Proposal for adding O_DIRECT support into JDK 9

[Lu, Yingqi]

Hi Brian/Alan,

After taking another look at the code, I could not find an easy way to check the channel/file position inside IOUtil class for non-positional read/write (position is not provided as a parameter to FileChannel.read/write).

Do you think it is proper to just leave the channel position check inside FileChannelImpl like it is now?

Thanks,
Lucy

From: nio-dev [mailto:nio-dev-bounces at openjdk.java.net] On Behalf Of Lu, Yingqi
Sent: Thursday, September 14, 2017 9:40 AM
To: Alan Bateman <Alan.Bateman at oracle.com>; Brian Burkhalter <brian.burkhalter at oracle.com>
Cc: nio-dev at openjdk.java.net
Subject: RE: Proposal for adding O_DIRECT support into JDK 9

Hi Alan,

The remaining check in FileChannelImpl is to check if channel/file position is aligned. I think we still need the check, but we can move it to IOUtil class. Please let me know if there is anything I misunderstand.

I will submit version 13 today to incorporate your feedback.

Thanks,
Lucy

From: Alan Bateman [mailto:Alan.Bateman at oracle.com]
Sent: Thursday, September 14, 2017 5:25 AM
To: Lu, Yingqi <yingqi.lu at intel.com<mailto:yingqi.lu at intel.com>>; Brian Burkhalter <brian.burkhalter at oracle.com<mailto:brian.burkhalter at oracle.com>>
Cc: nio-dev at openjdk.java.net<mailto:nio-dev at openjdk.java.net>
Subject: Re: Proposal for adding O_DIRECT support into JDK 9


On 14/09/2017 00:50, Lu, Yingqi wrote:
Hi Brian,

Thank you for the feedback. Here is a newer version of the patch http://cr.openjdk.java.net/~kkharbas/8164900/webrev.12/<http://cr.openjdk.java.net/%7Ekkharbas/8164900/webrev.12/>

Just catching up on this again.

The check for alignment of the position and size of the remaining in readInto/writeFrom{NativeBuffer} looks right now. There is still a duplicate check in FileChannelImpl.read/write that I assume can be removed (it is easily defeated). I assume the protected modifier can be dropped from the new check methods in Util.

It looks like the first sentence in the FileStore::getBlockSize javadoc has been dropped in this version. A sentence to say that to returns the block size would be good to have as the first sentence.

-Alan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/nio-dev/attachments/20170914/854bb933/attachment-0001.html>