hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chris Douglas (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-10924) Add a new instrumented read-write lock
Date Mon, 10 Oct 2016 06:03:21 GMT

    [ https://issues.apache.org/jira/browse/HDFS-10924?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15561397#comment-15561397

Chris Douglas commented on HDFS-10924:

bq. Personally feels the following change would be more readable: [snip]
This follows [guidance|https://docs.oracle.com/javase/8/docs/api/java/lang/System.html#nanoTime--]
for {{System.nanoTime}}.

The HDFS {{InstrumentedLock}} class contains nearly identical functionality. If Common will
introduce other instrumented locks, then this should replace what's in HDFS.

{{AutoCloseableReadLock}} and {{AutoCloseableReadLock}} are redundant with the base impl,
mod {{isLocked}} (which returns {{isWriteLocked}} for {{AutoCloseableReadLock}}). As in {{o.a.h.hdfs.InstrumentedLock}},
this should not build around {{AutoCloseableLock}} as the base API, but the classes in {{java.util.concurrent.locks}}.
Specifically, an {{InstrumentedReadWriteLock}} should return instrumented locks for {{readLock()}}
and {{writeLock()}}. It's up to the caller to wrap what's returned in an {{AutoCloseableLock}},
if desired.

> Add a new instrumented read-write lock
> --------------------------------------
>                 Key: HDFS-10924
>                 URL: https://issues.apache.org/jira/browse/HDFS-10924
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: datanode
>            Reporter: Jingcheng Du
>            Assignee: Jingcheng Du
>         Attachments: HDFS-10924-2.patch, HDFS-10924-3.patch, HDFS-10924-4.patch, HDFS-10924.patch
> Add a new instrumented read-write lock in hadoop common, so that the HDFS-9668 can use
this to improve the locking in FsDatasetImpl

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org

View raw message