oertl commented on PR #20359:
URL: https://github.com/apache/kafka/pull/20359#issuecomment-3489667902

   @ijuma 
   > Is it worth doing a separate PR that brings back these optimizations? 
What's the performance difference of adding them back?
   
   The biggest performance gain comes from the main loop. The code could 
potentially be further optimized for small inputs. However, these (potential) 
additional performance gains are not so clear and might require some 
benchmarking on different platforms to get confident results.
   
   > Also, please specify the Java version used for your benchmark runs.
   
   The version of Java was OpenJDK 24. However, I no longer have all the 
details about it. The benchmark was part of this PR, though, so you can easily 
run it again.
   
   > There has been work at the JVM level to automatically perform 
optimizations like this one (an example is 
https://github.com/openjdk/jdk/pull/24023).
   
   Interesting, but it does not seem to have been merged yet.
   
   


-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to