Github user rxin commented on the issue: https://github.com/apache/spark/pull/19222 Sorry this thread is too long for me to follow. I might be bringing up a point that has been brought up before. @kiszk did your perf tests take into account megamorphic callsites? It seems to me from a quick cursory look the benchmark result might not be accurate for real workloads if there are only one implementations of the MemoryBlock loaded.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org