banach-space wrote: Given the growing number of OpenMP and/or "offloading" flags, I agree with @AnastasiaStulova that it would be good to clarify the overall goal/design. That's not clear to me.
Is there are reference implementation that Flang is meant to follow? For example Clang or GFortran? Could we re-use more of Clang logic here? There's also a question how various flags impact one another and that seems also confusing (see e.g. this [question](https://flang-compiler.slack.com/archives/C01PY03PP9P/p1721378390015629) by @DominikAdamski ). Now, this is not really my area of expertise and I tend to stay hands-off. I rely on @jhuber6's [expertise in the area](https://llvm.org/devmtg/2022-04-03/slides/Improving.the.OpenMP.Offloading.Driver.LTO.libraries.and.toolchains.pdf) :) But it does sound like it would be good to discuss the long term goals here. That's not a blocker for me for this particular change. https://github.com/llvm/llvm-project/pull/100152 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
