lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Michael McCandless (JIRA)" <>
Subject [jira] Commented: (LUCENE-1410) PFOR implementation
Date Fri, 03 Oct 2008 19:19:44 GMT


Michael McCandless commented on LUCENE-1410:

Q: We should look at the asm that's produced?
A: Maybe. But would that be java bytecodes or x86 or powerpc? I'd prefer to wait with that.
There are some nice machine instrns to this on the various architectures, but that would require
to use native code.
I was thinking local CPU's native asm, so that we could at least see if the benefits of pfor
(no hard-for-cpu-to-predict if statement inside inner loop) "survive" through the Java stack.
 I'll try to look.

Q: But can I also get the "bytes consumed" when I ask decompress() to run?
A; Yes, but only after decompression or after compression.
The number of exceptions is not explicitly coded in the header, so the size to encode the
exceptions is not known beforehand. That could be changed, but than the header will get bigger.

OK so it is self-punctuating, so, we wouldn't need to encode block size in bytes into the

> PFOR implementation
> -------------------
>                 Key: LUCENE-1410
>                 URL:
>             Project: Lucene - Java
>          Issue Type: New Feature
>          Components: Other
>            Reporter: Paul Elschot
>            Priority: Minor
>         Attachments: LUCENE-1410b.patch,,
>   Original Estimate: 21840h
>  Remaining Estimate: 21840h
> Implementation of Patched Frame of Reference.

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:
For additional commands, e-mail:

View raw message