@mentalguy @bascule @mperham @halorgium Also some really good reading: https://blogs.oracle.com/dave/entry/a_race_in_locksupport_park …
@evanphx @mentalguy @mperham @halorgium I'd still like to know why this particular benchmark is deadlocking. It also deadlocks on rbx...
-
-
@bascule@mentalguy@mperham@halorgium You're using 2 mutexes instead of one in FC, for one. 2nd, wait isn't unlocking the right mutex. -
@evanphx@mentalguy@mperham@halorgium forget about FastCondition, it deadlocks with ConditionVariable -
@fowlduck yeah, I've got it... the other thread isn't guaranteed to be scheduled, even on MRI -
@bascule you would think that the first thread waiting on the mutex would be granted the lock immediately upon signal, though -
@fowlduck it's hard to keep "how the kernel schedules threads" and "how MRI schedules threads" separate in my head o_O -
@bascule that’s an argument over having separate scheduling, I suppose
End of conversation
New conversation -
-
-
@bascule@evanphx@mentalguy@mperham you have tried to wait on the mutex from both Threads. lock-free is good in this respect. -
@halorgium@evanphx@mentalguy@mperham perhaps I need to make a diagram, but the "waiter" thread releases the mutex when it waits...
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.