CH-Abhinav commented on PR #16691:
URL: https://github.com/apache/lucene/pull/16691#issuecomment-6921289267

   So should I try to benchmark the PR again or we can close this? I mean it 
isn't that big of a deal this will give real but small benefits. Scalar code is 
optimal for C2 but when compared with SIMD we can always have SIMD perform 
better than scalar code. Whatever you say we will proceed with that. The actual 
PR was with multiple SIMD vectors which gave a lot of performance and I am more 
inclined to that over single SIMD vector but Robert said to make it single 
vector so yeah.
   
   That multiple vector version will definitely beat scalar with C2(excluding 
GraalVM which idk why) . What do you think ? 


-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to