Github user squito commented on the issue: https://github.com/apache/spark/pull/21811 > Does it make sense to release byteChannel at deallocate()? you could, just to let GC kick in a *bit* earlier, but I don't think its going to make a big difference. (Netty's ByteBufs must be since they may be owned by netty's internal pools and never gc'ed)
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org