[
https://issues.apache.org/jira/browse/HADOOP-11505?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15107415#comment-15107415
]
Alan Burlison commented on HADOOP-11505:
----------------------------------------
bq. perhaps it is a SNAPSHOT dependency?
Could be, I'll have a dig around and see if I can figure it out. The odd thing
is that if I rebuilt from the repo root nothing new is pulled down into the
maven repo but the build succeeds - most odd...
bq. it would be a great service to everyone if this time could be reduced
I've had a poke at this but haven't been successful so far, my first stab was
to try to change the antrun bits in the pom.xml so that cmake only got called
to regenerate the Makefile if the corresponding CMakeLists.txt was newer than
it but I got lost in the twisty maze between where maven finishes and the
antrun plugin starts. Ugh :-( The other potential problem is the javah parts,
I'm not sure but it looks like javah is getting called unnecessarily as well,
and of course if the header files are regenerated that will most likely trigger
a make recompile anyway. I know way more about make than I do about ant, and
way more about ant than maven plus the maven documentation is generally
dreadful anyway, sigh ...
bq. I can't duplicate this
It's completely reproducible for me, but I have a different set of patches
applied to you. I'll check to see if it behaves the same way on Linux tomorrow.
> Various native parts use bswap incorrectly and unportably
> ---------------------------------------------------------
>
> 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: Alan Burlison
> Fix For: 3.0.0
>
> Attachments: HADOOP-11505.001.patch, HADOOP-11505.003.patch,
> HADOOP-11505.004.patch, HADOOP-11505.005.patch, HADOOP-11505.006.patch,
> HADOOP-11505.007.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)
