[GlobalISel] Translate insertelement and extractelement

Reviewers: qcolombet, aditya_nandakumar, dsanders, ab, t.p.northover, javed.absar

Reviewed By: qcolombet

Subscribers: dberris, rovka, llvm-commits, kristof.beyls

Differential Revision: https://reviews.llvm.org/D30761

llvm-svn: 297495

llvm/include/llvm/CodeGen/GlobalISel/IRTranslator.h

@@ -304,6 +304,10 @@ private:
 
   bool translateVAArg(const User &U, MachineIRBuilder &MIRBuilder);
 
+  bool translateInsertElement(const User &U, MachineIRBuilder &MIRBuilder);
+
+  bool translateExtractElement(const User &U, MachineIRBuilder &MIRBuilder);
+
   // Stubs to keep the compiler happy while we implement the rest of the
   // translation.
   bool translateResume(const User &U, MachineIRBuilder &MIRBuilder) {
@@ -342,12 +346,6 @@ private:
   bool translateUserOp2(const User &U, MachineIRBuilder &MIRBuilder) {
     return false;
   }
-  bool translateExtractElement(const User &U, MachineIRBuilder &MIRBuilder) {
-    return false;
-  }
-  bool translateInsertElement(const User &U, MachineIRBuilder &MIRBuilder) {
-    return false;
-  }
   bool translateShuffleVector(const User &U, MachineIRBuilder &MIRBuilder) {
     return false;
   }

llvm/include/llvm/CodeGen/GlobalISel/MachineIRBuilder.h

@@ -594,6 +594,30 @@ public:
   /// \return a MachineInstrBuilder for the newly created instruction.
   MachineInstrBuilder buildSelect(unsigned Res, unsigned Tst,
                                   unsigned Op0, unsigned Op1);
+
+  /// Build and insert \p Res<def> = G_INSERT_VECTOR_ELT \p Val,
+  /// \p Elt, \p Idx
+  ///
+  /// \pre setBasicBlock or setMI must have been called.
+  /// \pre \p Res and \p Val must be a generic virtual register
+  //       with the same vector type.
+  /// \pre \p Elt and \p Idx must be a generic virtual register
+  ///      with scalar type.
+  ///
+  /// \return The newly created instruction.
+  MachineInstrBuilder buildInsertVectorElement(unsigned Res, unsigned Val,
+                                               unsigned Elt, unsigned Idx);
+
+  /// Build and insert \p Res<def> = G_EXTRACT_VECTOR_ELT \p Val, \p Idx
+  ///
+  /// \pre setBasicBlock or setMI must have been called.
+  /// \pre \p Res must be a generic virtual register with scalar type.
+  /// \pre \p Val must be a generic virtual register with vector type.
+  /// \pre \p Idx must be a generic virtual register with scalar type.
+  ///
+  /// \return The newly created instruction.
+  MachineInstrBuilder buildExtractVectorElement(unsigned Res, unsigned Val,
+                                                unsigned Idx);
 };
 
 } // End namespace llvm.

llvm/include/llvm/Target/GenericOpcodes.td

@@ -519,4 +519,22 @@ def G_BRINDIRECT : Instruction {
   let isTerminator = 1;
 }
 
+//------------------------------------------------------------------------------
+// Vector ops
+//------------------------------------------------------------------------------
+
+// Generic insertelement.
+def G_INSERT_VECTOR_ELT : Instruction {
+  let OutOperandList = (outs type0:$dst);
+  let InOperandList = (ins type0:$src, type1:$elt, type2:$idx);
+  let hasSideEffects = 0;
+}
+
+// Generic extractelement.
+def G_EXTRACT_VECTOR_ELT : Instruction {
+  let OutOperandList = (outs type0:$dst);
+  let InOperandList = (ins type1:$src, type2:$idx);
+  let hasSideEffects = 0;
+}
+
 // TODO: Add the other generic opcodes.

llvm/include/llvm/Target/TargetOpcodes.def

@@ -395,6 +395,12 @@ HANDLE_TARGET_OPCODE(G_PTR_MASK)
 /// Generic BRANCH instruction. This is an unconditional branch.
 HANDLE_TARGET_OPCODE(G_BR)
 
+/// Generic insertelement.
+HANDLE_TARGET_OPCODE(G_INSERT_VECTOR_ELT)
+
+/// Generic extractelement.
+HANDLE_TARGET_OPCODE(G_EXTRACT_VECTOR_ELT)
+
 // TODO: Add more generic opcodes as we move along.
 
 /// Marker for the end of the generic opcode.

llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp

@@ -976,6 +976,36 @@ bool IRTranslator::translateVAArg(const User &U, MachineIRBuilder &MIRBuilder) {
   return true;
 }
 
