dmatth1 commented on PR #10011: URL: https://github.com/apache/arrow-rs/pull/10011#issuecomment-4607598032
Interesting. Looks like for `Neoverse-V2` wins are more minimal (hit win is ~1.3–1.6×, miss is parity or perhaps even a bit behind). I think on Neoverse-V2 scalar short-circuit is similar in speed to the vector kernel whereas for AVX2 and M1 Neon it's not (hence the wins on miss for those platforms too). Should we extend the benchmark bot to run on x86 so we have those benchmarks auto-populated here? -- 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]
