This change will check commit interrupt signal coming from other channels, e.g. newHead, reSubmit, before checking transaction level interrupt. This is to fix a corner case where an interrupt signal coming from another channel getting intercepted by transaction level interrupt, which resulted in an uncleaned state (potentially deadlock).
* eth, core: track block announcement and import timings
* eth: track total delay
* eth: add more context and prefix in logs
* eth/fetcher: fix invalid tracking of received at time
* put logs behind log.enable-block-tracking flag
* fix lint
* improve logging
* add comments
* add comments
This pull request resets the txlookup cache if chain reorg happens,
preventing them from remaining reachable. It addresses failures in
the hive tests.
Co-authored-by: rjl493456442 <garyrong0905@gmail.com>
* changed parallel universe HF to napoli HF
* integration-tests: update ipc path on ci tests (#1127)
* integration-tests: update ipc path on ci tests
* added devnetBorFlags in matic-cli-config.yml
---------
Co-authored-by: Pratik Patil <pratikspatil024@gmail.com>
* Revert "integration-tests: update ipc path on ci tests (#1127)"
This reverts commit 0660fac39b.
* using cancun for block-stm metadata instead of napoli
* added a check to verify tx dependencies and test cases
* fix in snapshot.chainConfig
* removed snapshot.*params.BorConfig using snapshot.chainConfig.Bor instead
* removed unnecessary if statement in ParallelStateProcessor.Process
* addressed comment
* small fix in commitTransactions
* dependency calculation bug fix in miner/worker.go
---------
Co-authored-by: Manav Darji <manavdarji.india@gmail.com>
Co-authored-by: Arpit Temani <temaniarpit27@gmail.com>
* should use Lock when mutating the flag
* same problem in MVHashMap.Write
* hole the rlock while reading WriteCell
---------
Co-authored-by: zhiqiangxu <652732310@qq.com>
