Github user squito commented on the issue:
https://github.com/apache/spark/pull/21456
From my understanding, that option is only available with G1GC, which is
not really a good fit for spark (forget the exact details but something about
humongous allocations which are common with all the large byte buffers typical
for spark).--- --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
