Prevents when two threads execute the assertions and two threads passed the check before any one is acquiring the SessionLock.
Diff Detail
Diff Detail
Event Timeline
Comment Actions
Does the original bug show up with TSan on any existing/checked in tests, or could TSan catch it in some new test that isn't checked in yet/should be checked in?
Comment Actions
I think it's more of an race condition than a data race. I will write a test case and check whether TSan can catch it. But the idea here is createJITDylib and getJITdylibByName must be executed by the same thread,
Comment Actions
Tentatively LGTM. Was this is generating a TSan error, or a crash? I would expect this to pass TSan, so writing a test case is probably not practical.