Closed laurynas-biveinis closed 1 year ago
@luqun has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.
@laurynas-biveinis has updated the pull request. You must reimport the pull request before landing.
@luqun has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.
@laurynas-biveinis has updated the pull request. You must reimport the pull request before landing.
Rebased and retested with no changes
@luqun has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.
Please rebase this PR.
@laurynas-biveinis has updated the pull request. You must reimport the pull request before landing.
@luqun , rebased
@luqun has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.
This has a minor effect for DD: by making MyRocks behave like InnoDB WRT locking, at least one more existing testcase can be run as-is under both DDSEs (main.status)
This has a minor effect for DD: by making MyRocks behave like InnoDB WRT locking, at least one more existing testcase can be run as-is under both DDSEs (main.status)
Currently, do you try to run main test suite under myrocks dd and find out main.status need this change?
This has a minor effect for DD: by making MyRocks behave like InnoDB WRT locking, at least one more existing testcase can be run as-is under both DDSEs (main.status)
Currently, do you try to run main test suite under myrocks dd and find out main.status need this change?
Yes, but that was not the motivation to do it. If this PR did not exist, I'd just skip the testcase
Since 5.7, all the THR_LOCK uses cases for InnoDB are fully covered by MDL locking and thus InnoDB removed THR_LOCK-based locking. The same MDL lock coverage applies to MyRocks as well, thus remove THR_LOCK-based locking there too.
References: