[PATCH] D120403: Lower Fortran intrinsic to a runtime call/llvm intrinsic

[Kiran Chandramohan via Phabricator via llvm-commits]

kiranchandramohan added a comment.

> @clementval wrote: Please use the Co-authored-by:  notation in the commit message otherwise the contributions is not correctly attributed.

Done

In D120403#3345844 <https://reviews.llvm.org/D120403#3345844>, @clementval wrote:

> In D120403#3345805 <https://reviews.llvm.org/D120403#3345805>, @kiranchandramohan wrote:
>
>> Removed code dealing with relaxed and precise pgmath intrinsics since this is not tested. 
>> Also removed flags which can switch between precise, relaxed, fast or llvm intrinsics.
>
> Would be better to remove the code before review.

Sorry about that. I will bring the removed code soon in another review. Submitting this now.



================
Comment at: flang/lib/Lower/IntrinsicCall.cpp:153-167
+enum MathRuntimeVersion {
+  fastVersion,
+  relaxedVersion,
+  preciseVersion,
+  llvmOnly
+};
+llvm::cl::opt<MathRuntimeVersion> mathRuntimeVersion(
----------------
kiranchandramohan wrote:
> Note: These flags are not tested in this patch since the `abs` intrinsic does not have fast/relaxed/precise versions for float, it only has an llvm or math runtime lowering for float. Can remove if necessary.
done


================
Comment at: flang/test/Lower/Intrinsics/abs.f90:1
+! RUN: bbc -emit-fir %s -o - | FileCheck %s
+
----------------
awarzynski wrote:
> Shall we test with `flang-new` as well? I think that that would be helpful to make sure that the two stay in sync.
done.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D120403/new/

https://reviews.llvm.org/D120403