eejbyfeldt commented on code in PR #847: URL: https://github.com/apache/datafusion-comet/pull/847#discussion_r1722788439
########## kube/Dockerfile: ########## @@ -32,14 +31,38 @@ ENV RUSTFLAGS="-C debuginfo=line-tables-only -C incremental=false" ENV SPARK_VERSION=3.4 ENV SCALA_VERSION=2.12 +# copy source files to Docker image +RUN mkdir /comet +WORKDIR /comet + +# build native code first so that this layer can be re-used +# if only Scala code gets modified +COPY rust-toolchain.toml /comet/rust-toolchain.toml +COPY native /comet/native +RUN cd native && RUSTFLAGS="-Ctarget-cpu=native" cargo build --release Review Comment: Is the `-Ctarget-cpu=native` a good idea? My guess would have been that github runners might use a range of different generations of cpus and therefore have different features? If that is the case different runs here might for example enable different simd instructions and cause varying performance. Would it not be better to target some specific set of features and document that? And user what something else they should build from source. -- 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: github-unsubscr...@datafusion.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org For additional commands, e-mail: github-h...@datafusion.apache.org
