[llvm] r338243 - AMDGPU: Make fneg combine handle fcanonicalize

[Matt Arsenault via llvm-commits]

Author: arsenm
Date: Mon Jul 30 05:16:47 2018
New Revision: 338243

URL: http://llvm.org/viewvc/llvm-project?rev=338243&view=rev
Log:
AMDGPU: Make fneg combine handle fcanonicalize

Modified:
    llvm/trunk/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
    llvm/trunk/test/CodeGen/AMDGPU/fneg-combines.ll

Modified: llvm/trunk/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/AMDGPUISelLowering.cpp?rev=338243&r1=338242&r2=338243&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/AMDGPUISelLowering.cpp (original)
+++ llvm/trunk/lib/Target/AMDGPU/AMDGPUISelLowering.cpp Mon Jul 30 05:16:47 2018
@@ -554,6 +554,7 @@ static bool fnegFoldsIntoOp(unsigned Opc
   case ISD::FTRUNC:
   case ISD::FRINT:
   case ISD::FNEARBYINT:
+  case ISD::FCANONICALIZE:
   case AMDGPUISD::RCP:
   case AMDGPUISD::RCP_LEGACY:
   case AMDGPUISD::RCP_IFLAG:
@@ -3595,6 +3596,7 @@ SDValue AMDGPUTargetLowering::performFNe
   case ISD::FRINT:
   case ISD::FNEARBYINT: // XXX - Should fround be handled?
   case ISD::FSIN:
+  case ISD::FCANONICALIZE:
   case AMDGPUISD::RCP:
   case AMDGPUISD::RCP_LEGACY:
   case AMDGPUISD::RCP_IFLAG:

Modified: llvm/trunk/test/CodeGen/AMDGPU/fneg-combines.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/fneg-combines.ll?rev=338243&r1=338242&r2=338243&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/AMDGPU/fneg-combines.ll (original)
+++ llvm/trunk/test/CodeGen/AMDGPU/fneg-combines.ll Mon Jul 30 05:16:47 2018
@@ -1725,6 +1725,26 @@ define amdgpu_kernel void @v_fneg_nearby
 }
 
 ; --------------------------------------------------------------------------------
+; fcanonicalize tests
+; --------------------------------------------------------------------------------
+
+; GCN-LABEL: {{^}}v_fneg_canonicalize_f32:
+; GCN: {{buffer|flat}}_load_dword [[A:v[0-9]+]]
+; GCN: v_mul_f32_e64 [[RESULT:v[0-9]+]], 1.0, -[[A]]
+; GCN: buffer_store_dword [[RESULT]]
+define amdgpu_kernel void @v_fneg_canonicalize_f32(float addrspace(1)* %out, float addrspace(1)* %a.ptr) #0 {
+  %tid = call i32 @llvm.amdgcn.workitem.id.x()
+  %tid.ext = sext i32 %tid to i64
+  %a.gep = getelementptr inbounds float, float addrspace(1)* %a.ptr, i64 %tid.ext
+  %out.gep = getelementptr inbounds float, float addrspace(1)* %out, i64 %tid.ext
+  %a = load volatile float, float addrspace(1)* %a.gep
+  %trunc = call float @llvm.canonicalize.f32(float %a)
+  %fneg = fsub float -0.0, %trunc
+  store float %fneg, float addrspace(1)* %out.gep
+  ret void
+}
+
+; --------------------------------------------------------------------------------
 ; vintrp tests
 ; --------------------------------------------------------------------------------
 
@@ -2117,6 +2137,7 @@ declare float @llvm.trunc.f32(float) #1
 declare float @llvm.round.f32(float) #1
 declare float @llvm.rint.f32(float) #1
 declare float @llvm.nearbyint.f32(float) #1
+declare float @llvm.canonicalize.f32(float) #1
 declare float @llvm.minnum.f32(float, float) #1
 declare float @llvm.maxnum.f32(float, float) #1