[ 
https://issues.apache.org/jira/browse/HADOOP-15033?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16300132#comment-16300132
 ] 

ASF GitHub Bot commented on HADOOP-15033:
-----------------------------------------

Github user dchuyko commented on the issue:

    https://github.com/apache/hadoop/pull/291
  
    Ok, I changed the patch according to the review.
    
    1. There's now isJavaSpecAtLeast() in Shell. This is a part I like.
    
    2. In DataChecksum there are now: racy state "useJava9Crc32C" which is set 
to the above by default and falls to false on error. New Java9Crc32CFactory now 
holds static method handles stuff to let it be initialized on first usage. And 
there is now also LOG to possibly log errors that never happen. To me it looks 
probably like too much meat for code that won't be called ever. And in case of 
NOP logger no one will see the error if it happens still. From the other hand 
all that may look more safe.


> Use java.util.zip.CRC32C for Java 9 and above
> ---------------------------------------------
>
>                 Key: HADOOP-15033
>                 URL: https://issues.apache.org/jira/browse/HADOOP-15033
>             Project: Hadoop Common
>          Issue Type: Improvement
>          Components: performance, util
>    Affects Versions: 3.0.0
>            Reporter: Dmitry Chuyko
>              Labels: Java9, common, jdk9
>         Attachments: HADOOP-15033.001.patch, HADOOP-15033.001.patch, 
> HADOOP-15033.002.patch, HADOOP-15033.003.patch, HADOOP-15033.003.patch, 
> HADOOP-15033.004.patch
>
>
> java.util.zip.CRC32C implementation is available since Java 9.
> https://docs.oracle.com/javase/9/docs/api/java/util/zip/CRC32C.html
> Platform specific assembler intrinsics make it more effective than any pure 
> Java implementation.
> Hadoop is compiled against Java 8 but class constructor may be accessible 
> with method handle on 9 to instances implementing Checksum in runtime.



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to