[Impala-CR](cdh5-trunk) Specify whether to clone IR function in GetFunction() instead of ReplaceCallSites()

["Skye Wanderman-Milne (Code Review)" <ge...@cloudera.org>]

Skye Wanderman-Milne has posted comments on this change.

Change subject: Specify whether to clone IR function in GetFunction() instead of ReplaceCallSites()
......................................................................


Patch Set 1:

(4 comments)

http://gerrit.cloudera.org:8080/#/c/2112/1/be/src/codegen/llvm-codegen.cc
File be/src/codegen/llvm-codegen.cc:

Line 550: LlvmCodeGenTest
> If this is only needed for the test, maybe we can just get the test to do i
Done. I changed the test to not modify an already-jitted function since we only do that in the test, and won't be able to do it once we switch to MCJIT anyway.


http://gerrit.cloudera.org:8080/#/c/2112/1/be/src/exec/aggregation-node.cc
File be/src/exec/aggregation-node.cc:

Line 809: int replaced
> nits: Defining this in the outer scope (like before) also seems fine. Same 
Done


http://gerrit.cloudera.org:8080/#/c/2112/1/be/src/exec/partitioned-hash-join-node.cc
File be/src/exec/partitioned-hash-join-node.cc:

Line 1713:  // Depends on join_op_
> Missed this one. I'm not sure that it matters too much though.
Done


Line 1717:   switch (join_op_) {
> ProcessProbeBatch() seems rather complicated and we may easily get some of 
I ran query_test/test_join_queries.py. I'm assuming that exercises all the different joins.


-- 
To view, visit http://gerrit.cloudera.org:8080/2112
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I8eabb5c062ee223c5de9df40aacfdc9dcda5ba63
Gerrit-PatchSet: 1
Gerrit-Project: Impala
Gerrit-Branch: cdh5-trunk
Gerrit-Owner: Skye Wanderman-Milne <sk...@cloudera.com>
Gerrit-Reviewer: Michael Ho <kw...@cloudera.com>
Gerrit-Reviewer: Skye Wanderman-Milne <sk...@cloudera.com>
Gerrit-Reviewer: Tim Armstrong <ta...@cloudera.com>
Gerrit-HasComments: Yes