Ngone51 commented on a change in pull request #32287:
URL: https://github.com/apache/spark/pull/32287#discussion_r620785987
##########
File path:
core/src/main/scala/org/apache/spark/storage/ShuffleBlockFetcherIterator.scala
##########
@@ -613,6 +634,12 @@ final class ShuffleBlockFetcherIterator(
}
if (isNetworkReqDone) {
reqsInFlight -= 1
+ if (!buf.isInstanceOf[NettyManagedBuffer]) {
Review comment:
> For me this means we do not necessary have some extra free off-heap
memory by this.
Yes. that's true.
(So the original assuming here is that whenever there's a completed request,
there would be some freed memory.)
I have another idea here:
https://github.com/apache/spark/pull/32287#discussion_r620779976
Could you also take a look?
--
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:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
