[
https://issues.apache.org/jira/browse/HADOOP-11665?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15020937#comment-15020937
]
Steve Loughran commented on HADOOP-11665:
-----------------------------------------
Given x86 is the CPU for almost all hadoop systems, it's assumptions live in
the code: byte ordering, whether misaligned accesses are permitted, etc.
There's no assumptions about cache lines in the code —but that's because
there's been no optimisation of struct layout for cache efficiency
> Provide and unify cross platform byteorder support in native code
> -----------------------------------------------------------------
>
> Key: HADOOP-11665
> URL: https://issues.apache.org/jira/browse/HADOOP-11665
> Project: Hadoop Common
> Issue Type: Bug
> Components: native, util
> Affects Versions: 2.4.1, 2.6.0
> Environment: PowerPC Big Endian & other Big Endian platforms, Sparc,
> Arm
> Reporter: Binglin Chang
> Assignee: Binglin Chang
> Labels: BB2015-05-TBR
> Attachments: HADOOP-11665.001.patch, HADOOP-11665.002.patch
>
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
