[PATCH] D98516: [WIP] Introduce the "implements" string attribute

[Artem Belevich via Phabricator via llvm-commits]

tra added a comment.

Nice.

And, also, a bit scary as it can potentially be used to replace an arbitrary function, not just a math intrinsic normally mapped to libcall. 
I can see how it may be useful to allow function substitution for uses other than libm on GPUs, but it's also going to be way too easy to make a mess. E.g. if/when someone substitutes `@llvm.lifetime.start` or some other special purpose intrinsic llvm relies on.

Perhaps it would make sense to only allow whitelisted functions?



================
Comment at: llvm/test/Transforms/ImplementsAttrResolver/intrinsic_implementation.ll:6
+
+define internal float @my_sin(float %d) "implements"="llvm.sin.f32" {
+; CHECK-LABEL: @my_sin(
----------------
What happens if the function has different signature than that of the intrinsic?


================
Comment at: llvm/test/Transforms/ImplementsAttrResolver/intrinsic_implementation.ll:6
+
+define internal float @my_sin(float %d) "implements"="llvm.sin.f32" {
+; CHECK-LABEL: @my_sin(
----------------
tra wrote:
> What happens if the function has different signature than that of the intrinsic?
We may want to be able provide different implementations based on fast-math flags.
libdevice has few functions that have distinct fast-math implementations.



Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D98516/new/

https://reviews.llvm.org/D98516