[PATCH] D72439: [mlir] Added Abs, Ceil and Cos to GPU lowering conversion.

[Alexander Belyaev via Phabricator via llvm-commits]

pifon2a requested changes to this revision.
pifon2a added inline comments.
This revision now requires changes to proceed.


================
Comment at: mlir/test/Conversion/GPUToROCDL/gpu-to-rocdl.mlir:48
+    // CHECK: llvm.call @_ocml_fabs_f32(%{{.*}}) : (!llvm.float) -> !llvm.float
+    %result_f32 = std.absf %fabs_f32 : f32
+    // CHECK: llvm.call @_ocml_fabs_f32(%{{.*}}) : (!llvm.float) -> !llvm.float
----------------
ftynse wrote:
> What's the point of exercising the same operation twice?
I think this is copied from `expf` test where I was checking that the symbols are not added twice to the symbol table. I think here it is enough to exercise the op only once.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D72439/new/

https://reviews.llvm.org/D72439