Balazs Hevele has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/24089 )

Change subject: IMPALA-14850: Codegen tuple DeepCopy for hash join
......................................................................


Patch Set 23:

(4 comments)

http://gerrit.cloudera.org:8080/#/c/24089/22/be/src/runtime/buffered-tuple-stream-test.cc
File be/src/runtime/buffered-tuple-stream-test.cc:

http://gerrit.cloudera.org:8080/#/c/24089/22/be/src/runtime/buffered-tuple-stream-test.cc@406
PS22, Line 406:
> maybe "use_codegen"?
Done


http://gerrit.cloudera.org:8080/#/c/24089/22/be/src/runtime/buffered-tuple-stream-test.cc@579
PS22, Line 579: eStreamTest : pub
> This is not new code, but this seems to be also frequently used, just nulla
All Test classes have different variants, e.g:
-SimpleTupleStreamTest has a single non-nullable tuple
-MultiTupleStreamTest has 3 non-nullable tuples
-MultiNullableTupleStreamTest has 1 non-nullable and 2 nullable tuples
Hence, I think it's best to keep them separate.


http://gerrit.cloudera.org:8080/#/c/24089/22/be/src/runtime/buffered-tuple-stream-test.cc@605
PS22, Line 605:   }
              : };
              :
              : // Tests with multiple NULLable tuples per row.
              : class MultiNullableTupleStreamTest : public 
SimpleTupleStreamTest {
              :  protected:
              :   virtual void CreateDescriptors() {
              :     vector<bool> nullable_tuples;
              :     nullable_tuples.push_back(false);
              :     nullable_tuples.push_back(true);
> This functions look the same - can't we unify them somehow?
I modified the base class so that it handles nullptr for unavailable descs, so 
these classes don't need to override the function anymore.


http://gerrit.cloudera.org:8080/#/c/24089/22/be/src/runtime/buffered-tuple-stream-test.cc@2067
PS22, Line 2067:  stream(
> Not sure if worth the effort, but TestValues could pick up descriptor and c
I added a function to get codegen function from the descriptor. With that, the 
call sites remain unchanged.
Currently, the template argument is not enough to deduce the row 
descriptor/codegen fn, e.g. both int_desc_ and big_row_desc_ call it with <int>.
I would prefer not to add further template magic with this change.



--
To view, visit http://gerrit.cloudera.org:8080/24089
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I63e32babdbaf56095478c6c66afb9cb91189f946
Gerrit-Change-Number: 24089
Gerrit-PatchSet: 23
Gerrit-Owner: Balazs Hevele <[email protected]>
Gerrit-Reviewer: Balazs Hevele <[email protected]>
Gerrit-Reviewer: Csaba Ringhofer <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Joe McDonnell <[email protected]>
Gerrit-Reviewer: Michael Smith <[email protected]>
Gerrit-Reviewer: Yida Wu <[email protected]>
Gerrit-Comment-Date: Wed, 08 Jul 2026 08:34:26 +0000
Gerrit-HasComments: Yes

Reply via email to