[
https://issues.apache.org/jira/browse/SPARK-24498?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16543336#comment-16543336
]
Marco Gaido commented on SPARK-24498:
-------------------------------------
[~maropu] yes, I remembered I had some troubles compiling the generated code
with the jdk compiler too. There is also one case (which I saw in the branch
you prepared you addressed generating the proper code according to the chosen
compiler) in which there isn't really a way to make both them happy. In other
cases, when there is a form which works fine on both, I think it would be great
to use it. So I agree with your proposal.
My only concern is that as of now we have no way to check the compilation with
JDK, so it would be probably hard to enforce that we correct all the problems
and/or we don't introduce new ones. So the risk is that the effort spent on
that task could be not so useful...
> Add JDK compiler for runtime codegen
> ------------------------------------
>
> Key: SPARK-24498
> URL: https://issues.apache.org/jira/browse/SPARK-24498
> Project: Spark
> Issue Type: New Feature
> Components: SQL
> Affects Versions: 2.3.0
> Reporter: Xiao Li
> Priority: Major
>
> In some cases, JDK compiler can generate smaller bytecode and take less time
> in compilation compared to Janino. However, in some cases, Janino is better.
> We should support both for our runtime codegen. Janino will be still our
> default runtime codegen compiler.
> See the related JIRAs in DRILL:
> - https://issues.apache.org/jira/browse/DRILL-1155
> - https://issues.apache.org/jira/browse/DRILL-4778
> - https://issues.apache.org/jira/browse/DRILL-5696
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
