harmony-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Paulex Yang (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HARMONY-3037) [classlib][nio]DatagramChannelImpl and SocketChannelImpl lack support for direct byte buffer
Date Mon, 29 Jan 2007 10:21:49 GMT

    [ https://issues.apache.org/jira/browse/HARMONY-3037?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12468213
] 

Paulex Yang commented on HARMONY-3037:
--------------------------------------

Leo, I still have some concerns on your new patch, for example, the DatagramSocketImpl.write()/send(),
seems you tried to avoid the array copy by get ByteBuffer.array() directly, but if the ByteBuffer
is got by ByteBuffer.slice(), the arrayOffset() needs to be considered. Would you have a look
on this issue?

> [classlib][nio]DatagramChannelImpl  and SocketChannelImpl lack support for direct byte
buffer 
> ----------------------------------------------------------------------------------------------
>
>                 Key: HARMONY-3037
>                 URL: https://issues.apache.org/jira/browse/HARMONY-3037
>             Project: Harmony
>          Issue Type: Improvement
>          Components: Classlib
>            Reporter: Leo Li
>         Attachments: patch-3037.zip
>
>
> As spec says, "Given a direct byte buffer, the Java virtual machine will make a best
effort to perform native I/O operations directly upon it. That is, it will attempt to avoid
copying the buffer's content to (or from) an intermediate buffer before (or after) each invocation
of one of the underlying operating system's native I/O operation". However, in Harmony nio,
DatagramChannelImpl and SocketChannelImpl do not differentiate whether the byte buffer is
direct or not.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message