hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jitendra Nath Pandey (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HIVE-6639) Vectorization: Partition column names are not picked up.
Date Mon, 17 Mar 2014 18:49:45 GMT

    [ https://issues.apache.org/jira/browse/HIVE-6639?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13938195#comment-13938195
] 

Jitendra Nath Pandey commented on HIVE-6639:
--------------------------------------------

Ran tests locally. Only failures were show_create_table_serde.q and metadata_only_queries_with_filters.q
which are unrelated to this patch.

> Vectorization: Partition column names are not picked up.
> --------------------------------------------------------
>
>                 Key: HIVE-6639
>                 URL: https://issues.apache.org/jira/browse/HIVE-6639
>             Project: Hive
>          Issue Type: Bug
>            Reporter: Jitendra Nath Pandey
>            Assignee: Jitendra Nath Pandey
>         Attachments: HIVE-6639.2.patch, HIVE-6639.3.patch, HIVE-6639.4.patch, HIVE-6639.5.patch,
HIVE-6639.5.patch, HIVE-6639.6.patch
>
>
> The vectorized plan generation finds the list of partitioning columns from pruned-partition-list
using table scan operator. In some cases the list is coming as null. TPCDS query 27 can reproduce
this issue if the store_sales table is partitioned on ss_store_sk. The exception stacktrace
is :
> {code}
> Caused by: java.lang.NullPointerException
> 	at org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getInputColumnIndex(VectorizationContext.java:166)
> 	at org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getColumnVectorExpression(VectorizationContext.java:240)
> 	at org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getVectorExpression(VectorizationContext.java:287)
> 	at org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getVectorExpressions(VectorizationContext.java:267)
> 	at org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getVectorExpressions(VectorizationContext.java:255)
> 	at org.apache.hadoop.hive.ql.exec.vector.VectorMapJoinOperator.<init>(VectorMapJoinOperator.java:116)
> 	... 42 more
> {code}



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message