[X86] Combine FRINT + FP_TO_SINT to LRINT (llvm#126477)

Based on Craig's suggestion on llvm#126217

Alive2: https://alive2.llvm.org/ce/z/9XNpWt

Author: phoebewang

llvm/lib/Target/X86/X86ISelLowering.cpp

@@ -2682,6 +2682,7 @@ X86TargetLowering::X86TargetLowering(const X86TargetMachine &TM,
                        ISD::ZERO_EXTEND_VECTOR_INREG,
                        ISD::SINT_TO_FP,
                        ISD::UINT_TO_FP,
+                       ISD::FP_TO_SINT,
                        ISD::STRICT_SINT_TO_FP,
                        ISD::STRICT_UINT_TO_FP,
                        ISD::FP_TO_SINT_SAT,
@@ -56494,6 +56495,17 @@ static SDValue combineSIntToFP(SDNode *N, SelectionDAG &DAG,
   return SDValue();
 }
 
+static SDValue combineFPToSInt(SDNode *N, SelectionDAG &DAG,
+                               const X86Subtarget &Subtarget) {
+  EVT VT = N->getValueType(0);
+  SDValue Src = N->getOperand(0);
+  if (Subtarget.hasSSE2() && Src.getOpcode() == ISD::FRINT &&
+      VT.getScalarType() == MVT::i32 && Src.hasOneUse())
+    return DAG.getNode(ISD::LRINT, SDLoc(N), VT, Src.getOperand(0));
+
+  return SDValue();
+}
+
 // Custom handling for VCVTTPS2QQS/VCVTTPS2UQQS
 static SDValue combineFP_TO_xINT_SAT(SDNode *N, SelectionDAG &DAG,
                                      const X86Subtarget &Subtarget) {
@@ -59572,6 +59584,7 @@ SDValue X86TargetLowering::PerformDAGCombine(SDNode *N,
   case ISD::UINT_TO_FP:
   case ISD::STRICT_UINT_TO_FP:
     return combineUIntToFP(N, DAG, Subtarget);
+  case ISD::FP_TO_SINT:     return combineFPToSInt(N, DAG, Subtarget);
   case ISD::LRINT:
   case ISD::LLRINT:         return combineLRINT_LLRINT(N, DAG, Subtarget);
   case ISD::FADD:

llvm/test/CodeGen/X86/rint-conv.ll

@@ -0,0 +1,114 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc < %s -mtriple=i686-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=X86
+; RUN: llc < %s -mtriple=x86_64-unknown | FileCheck %s --check-prefixes=X64
+; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX
+
+define i32 @combine_f32(float %x) nounwind {
+; X86-LABEL: combine_f32:
+; X86:       # %bb.0: # %entry
+; X86-NEXT:    cvtss2si {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    retl
+;
+; X64-LABEL: combine_f32:
+; X64:       # %bb.0: # %entry
+; X64-NEXT:    cvtss2si %xmm0, %eax
+; X64-NEXT:    retq
+;
+; AVX-LABEL: combine_f32:
+; AVX:       # %bb.0: # %entry
+; AVX-NEXT:    vcvtss2si %xmm0, %eax
+; AVX-NEXT:    retq
+entry:
+  %0 = tail call float @llvm.rint.f32(float %x)
+  %1 = fptosi float %0 to i32
+  ret i32 %1
+}
+
+define i32 @combine_f64(double %x) nounwind {
+; X86-LABEL: combine_f64:
+; X86:       # %bb.0: # %entry
+; X86-NEXT:    cvtsd2si {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    retl
+;
+; X64-LABEL: combine_f64:
+; X64:       # %bb.0: # %entry
+; X64-NEXT:    cvtsd2si %xmm0, %eax
+; X64-NEXT:    retq
+;
+; AVX-LABEL: combine_f64:
+; AVX:       # %bb.0: # %entry
+; AVX-NEXT:    vcvtsd2si %xmm0, %eax
+; AVX-NEXT:    retq
+entry:
+  %0 = tail call double @llvm.rint.f32(double %x)
+  %1 = fptosi double %0 to i32
+  ret i32 %1
+}
+
+define <4 x i32> @combine_v4f32(<4 x float> %x) nounwind {
+; X86-LABEL: combine_v4f32:
+; X86:       # %bb.0: # %entry
+; X86-NEXT:    cvtps2dq %xmm0, %xmm0
+; X86-NEXT:    retl
+;
+; X64-LABEL: combine_v4f32:
+; X64:       # %bb.0: # %entry
+; X64-NEXT:    cvtps2dq %xmm0, %xmm0
+; X64-NEXT:    retq
+;
+; AVX-LABEL: combine_v4f32:
+; AVX:       # %bb.0: # %entry
+; AVX-NEXT:    vcvtps2dq %xmm0, %xmm0
+; AVX-NEXT:    retq
+entry:
+  %0 = tail call <4 x float> @llvm.rint.v4f32(<4 x float> %x)
+  %1 = fptosi <4 x float> %0 to <4 x i32>
+  ret <4 x i32> %1
+}
+
+define <2 x i32> @combine_v2f64(<2 x double> %x) nounwind {
+; X86-LABEL: combine_v2f64:
+; X86:       # %bb.0: # %entry
+; X86-NEXT:    cvtpd2dq %xmm0, %xmm0
+; X86-NEXT:    retl
+;
+; X64-LABEL: combine_v2f64:
+; X64:       # %bb.0: # %entry
+; X64-NEXT:    cvtpd2dq %xmm0, %xmm0
+; X64-NEXT:    retq
+;
+; AVX-LABEL: combine_v2f64:
+; AVX:       # %bb.0: # %entry
+; AVX-NEXT:    vcvtpd2dq %xmm0, %xmm0
+; AVX-NEXT:    retq
+entry:
+  %0 = tail call <2 x double> @llvm.rint.v2f64(<2 x double> %x)
+  %1 = fptosi <2 x double> %0 to <2 x i32>
+  ret <2 x i32> %1
+}
+
+define <4 x i32> @combine_v4f64(<4 x double> %x) nounwind {
+; X86-LABEL: combine_v4f64:
+; X86:       # %bb.0: # %entry
+; X86-NEXT:    cvtpd2dq %xmm1, %xmm1
+; X86-NEXT:    cvtpd2dq %xmm0, %xmm0
+; X86-NEXT:    movlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X86-NEXT:    retl
+;
+; X64-LABEL: combine_v4f64:
+; X64:       # %bb.0: # %entry
+; X64-NEXT:    cvtpd2dq %xmm1, %xmm1
+; X64-NEXT:    cvtpd2dq %xmm0, %xmm0
+; X64-NEXT:    movlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X64-NEXT:    retq
+;
+; AVX-LABEL: combine_v4f64:
+; AVX:       # %bb.0: # %entry
+; AVX-NEXT:    vcvtpd2dq %ymm0, %xmm0
+; AVX-NEXT:    vzeroupper
+; AVX-NEXT:    retq
+entry:
+  %0 = tail call <4 x double> @llvm.rint.v4f64(<4 x double> %x)
+  %1 = fptosi <4 x double> %0 to <4 x i32>
+  ret <4 x i32> %1
+}