hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Allen Wittenauer (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (HADOOP-11505) hadoop-mapreduce-client-nativetask uses bswap where be32toh is needed, doesn't work on non-x86
Date Thu, 26 Nov 2015 17:55:11 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-11505?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15029137#comment-15029137
] 

Allen Wittenauer edited comment on HADOOP-11505 at 11/26/15 5:54 PM:
---------------------------------------------------------------------

Yeah, Darwin (at least as of El Capitan aka 10.11) doesn't appear to provide either of those
options.  Instead, it uses <libkern/OSByteOrder.h> and it's own functions: 

{code}
OSSwapHostToBigInt16(x)
OSSwapHostToBigInt32(x)
OSSwapHostToBigInt64(x)
OSSwapHostToLittleInt16(x)
OSSwapHostToLittleInt32(x)
OSSwapHostToLittleInt64(x)
{code}

and vice versa:

{code}
OSSwapLittleToHostInt16(x)
OSSwapLittleToHostInt32(x)
OSSwapLittleToHostInt64(x)
OSSwapBigToHostInt16(x)
OSSwapBigToHostInt32(x)
OSSwapBigToHostInt64(x)
{code}

I think that covers all of the *relevant* ones.  There are more, however....


was (Author: aw):
Yeah, Darwin (at least as of El Capitan aka 10.11) doesn't appear to provide either of those
options.  Instead, it uses <libkern/OSByteOrder.h> and it's own functions: 

{code}
OSSwapHostToBigInt16(x)
OSSwapHostToBigInt32(x)
OSSwapHostToBigInt64(x)
OSSwapHostToLittleInt16(x)
OSSwapHostToLittleInt32(x)
OSSwapHostToLittleInt64(x)
{code}

and vice versa:

{code}
OSSwapLittleToHostInt16(x)
OSSwapLittleToHostInt32(x)
OSSwapLittleToHostInt64(x)
OSSwapBigToHostInt16(x)
OSSwapBigToHostInt32(x)
OSSwapBigToHostInt64(x)
{code}

I think that covers them all. 

> 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
>         Attachments: HADOOP-11505.001.patch, HADOOP-11505.003.patch, HADOOP-11505.004.patch,
HADOOP-11505.005.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)

Mime
View raw message