lucenenet-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Nicholas Paldino (JIRA)" <>
Subject [jira] Commented: (LUCENENET-230) TestOpenBitSet
Date Fri, 13 Nov 2009 18:47:39 GMT


Nicholas Paldino commented on LUCENENET-230:

I believe I commented on this elsewhere, but this is a horrible idea.  You are trying to second
guess the JIT which in general is a very bad idea.

Also, there have been no mentions that this actually impacts performance.  You say "I believe"
that it will cause  performance issue.  

Because you have no substantial evidence that removing the method call will impact performance
in a noticeable way, it shouldn't be done, as you are increasing the complexity of the code
and reducing maintainability for a benefit that you don't know is there.

In other words, this is premature optimization.  It's an anti-best-practice and shouldn't
be applied here.

That being said, the patch for the fix for URShift should be applied, but it should not be
replaced in OpenBitSetIterator or OpenBitSet.  Also, an attempt to replace calls to URShift
in other areas of the code bring up the same issues and should NOT be made.

> TestOpenBitSet
> --------------
>                 Key: LUCENENET-230
>                 URL:
>             Project: Lucene.Net
>          Issue Type: Bug
>            Reporter: Michael Garski
>         Attachments: LUCENENET-230.patch
> I'm working through the tests in TestOpenBitSet.  I'll be merging in Doug's code from
2.4 to handle growing the BitArrays.

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

View raw message