[
https://issues.apache.org/jira/browse/MAHOUT-392?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12865429#action_12865429
]
Sean Owen commented on MAHOUT-392:
----------------------------------
I won't argue about it, since it's tiny, but they are not inlined by javac.
They may be inlined by a JIT, if/when it decides to compile and optimize the
method some time after the method has been used repeatedly. If this were a
performance-critical block, I wouldn't rely on this behavior happening at all,
or happening to the extent I'd like, and would help the compiler by just doing
it by hand. One thing javac can do, strangely, is inline static final
primitives. There's a limit to think thinking -- at some point obscuring code
by hand isn't worth it.
> Test cases for logGamma, Distribution.normal and Distribution.beta, fix for
> Distribution.normal
> -----------------------------------------------------------------------------------------------
>
> Key: MAHOUT-392
> URL: https://issues.apache.org/jira/browse/MAHOUT-392
> Project: Mahout
> Issue Type: Bug
> Reporter: Ted Dunning
> Fix For: 0.4
>
> Attachments: MAHOUT-392.patch
>
>
> I need access to the normal distribution to write test cases for random
> matrices. While testing that, I found it wasn't quite right so I provided an
> alternative implementation.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
