JoshRosen commented on a change in pull request #27246:
[SPARK-30536][CORE][SQL] Sort-merge join operator spilling performance
improvements
URL: https://github.com/apache/spark/pull/27246#discussion_r369918625
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/ExternalAppendOnlyUnsafeRowArray.scala
##########
@@ -106,6 +107,8 @@ private[sql] class ExternalAppendOnlyUnsafeRowArray(
def add(unsafeRow: UnsafeRow): Unit = {
if (numRows < numRowsInMemoryBufferThreshold) {
inMemoryBuffer += unsafeRow.copy()
+ numRows += 1
Review comment:
I think that we need to increment `numRows` and `modificationsCount` in both
branches, not just this one: if we only update it in this branch then
`length()` will return the wrong result.
If you want to know how many rows are in the in-memory buffer then I'd
define a separate `numRowBufferedInMemory` variable; that will also help to
make the code in `MergerIterator` easier to understand.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
