From d709efb0da64fa286362cedabccdb316a2461d24 Mon Sep 17 00:00:00 2001
From: Konstantin Zhuravlyov <kzhuravl_dev@outlook.com>
Date: Thu, 17 Nov 2016 04:28:37 +0000
Subject: [PATCH] [AMDGPU] Custom lower f16 = fp_round f64

llvm-svn: 287203
---
 llvm/lib/Target/AMDGPU/SIISelLowering.cpp | 20 ++++++++++++++++++++
 llvm/lib/Target/AMDGPU/SIISelLowering.h   |  3 +++
 2 files changed, 23 insertions(+)

diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index b72415b..79953b7 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -285,6 +285,7 @@ SITargetLowering::SITargetLowering(const TargetMachine &TM,
     AddPromotedToType(ISD::STORE, MVT::f16, MVT::i16);
 
     // F16 - VOP1 Actions.
+    setOperationAction(ISD::FP_ROUND, MVT::f16, Custom);
     setOperationAction(ISD::FCOS, MVT::f16, Promote);
     setOperationAction(ISD::FSIN, MVT::f16, Promote);
     setOperationAction(ISD::FP_TO_SINT, MVT::f16, Promote);
@@ -1832,6 +1833,8 @@ SDValue SITargetLowering::LowerOperation(SDValue Op, SelectionDAG &DAG) const {
 
   case ISD::ConstantFP:
     return lowerConstantFP(Op, DAG);
+  case ISD::FP_ROUND:
+    return lowerFP_ROUND(Op, DAG);
   }
   return SDValue();
 }
@@ -2043,6 +2046,23 @@ SDValue SITargetLowering::lowerConstantFP(SDValue Op, SelectionDAG &DAG) const {
   return SDValue();
 }
 
+SDValue SITargetLowering::lowerFP_ROUND(SDValue Op, SelectionDAG &DAG) const {
+  EVT DstVT = Op.getValueType();
+  EVT SrcVT = Op.getOperand(0).getValueType();
+
+  assert(DstVT == MVT::f16 &&
+         "Do not know how to custom lower FP_ROUND for non-f16 type");
+
+  if (SrcVT != MVT::f64)
+    return Op;
+
+  SDLoc DL(Op);
+  SDValue Src = Op.getOperand(0);
+  SDValue FpToFp16 = DAG.getNode(ISD::FP_TO_FP16, DL, MVT::i32, Src);
+  SDValue Trunc = DAG.getNode(ISD::TRUNCATE, DL, MVT::i16, FpToFp16);
+  return DAG.getNode(ISD::BITCAST, DL, MVT::f16, Trunc);;
+}
+
 SDValue SITargetLowering::getSegmentAperture(unsigned AS,
                                              SelectionDAG &DAG) const {
   SDLoc SL;
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.h b/llvm/lib/Target/AMDGPU/SIISelLowering.h
index 2cbfe11..2e14f49 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.h
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.h
@@ -56,6 +56,9 @@ class SITargetLowering final : public AMDGPUTargetLowering {
   /// \brief Custom lowering for ISD::ConstantFP.
   SDValue lowerConstantFP(SDValue Op, SelectionDAG &DAG) const;
 
+  /// \brief Custom lowering for ISD::FP_ROUND for MVT::f16.
+  SDValue lowerFP_ROUND(SDValue Op, SelectionDAG &DAG) const;
+
   SDValue getSegmentAperture(unsigned AS, SelectionDAG &DAG) const;
   SDValue lowerADDRSPACECAST(SDValue Op, SelectionDAG &DAG) const;
   SDValue lowerTRAP(SDValue Op, SelectionDAG &DAG) const;
-- 
2.7.4