[
https://issues.apache.org/jira/browse/HADOOP-7753?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13136658#comment-13136658
]
Eli Collins commented on HADOOP-7753:
-------------------------------------
Aside from Nathan's great feedback I have the following minor comments based on
the previous version of the patch. Will check out the latest.
Shouldn't the catch of the ULE in syncFileRangeIfPossible set
syncFileRangePossible to false?
In manual_sync_file_range in NativeIO.c might be worth referencing RedHat BZ
518581 which has the rationale as to why the glibc API isnt't exposed w the
recommendation for the direct syscall.
In NativeIO.java would be good to javadoc that these methods throw IOEs with
the relevant errno.
I found the wording "each readahead overlaps 50% with the previous readahead" a
little confusing (had to read the code to understand the comment), perhaps
something like "only issue a new readahead if we've already advanced past the
halfway mark of the last one" would be more clear.
> Support fadvise and sync_data_range in NativeIO, add ReadaheadPool class
> ------------------------------------------------------------------------
>
> Key: HADOOP-7753
> URL: https://issues.apache.org/jira/browse/HADOOP-7753
> Project: Hadoop Common
> Issue Type: Sub-task
> Components: io, native
> Affects Versions: 0.23.0
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Attachments: hadoop-7753.txt, hadoop-7753.txt, hadoop-7753.txt,
> hadoop-7753.txt
>
>
> This JIRA adds JNI wrappers for sync_data_range and posix_fadvise. It also
> implements a ReadaheadPool class for future use from HDFS and MapReduce.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
