[
https://issues.apache.org/jira/browse/LUCENE-1796?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12741402#action_12741402
]
Mark Miller commented on LUCENE-1796:
-------------------------------------
Sorry - to be a bit more clear:
Lucene trunk without your patch was like 90 seconds - twice as slow as a
previous version of Lucene on Yoniks short doc test.
The numbers reported above are for Lucene pre the TokenStream improvements, and
post with your patch - so now the numbers are much closer - though its still a
tad slower.
You second patch didnt change things for me from your first patch - same perf
in this test.
bq. If so, the problem was really this reflection calls.
Yes - which you can clearly see from the profiling results I got yesterday:
http://myhardshadow.com/images/before.png
http://myhardshadow.com/images/after.png
isMethodOverriden was called way too often for its speed - its a highly
inefficient call if you trace through it.
> Speed up repeated TokenStream init
> ----------------------------------
>
> Key: LUCENE-1796
> URL: https://issues.apache.org/jira/browse/LUCENE-1796
> Project: Lucene - Java
> Issue Type: Improvement
> Reporter: Mark Miller
> Assignee: Uwe Schindler
> Fix For: 2.9
>
> Attachments: LUCENE-1796.patch, LUCENE-1796.patch
>
>
> by caching isMethodOverridden results
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: java-dev-h...@lucene.apache.org
