[IntrinsicEmitter] Add overloaded type VecOfBitcastsToInt for SVE intrinsics

[kmclaughlin]

Summary

This allows intrinsics such as the following to be defined:

• declare <n x 4 x i32> @llvm.something.nxv4f32(<n x 4 x i32>, <n x 4 x i1>, <n x 4 x float>)

...where <n x 4 x i32> is derived from <n x 4 x float>, but
the element needs bitcasting to int.

[c-rhodes]

LGTM

Looks like a pretty straightforward change. Might be worth waiting a day or so before committing incase anyone else has any comments.

llvm/include/llvm/IR/Intrinsics.h

#undef GET_INTRINSIC_ENUM_VALUES
   /// intrinsic. This is returned by getIntrinsicInfoTableEntries.
   struct IITDescriptor {
     enum IITDescriptorKind {
       Void, VarArg, MMX, Token, Metadata, Half, Float, Double, Quad,
       Integer, Vector, Pointer, Struct,
       Argument, ExtendArgument, TruncArgument, HalfVecArgument,
       SameVecWidthArgument, PtrToArgument, PtrToElt, VecOfAnyPtrsToElt,
       VecElementArgument, ScalableVecArgument, Subdivide2Argument,
-      Subdivide4Argument
+      Subdivide4Argument, VecOfBitcastsToInt
     } Kind;
 
     union {
       unsigned Integer_Width;
       unsigned Float_Width;
       unsigned Vector_Width;
       unsigned Pointer_AddressSpace;
       unsigned Struct_NumElements;
enum ArgKind {
       AK_MatchType = 7
     };
 
     unsigned getArgumentNumber() const {
       assert(Kind == Argument || Kind == ExtendArgument ||
              Kind == TruncArgument || Kind == HalfVecArgument ||
              Kind == SameVecWidthArgument || Kind == PtrToArgument ||
              Kind == PtrToElt || Kind == VecElementArgument ||
-             Kind == Subdivide2Argument || Kind == Subdivide4Argument);
+             Kind == Subdivide2Argument || Kind == Subdivide4Argument ||
+             Kind == VecOfBitcastsToInt);
       return Argument_Info >> 3;
     }
     ArgKind getArgumentKind() const {
       assert(Kind == Argument || Kind == ExtendArgument ||
              Kind == TruncArgument || Kind == HalfVecArgument ||
              Kind == SameVecWidthArgument || Kind == PtrToArgument ||
              Kind == VecElementArgument || Kind == Subdivide2Argument ||
-             Kind == Subdivide4Argument);
+             Kind == Subdivide4Argument || Kind == VecOfBitcastsToInt);
       return (ArgKind)(Argument_Info & 7);
     }
 
     // VecOfAnyPtrsToElt uses both an overloaded argument (for address space)
     // and a reference argument (for matching vector width and element types)
     unsigned getOverloadArgNumber() const {
       assert(Kind == VecOfAnyPtrsToElt);
       return Argument_Info >> 16;

llvm/include/llvm/IR/Intrinsics.td

 class LLVMHalfElementsVectorType<int num> : LLVMMatchType<num>;
 
 // Match the type of another intrinsic parameter that is expected to be a
 // vector type (i.e. <N x iM>) but with each element subdivided to
 // form a vector with more elements that are smaller than the original.
 class LLVMSubdivide2VectorType<int num> : LLVMMatchType<num>;
 class LLVMSubdivide4VectorType<int num> : LLVMMatchType<num>;
 
+// Match the element count and bit width of another intrinsic parameter, but
+// change the element type to an integer.
+class LLVMVectorOfBitcastsToInt<int num> : LLVMMatchType<num>;
+
 def llvm_void_ty       : LLVMType<isVoid>;
 let isAny = 1 in {
   def llvm_any_ty        : LLVMType<Any>;
   def llvm_anyint_ty     : LLVMType<iAny>;
   def llvm_anyfloat_ty   : LLVMType<fAny>;
   def llvm_anyvector_ty  : LLVMType<vAny>;
 }
 def llvm_i1_ty         : LLVMType<i1>;

llvm/lib/IR/Function.cpp

enum IIT_Info {
   IIT_V1024 = 37,
   IIT_STRUCT6 = 38,
   IIT_STRUCT7 = 39,
   IIT_STRUCT8 = 40,
   IIT_F128 = 41,
   IIT_VEC_ELEMENT = 42,
   IIT_SCALABLE_VEC = 43,
   IIT_SUBDIVIDE2_ARG = 44,
-  IIT_SUBDIVIDE4_ARG = 45
+  IIT_SUBDIVIDE4_ARG = 45,
+  IIT_VEC_OF_BITCASTS_TO_INT = 46
 };
 
 static void DecodeIITType(unsigned &NextElt, ArrayRef<unsigned char> Infos,
                       SmallVectorImpl<Intrinsic::IITDescriptor> &OutputTable) {
   using namespace Intrinsic;
 
