paleolimbot commented on PR #280:
URL: https://github.com/apache/arrow-nanoarrow/pull/280#issuecomment-1750684511

   > does this program generate any difference for you? 
   
   With my (apple aarch64) `clang`, the compiler correctly deduces that it can 
do all the work at compile time and gives me a time of zero for both! (Except 
at `-O0`, where it gives me slightly better timing for the 'no shifts' option).
   
   Either way I'm happy to merge this (perhaps with a rebase to clear up the 
CI). In the unlikely event I discover any performance regression I'm happy to 
revisit.
   
   I did try briefly my macro approach before 0.3.0 to see if that could be a 
quick win, but it seemed to *decrease* performance in the R package for the 
"unpack 32" case in my first benchmark so I moved on to other things.


-- 
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