Github user jerryshao commented on a diff in the pull request:
https://github.com/apache/spark/pull/2824#discussion_r19065507
--- Diff: core/src/main/scala/org/apache/spark/util/Utils.scala ---
@@ -272,20 +272,38 @@ private[spark] object Utils extends Logging {
/** Copy all data from an InputStream to an OutputStream */
def copyStream(in: InputStream,
out: OutputStream,
- closeStreams: Boolean = false): Long =
+ closeStreams: Boolean = false,
+ transferToEnabled: Boolean = true): Long =
--- End diff --
IMHO, I think NIO way will gain much better performance, especially in
shuffle. Though we met some issues with specific kernel version, mostly this
exception will not be met, I think we should balance the trade-off between
seldom met exception and performance gain.
If we hopes people to never use `transferTo`, why not directly remove this
code path and configuration, this will be more straightforward.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
