[ https://issues.apache.org/jira/browse/ORC-166?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15933208#comment-15933208 ]
Prasanth Jayachandran commented on ORC-166: ------------------------------------------- Some comments: - Can you please add test case for this? Something that uses BufferPoolMXBean to monitor direct BB allocations before and after reading ORC file. https://github.com/apache/hive/blob/master/llap-server/src/java/org/apache/hadoop/hive/llap/metrics/LlapDaemonJvmMetrics.java#L86-L88 - In CodecPool, use ConcurrentHashMap instead of synchronized blocks? - nit: Can you also add the comment "// Note: shim path does not care about levels and strategies (only used for decompression)." to ZlibCodec.reset() where it's more relevant? > add codec pool to ORC; make sure end is called on underlying codecs > ------------------------------------------------------------------- > > Key: ORC-166 > URL: https://issues.apache.org/jira/browse/ORC-166 > Project: ORC > Issue Type: Bug > Reporter: Sergey Shelukhin > Assignee: Sergey Shelukhin > Attachments: ORC-166.patch > > > Subj -- This message was sent by Atlassian JIRA (v6.3.15#6346)