   IIT_Info Info = IIT_Info(Infos[NextElt++]);
   unsigned StructElts = 2;
case IIT_VEC_ELEMENT: {
     return;
   }
   case IIT_SCALABLE_VEC: {
     OutputTable.push_back(IITDescriptor::get(IITDescriptor::ScalableVecArgument,
                                              0));
     DecodeIITType(NextElt, Infos, OutputTable);
     return;
   }
+  case IIT_VEC_OF_BITCASTS_TO_INT: {
+    unsigned ArgInfo = (NextElt == Infos.size() ? 0 : Infos[NextElt++]);
+    OutputTable.push_back(IITDescriptor::get(IITDescriptor::VecOfBitcastsToInt,
+                                             ArgInfo));
+    return;
+  }
   }
   llvm_unreachable("unhandled");
 }
 
 #define GET_INTRINSIC_GENERATOR_GLOBAL
 #include "llvm/IR/IntrinsicImpl.inc"
 #undef GET_INTRINSIC_GENERATOR_GLOBAL
 
case IITDescriptor::PtrToElt: {
     return PointerType::getUnqual(EltTy);
   }
   case IITDescriptor::VecElementArgument: {
     Type *Ty = Tys[D.getArgumentNumber()];
     if (VectorType *VTy = dyn_cast<VectorType>(Ty))
       return VTy->getElementType();
     llvm_unreachable("Expected an argument of Vector Type");
   }
+  case IITDescriptor::VecOfBitcastsToInt: {
+    Type *Ty = Tys[D.getArgumentNumber()];
+    VectorType *VTy = dyn_cast<VectorType>(Ty);
+    assert(VTy && "Expected an argument of Vector Type");
+    return VectorType::getInteger(VTy);
+  }
   case IITDescriptor::VecOfAnyPtrsToElt:
     // Return the overloaded type (which determines the pointers address space)
     return Tys[D.getOverloadArgNumber()];
   case IITDescriptor::ScalableVecArgument: {
     Type *Ty = DecodeFixedType(Infos, Tys, Context);
     return VectorType::get(Ty->getVectorElementType(),
                            { Ty->getVectorNumElements(), true });
   }
switch (D.Kind) {
     }
     case IITDescriptor::ScalableVecArgument: {
       VectorType *VTy = dyn_cast<VectorType>(Ty);
       if (!VTy || !VTy->isScalable())
         return true;
       return matchIntrinsicType(VTy, Infos, ArgTys, DeferredChecks,
                                 IsDeferredCheck);
     }
+    case IITDescriptor::VecOfBitcastsToInt: {
+      if (D.getArgumentNumber() >= ArgTys.size())
+        return IsDeferredCheck || DeferCheck(Ty);
+      auto *ReferenceType = dyn_cast<VectorType>(ArgTys[D.getArgumentNumber()]);
+      auto *ThisArgVecTy = dyn_cast<VectorType>(Ty);
+      if (!ThisArgVecTy || !ReferenceType)
+        return true;
+      return ThisArgVecTy != VectorType::getInteger(ReferenceType);
+    }
   }
   llvm_unreachable("unhandled");
 }
 
 Intrinsic::MatchIntrinsicTypesResult
 Intrinsic::matchIntrinsicSignature(FunctionType *FTy,
                                    ArrayRef<Intrinsic::IITDescriptor> &Infos,
                                    SmallVectorImpl<Type *> &ArgTys) {

llvm/utils/TableGen/IntrinsicEmitter.cpp

enum IIT_Info {
   IIT_V1024 = 37,
   IIT_STRUCT6 = 38,
   IIT_STRUCT7 = 39,
   IIT_STRUCT8 = 40,
   IIT_F128 = 41,
   IIT_VEC_ELEMENT = 42,
   IIT_SCALABLE_VEC = 43,
   IIT_SUBDIVIDE2_ARG = 44,
-  IIT_SUBDIVIDE4_ARG = 45
+  IIT_SUBDIVIDE4_ARG = 45,
+  IIT_VEC_OF_BITCASTS_TO_INT = 46
 };
 
 static void EncodeFixedValueType(MVT::SimpleValueType VT,
                                  std::vector<unsigned char> &Sig) {
   if (MVT(VT).isInteger()) {
     unsigned BitWidth = MVT(VT).getSizeInBits();
     switch (BitWidth) {
     default: PrintFatalError("unhandled integer type width in intrinsic!");
if (R->isSubClassOf("LLVMMatchType")) {
     } else if (R->isSubClassOf("LLVMPointerToElt"))
       Sig.push_back(IIT_PTR_TO_ELT);
     else if (R->isSubClassOf("LLVMVectorElementType"))
       Sig.push_back(IIT_VEC_ELEMENT);
     else if (R->isSubClassOf("LLVMSubdivide2VectorType"))
       Sig.push_back(IIT_SUBDIVIDE2_ARG);
     else if (R->isSubClassOf("LLVMSubdivide4VectorType"))
       Sig.push_back(IIT_SUBDIVIDE4_ARG);
+    else if (R->isSubClassOf("LLVMVectorOfBitcastsToInt"))
+      Sig.push_back(IIT_VEC_OF_BITCASTS_TO_INT);
     else
       Sig.push_back(IIT_ARG);
     return Sig.push_back((Number << 3) | 7 /*IITDescriptor::AK_MatchType*/);
   }
 
   MVT::SimpleValueType VT = getValueType(R->getValueAsDef("VT"));
 
   unsigned Tmp = 0;