[
https://issues.apache.org/jira/browse/HADOOP-10047?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated HADOOP-10047:
-----------------------------
Attachment: HADOOP-10047-WIP.patch
[~cnauroth] - I do not see a way out of this other than a down-cast, without
adding a method to the interface & breaking source-compat for the API.
The patch attached is code-complete for Zlib, but does it slightly
inefficiently with 2 memory copies rather than passing the buffer directly into
the JNI layer.
This allows the API to accept both direct & indirect buffers, which was
convenient for testing.
> Allow Compressor/Decompressor APIs to expose a Direct ByteBuffer API
> --------------------------------------------------------------------
>
> Key: HADOOP-10047
> URL: https://issues.apache.org/jira/browse/HADOOP-10047
> Project: Hadoop Common
> Issue Type: New Feature
> Components: io
> Reporter: Gopal V
> Assignee: Gopal V
> Labels: compression
> Attachments: DirectCompressor.html, DirectDecompressor.html,
> HADOOP-10047-WIP.patch
>
>
> With the Zero-Copy reads in HDFS (HDFS-5260), it becomes important to perform
> all I/O operations without copying data into byte[] buffers or other buffers
> which wrap over them.
> This is a proposal for adding new DirectCompressor and DirectDecompressor
> interfaces to the io.compress, to indicate codecs which want to surface the
> direct buffer layer upwards.
> The implementation may or may not copy the buffers passed in, but should work
> with direct heap/mmap buffers and cannot assume .array() availability.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
