Tim Armstrong has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/16419 )

Change subject: IMPALA-10078: Proper codegen for KuduPartitionExpr
......................................................................


Patch Set 3:

(3 comments)

Great to see this finally removed. I think we need to reach closure on the 
testing, and then I'd like to do a final close review of the codegen logic just 
to make sure it's correct.

http://gerrit.cloudera.org:8080/#/c/16419/3//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/16419/3//COMMIT_MSG@12
PS3, Line 12:  - TODO: I don't think we have good test coverage on this, I 
tried to
One sanity check I can think of is to write to a hash partitioned table and 
check the profile to make sure that the different instances of the writer got 
similar numbers of rows. E.g. if you insert values [0,9999] into a hash 
partitioned table with 3 hash partitions, you could check, if, say each 
partition got between 3000 and 4000 values inserted into it.


http://gerrit.cloudera.org:8080/#/c/16419/3/be/src/exec/kudu-util-ir.cc
File be/src/exec/kudu-util-ir.cc:

http://gerrit.cloudera.org:8080/#/c/16419/3/be/src/exec/kudu-util-ir.cc@54
PS3, Line 54:   // TODO: codegen this to eliminate branching on type.
I think we can remove this since we use ColumnType::ToIR() and replace the 
argument with a constant.


http://gerrit.cloudera.org:8080/#/c/16419/3/be/src/exprs/kudu-partition-expr.cc
File be/src/exprs/kudu-partition-expr.cc:

http://gerrit.cloudera.org:8080/#/c/16419/3/be/src/exprs/kudu-partition-expr.cc@157
PS3, Line 157: "class.impala::Status"
Can you make this Status::LLVM_CLASS_NAME to fit the current pattern.



--
To view, visit http://gerrit.cloudera.org:8080/16419
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ifcae34f71b407837e2c5f1b97aa230e490a268df
Gerrit-Change-Number: 16419
Gerrit-PatchSet: 3
Gerrit-Owner: Daniel Becker <daniel.bec...@cloudera.com>
Gerrit-Reviewer: Csaba Ringhofer <csringho...@cloudera.com>
Gerrit-Reviewer: Daniel Becker <daniel.bec...@cloudera.com>
Gerrit-Reviewer: Impala Public Jenkins <impala-public-jenk...@cloudera.com>
Gerrit-Reviewer: Qifan Chen <qc...@cloudera.com>
Gerrit-Reviewer: Thomas Tauber-Marshall <tmarsh...@cloudera.com>
Gerrit-Reviewer: Tim Armstrong <tarmstr...@cloudera.com>
Gerrit-Comment-Date: Mon, 14 Sep 2020 16:30:35 +0000
Gerrit-HasComments: Yes

Reply via email to