dweiss commented on issue #1426: Do a bit count on 8 bytes from a long directly instead of reading 8 b… URL: https://github.com/apache/lucene-solr/pull/1426#issuecomment-612864603 I can't say for sure. Like I said - reasoning about hotspot optimizations is difficult and not reproducible across different environments. You could try to dump assembly code and check out the difference there. My experience tells me that even a different generated code shape could be the reason... I wanted to get this patch in on Java side mostly because if the reader does optimize getLong somehow then it automatically propagates to any calling code whereas a sequence of readByte() doesn't. These things accumulate over time and it's hard to keep track of them later on.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