+bool IRTranslator::translateInsertElement(const User &U,
+                                          MachineIRBuilder &MIRBuilder) {
+  // If it is a <1 x Ty> vector, use the scalar as it is
+  // not a legal vector type in LLT.
+  if (U.getType()->getVectorNumElements() == 1) {
+    unsigned Elt = getOrCreateVReg(*U.getOperand(1));
+    ValToVReg[&U] = Elt;
+    return true;
+  }
+  MIRBuilder.buildInsertVectorElement(
+      getOrCreateVReg(U), getOrCreateVReg(*U.getOperand(0)),
+      getOrCreateVReg(*U.getOperand(1)), getOrCreateVReg(*U.getOperand(2)));
+  return true;
+}
+
+bool IRTranslator::translateExtractElement(const User &U,
+                                           MachineIRBuilder &MIRBuilder) {
+  // If it is a <1 x Ty> vector, use the scalar as it is
+  // not a legal vector type in LLT.
+  if (U.getOperand(0)->getType()->getVectorNumElements() == 1) {
+    unsigned Elt = getOrCreateVReg(*U.getOperand(0));
+    ValToVReg[&U] = Elt;
+    return true;
+  }
+  MIRBuilder.buildExtractVectorElement(getOrCreateVReg(U),
+                                       getOrCreateVReg(*U.getOperand(0)),
+                                       getOrCreateVReg(*U.getOperand(1)));
+  return true;
+}
+
 bool IRTranslator::translatePHI(const User &U, MachineIRBuilder &MIRBuilder) {
   const PHINode &PI = cast<PHINode>(U);
   auto MIB = MIRBuilder.buildInstr(TargetOpcode::PHI);

llvm/lib/CodeGen/GlobalISel/MachineIRBuilder.cpp

@@ -582,6 +582,46 @@ MachineInstrBuilder MachineIRBuilder::buildSelect(unsigned Res, unsigned Tst,
       .addUse(Op1);
 }
 
+MachineInstrBuilder MachineIRBuilder::buildInsertVectorElement(unsigned Res,
+                                                               unsigned Val,
+                                                               unsigned Elt,
+                                                               unsigned Idx) {
+#ifndef NDEBUG
+  LLT ResTy = MRI->getType(Res);
+  LLT ValTy = MRI->getType(Val);
+  LLT EltTy = MRI->getType(Elt);
+  LLT IdxTy = MRI->getType(Idx);
+  assert(ResTy.isVector() && ValTy.isVector() && "invalid operand type");
+  assert(EltTy.isScalar() && IdxTy.isScalar() && "invalid operand type");
+  assert(ResTy.getNumElements() == ValTy.getNumElements() && "type mismatch");
+  assert(ResTy.getElementType() == EltTy && "type mismatch");
+#endif
+
+  return buildInstr(TargetOpcode::G_INSERT_VECTOR_ELT)
+      .addDef(Res)
+      .addUse(Val)
+      .addUse(Elt)
+      .addUse(Idx);
+}
+
+MachineInstrBuilder MachineIRBuilder::buildExtractVectorElement(unsigned Res,
+                                                                unsigned Val,
+                                                                unsigned Idx) {
+#ifndef NDEBUG
+  LLT ResTy = MRI->getType(Res);
+  LLT ValTy = MRI->getType(Val);
+  LLT IdxTy = MRI->getType(Idx);
+  assert(ValTy.isVector() && "invalid operand type");
+  assert(ResTy.isScalar() && IdxTy.isScalar() && "invalid operand type");
+  assert(ValTy.getElementType() == ResTy && "type mismatch");
+#endif
+
+  return buildInstr(TargetOpcode::G_EXTRACT_VECTOR_ELT)
+      .addDef(Res)
+      .addUse(Val)
+      .addUse(Idx);
+}
+
 void MachineIRBuilder::validateTruncExt(unsigned Dst, unsigned Src,
                                         bool IsExtend) {
 #ifndef NDEBUG

llvm/test/CodeGen/AArch64/GlobalISel/arm64-irtranslator.ll

@@ -1271,3 +1271,35 @@ define void @test_trivial_inlineasm() {
   call void asm "wibble", ""()
   ret void
 }
+
+define <2 x i32> @test_insertelement(<2 x i32> %vec, i32 %elt, i32 %idx){
+; CHECK-LABEL: name: test_insertelement
+; CHECK: [[VEC:%[0-9]+]](<2 x s32>) = COPY %d0
+; CHECK: [[ELT:%[0-9]+]](s32) = COPY %w0
+; CHECK: [[IDX:%[0-9]+]](s32) = COPY %w1
+; CHECK: [[RES:%[0-9]+]](<2 x s32>) = G_INSERT_VECTOR_ELT [[VEC]], [[ELT]](s32), [[IDX]](s32)
+; CHECK: %d0 = COPY [[RES]](<2 x s32>)
+  %res = insertelement <2 x i32> %vec, i32 %elt, i32 %idx
+  ret <2 x i32> %res
+}
+
+define i32 @test_extractelement(<2 x i32> %vec, i32 %idx) {
+; CHECK-LABEL: name: test_extractelement
+; CHECK: [[VEC:%[0-9]+]](<2 x s32>) = COPY %d0
+; CHECK: [[IDX:%[0-9]+]](s32) = COPY %w0
+; CHECK: [[RES:%[0-9]+]](s32) = G_EXTRACT_VECTOR_ELT [[VEC]](<2 x s32>), [[IDX]](s32)
+; CHECK: %w0 = COPY [[RES]](s32)
+  %res = extractelement <2 x i32> %vec, i32 %idx
+  ret i32 %res
+}
+
+define i32 @test_singleelementvector(i32 %elt){
+; CHECK-LABEL: name: test_singleelementvector
+; CHECK: [[ELT:%[0-9]+]](s32) = COPY %w0
+; CHECK-NOT: G_INSERT_VECTOR_ELT
+; CHECK-NOT: G_EXTRACT_VECTOR_ELT
+; CHECK: %w0 = COPY [[ELT]](s32)
+  %vec = insertelement <1 x i32> undef, i32 %elt, i32 0
+  %res = extractelement <1 x i32> %vec, i32 0
+  ret i32 %res
+}