[PATCH] D67406: [TLI][AMDGPU] AMDPAL does not have tan function

[Tim Renouf via Phabricator via llvm-commits]

tpr updated this revision to Diff 219581.
tpr added a comment.

V2: Disable all library functions, not just tan.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D67406/new/

https://reviews.llvm.org/D67406

Files:
  lib/Analysis/TargetLibraryInfo.cpp
  test/Transforms/InstCombine/AMDGPU/tan.ll


Index: test/Transforms/InstCombine/AMDGPU/tan.ll
===================================================================
--- /dev/null
+++ test/Transforms/InstCombine/AMDGPU/tan.ll
@@ -0,0 +1,21 @@
+; RUN: opt -mtriple=amdgcn--amdpal -S -instcombine <%s | FileCheck --check-prefixes=GCN %s
+
+; Check that sin/cos is not folded to tan on amdgcn.
+
+; GCN-LABEL: define amdgpu_ps float @llpc.shader.FS.main
+; GCN: call float @llvm.sin.f32
+; GCN: call float @llvm.cos.f32
+
+declare float @llvm.sin.f32(float) #0
+declare float @llvm.cos.f32(float) #0
+
+define amdgpu_ps float @llpc.shader.FS.main(float %arg) {
+.entry:
+  %tmp32 = call float @llvm.sin.f32(float %arg)
+  %tmp33 = call float @llvm.cos.f32(float %arg)
+  %tmp34 = fdiv reassoc nnan nsz arcp contract float 1.000000e+00, %tmp33
+  %tmp35 = fmul reassoc nnan nsz arcp contract float %tmp32, %tmp34
+  ret float %tmp35
+}
+
+attributes #0 = { nounwind readnone speculatable willreturn }
Index: lib/Analysis/TargetLibraryInfo.cpp
===================================================================
--- lib/Analysis/TargetLibraryInfo.cpp
+++ lib/Analysis/TargetLibraryInfo.cpp
@@ -104,17 +104,8 @@
   TLI.setShouldSignExtI32Param(ShouldSignExtI32Param);
 
   if (T.getArch() == Triple::r600 ||
-      T.getArch() == Triple::amdgcn) {
-    TLI.setUnavailable(LibFunc_ldexp);
-    TLI.setUnavailable(LibFunc_ldexpf);
-    TLI.setUnavailable(LibFunc_ldexpl);
-    TLI.setUnavailable(LibFunc_exp10);
-    TLI.setUnavailable(LibFunc_exp10f);
-    TLI.setUnavailable(LibFunc_exp10l);
-    TLI.setUnavailable(LibFunc_log10);
-    TLI.setUnavailable(LibFunc_log10f);
-    TLI.setUnavailable(LibFunc_log10l);
-  }
+      T.getArch() == Triple::amdgcn)
+    TLI.disableAllFunctions();
 
   // There are no library implementations of memcpy and memset for AMD gpus and
   // these can be difficult to lower in the backend.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D67406.219581.patch
Type: text/x-patch
Size: 1871 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20190910/eef419c2/attachment.bin>