[
https://issues.apache.org/jira/browse/HADOOP-11505?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14642807#comment-14642807
]
Alan Burlison commented on HADOOP-11505:
----------------------------------------
There's another potential problem lurking here as well - I'm assuming the
contents of KVBuffers are read in across the wire into instances of the the
struct, where the wire format is 2 32-bit length fields followed by two
variable-length concatenated byte buffers. However the code above, and
elsewhere in this struct, assumes that there is no padding between any of the
fields in the struct. That may not always be the case.
> hadoop-mapreduce-client-nativetask uses bswap where be32toh is needed,
> doesn't work on non-x86
> ----------------------------------------------------------------------------------------------
>
> Key: HADOOP-11505
> URL: https://issues.apache.org/jira/browse/HADOOP-11505
> Project: Hadoop Common
> Issue Type: Bug
> Affects Versions: 3.0.0
> Reporter: Colin Patrick McCabe
> Assignee: Colin Patrick McCabe
> Labels: BB2015-05-TBR
> Attachments: HADOOP-11505.001.patch, HADOOP-11505.003.patch,
> HADOOP-11505.004.patch
>
>
> hadoop-mapreduce-client-nativetask fails to use x86 optimizations in some
> cases. Also, on some alternate, non-x86, non-ARM architectures the generated
> code is incorrect. Thanks to Steve Loughran and Edward Nevill for finding
> this.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
