[llvm] [NVPTX] Support llvm.{exp2, log2} for f32 and vector of f32 (PR #120519)

[Princeton Ferro via llvm-commits]

================
@@ -0,0 +1,61 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc < %s -march=nvptx64 -mcpu=sm_20 -mattr=+ptx32 -nvptx-approx-exp2f32 | FileCheck --check-prefixes=CHECK %s
+; RUN: %if ptxas-12.0 %{ llc < %s -march=nvptx64 -mcpu=sm_20 -mattr=+ptx32 -nvptx-approx-exp2f32 | %ptxas-verify -arch=sm_20 %}
+target triple = "nvptx64-nvidia-cuda"
+
+; CHECK-LABEL: exp2_test
+define float @exp2_test(float %in) {
+; CHECK-LABEL: exp2_test(
+; CHECK:       {
+; CHECK-NEXT:    .reg .f32 %f<3>;
+; CHECK-EMPTY:
+; CHECK-NEXT:  // %bb.0: // %entry
+; CHECK-NEXT:    ld.param.f32 %f1, [exp2_test_param_0];
+; CHECK-NEXT:    ex2.approx.f32 %f2, %f1;
----------------
Prince781 wrote:

Yeah, I can't think of a good way to have a test for that.

https://github.com/llvm/llvm-project/pull/120519