impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alex Behm (Code Review)" <>
Subject [Impala-ASF-CR] IMPALA-5483: Automatically disable codegen for small queries
Date Fri, 23 Jun 2017 19:05:40 GMT
Alex Behm has posted comments on this change.

Change subject: IMPALA-5483: Automatically disable codegen for small queries

Patch Set 4:

File fe/src/main/java/org/apache/impala/planner/

Line 151: 
> Happy to do this but will defer the work until we have a clear decision abo
I'm on board with merging this change with a 50K default.
File fe/src/main/java/org/apache/impala/util/

Line 60:       maxRowsProcessed_ = Math.max(maxRowsProcessed_, numRows);
> In that case the scan returns every row it scans and hits the limit after p
I'm thinking about the case:
(missingStats && !scan.hasLimit() && !scan.getConjuncts().isEmpty())

That should be invalid imo, but today it is valid.
File testdata/workloads/functional-query/queries/QueryTest/codegen.test:

Line 2: ---- QUERY
> To confirm that it's actually plumbed all the way through to the backend.
Don't we have other tests that confirm the codegen_hint is honored? Validating plan choices
with EE tests and inspecting the profile seems odd to me.
File tests/common/

Line 139:                                       disable_codegen_rows_threshold_options=[5000],
> The reasoning below doesn't apply since disable_codegen is always on here. 
Works for me. I think we should generally reconsider the use of this function in various places,
e.g. we use it for our TPCH/DS functional tests.

To view, visit
To unsubscribe, visit

Gerrit-MessageType: comment
Gerrit-Change-Id: I273bcee58641f5b97de52c0b2caab043c914b32e
Gerrit-PatchSet: 4
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-Owner: Tim Armstrong <>
Gerrit-Reviewer: Alex Behm <>
Gerrit-Reviewer: Juan Yu <>
Gerrit-Reviewer: Michael Ho
Gerrit-Reviewer: Michael Ho <>
Gerrit-Reviewer: Tim Armstrong <>
Gerrit-HasComments: Yes

View raw message