimated effort to review: 3 🔵🔵🔵⚪⚪

🧪 No relevant tests

🔒 No security concerns identified

⚡ Recommended focus areas for review Performance Concern The new PartialEq implementation compares multiple fields, which could be expensive for frequent comparisons. Consider if all these comparisons are necessary or if a subset of fields would suffice. Error Handling The error handling for TransactionEvmInputMismatch is now done before upgrading the lock. Ensure this doesn't introduce any race conditions or inconsistencies. Concurrency Management The change from write lock to upgradable read lock improves concurrency, but ensure that all necessary modifications are done after upgrading to write lock.

[github-actions]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Score
Possible issue	Use a fallible lock upgrade to prevent potential panics and deadlocks The RwLockUpgradableReadGuard::upgrade operation might panic if another thread has acquired a write lock. Consider using a fallible upgrade method or restructure the code to avoid potential deadlocks. src/eth/storage/temporary/inmemory.rs [137] -let mut pending_block = RwLockUpgradableReadGuard::<InMemoryTemporaryStorageState>::upgrade(pending_block); +let mut pending_block = match RwLockUpgradableReadGuard::<InMemoryTemporaryStorageState>::try_upgrade(pending_block) { + Ok(guard) => guard, + Err(_) => return Err(StratusError::LockUpgradeFailure), +}; Suggestion importance[1-10]: 9 Why: This suggestion addresses a critical concurrency issue. Using a fallible upgrade method prevents potential panics and deadlocks, significantly improving the robustness and reliability of the code in multi-threaded scenarios.	9
General	Utilize the implemented PartialEq trait for more consistent and maintainable code Consider using the PartialEq trait implementation for EvmInput to simplify the comparison in save_pending_execution. This would make the code more consistent and easier to maintain. src/eth/executor/evm_input.rs [128-133] -if tx.evm_input != (&tx.input, &pending_block.block.header) { - let expected_input = EvmInput::from_eth_transaction(&tx.input, &pending_block.block.header); +let expected_input = EvmInput::from_eth_transaction(&tx.input, &pending_block.block.header); +if tx.evm_input != expected_input { return Err(StratusError::TransactionEvmInputMismatch { expected: Box::new(expected_input), actual: Box::new(tx.evm_input.clone()), }); } Suggestion importance[1-10]: 7 Why: The suggestion improves code consistency and readability by leveraging the newly implemented PartialEq trait. It simplifies the comparison logic, making it more maintainable and less error-prone.	7

[gabriel-aranha-cw]

Final benchmark:
Run ID: bench-4271986061

Git Info:

• Branch: main
• Commit: 701141a
• PR: enha: make save_pending_execution cheaper #1911

Configuration:

• Target Account Strategy: Default

RPS Stats: Max: 1232.00, Min: 535.00, Avg: 1020.91, StdDev: 71.21
TPS Stats: Max: 1237.00, Min: 874.00, Avg: 991.11, StdDev: 74.34

Plot: View Plot