[
https://issues.apache.org/jira/browse/HDFS-8157?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14529275#comment-14529275
]
Colin Patrick McCabe commented on HDFS-8157:
--------------------------------------------
Thanks for this, [~arpitagarwal].
I don't think we should add {{DataNode#skipNativeIoCheckForTesting}}. To
simulate locking memory without adding a dependency on NativeIO, then just
create a custom cache manipulator. This custom manipulator can always return
true for {{verifyCanMlock}}. There are some other unit tests doing this.
{code}
public void releaseReservedSpace(long bytesToRelease, boolean
releaseLockedMemory);
{code}
I would rather have a separate function for releasing the memory than overload
the meaning of this one.
Maybe I am missing something, but I don't understand the purpose behind
{{releaseRoundDown}}. Why would we round down to a page size when allocating
or releasing memory?
> Writes to RAM DISK reserve locked memory for block files
> --------------------------------------------------------
>
> Key: HDFS-8157
> URL: https://issues.apache.org/jira/browse/HDFS-8157
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode
> Reporter: Arpit Agarwal
> Assignee: Arpit Agarwal
> Attachments: HDFS-8157.01.patch
>
>
> Per discussion on HDFS-6919, the first step is that writes to RAM disk will
> reserve locked memory via the FsDatasetCache.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
