[mlir][LLVMIR] Add lowering support for vp.muladd intrinsic

Reviewed By: gysit

Differential Revision: https://reviews.llvm.org/D143449

Author: dcaballe

mlir/include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td

@@ -825,7 +825,8 @@ def LLVM_VPFRemOp : LLVM_VPBinaryF<"frem">;
 def LLVM_VPFNegOp : LLVM_VPUnaryF<"fneg">;
 
 // Float Ternary
-def LLVM_VPFmaOp  : LLVM_VPTernaryF<"fma">;
+def LLVM_VPFMulAddOp  : LLVM_VPTernaryF<"fmuladd">;
+def LLVM_VPFmaOp      : LLVM_VPTernaryF<"fma">;
 
 // Integer Reduction
 def LLVM_VPReduceAddOp  : LLVM_VPReductionI<"add">;

mlir/test/Target/LLVMIR/Import/intrinsic.ll

@@ -629,6 +629,8 @@ define void @vector_predication_intrinsics(<8 x i32> %0, <8 x i32> %1, <8 x floa
   %58 = call <8 x i64> @llvm.vp.ptrtoint.v8i64.v8p0(<8 x ptr> %6, <8 x i1> %11, i32 %12)
   ; CHECK: "llvm.intr.vp.inttoptr"(%{{.*}}, %{{.*}}, %{{.*}}) : (vector<8xi64>, vector<8xi1>, i32) -> !llvm.vec<8 x ptr>
   %59 = call <8 x ptr> @llvm.vp.inttoptr.v8p0.v8i64(<8 x i64> %4, <8 x i1> %11, i32 %12)
+  ; CHECK: "llvm.intr.vp.fmuladd"(%{{.*}}, %{{.*}}, %{{.*}}, %{{.*}}, %{{.*}}) : (vector<8xf32>, vector<8xf32>, vector<8xf32>, vector<8xi1>, i32) -> vector<8xf32>
+  %60 = call <8 x float> @llvm.vp.fmuladd.v8f32(<8 x float> %2, <8 x float> %3, <8 x float> %3, <8 x i1> %11, i32 %12)
   ret void
 }
 
@@ -762,6 +764,7 @@ declare <8 x float> @llvm.vp.fdiv.v8f32(<8 x float>, <8 x float>, <8 x i1>, i32)
 declare <8 x float> @llvm.vp.frem.v8f32(<8 x float>, <8 x float>, <8 x i1>, i32)
 declare <8 x float> @llvm.vp.fneg.v8f32(<8 x float>, <8 x i1>, i32)
 declare <8 x float> @llvm.vp.fma.v8f32(<8 x float>, <8 x float>, <8 x float>, <8 x i1>, i32)
+declare <8 x float> @llvm.vp.fmuladd.v8f32(<8 x float>, <8 x float>, <8 x float>, <8 x i1>, i32)
 declare i32 @llvm.vp.reduce.add.v8i32(i32, <8 x i32>, <8 x i1>, i32)
 declare i32 @llvm.vp.reduce.mul.v8i32(i32, <8 x i32>, <8 x i1>, i32)
 declare i32 @llvm.vp.reduce.and.v8i32(i32, <8 x i32>, <8 x i1>, i32)

mlir/test/Target/LLVMIR/llvmir-intrinsics.mlir

@@ -593,6 +593,9 @@ llvm.func @vector_predication_intrinsics(%A: vector<8xi32>, %B: vector<8xi32>,
   // CHECK: call <8 x float> @llvm.vp.fma.v8f32
   "llvm.intr.vp.fma" (%C, %D, %D, %mask, %evl) :
          (vector<8xf32>, vector<8xf32>, vector<8xf32>, vector<8xi1>, i32) -> vector<8xf32>
+  // CHECK: call <8 x float> @llvm.vp.fmuladd.v8f32
+  "llvm.intr.vp.fmuladd" (%C, %D, %D, %mask, %evl) :
+         (vector<8xf32>, vector<8xf32>, vector<8xf32>, vector<8xi1>, i32) -> vector<8xf32>
 
   // CHECK: call i32 @llvm.vp.reduce.add.v8i32
   "llvm.intr.vp.reduce.add" (%i, %A, %mask, %evl) :