[X86][SATCVT] Reduce MIN/MAXSS/D by conversion instruction result

CVTSD2SI returns INT_MIN/LONG_MIN when underflow happens.
VCVTPS/D2UD/QQ returns INT_MAX/LONG_MAX when overflow happens.

We can reduce one MIN/MAXSS/D instruction leveraging the result of the conversion instruction.

Partially fixes #136342

Author: phoebewang

llvm/lib/Target/X86/X86ISelLowering.cpp

@@ -21851,6 +21851,15 @@ X86TargetLowering::LowerFP_TO_INT_SAT(SDValue Op, SelectionDAG &DAG) const {
   assert(SatWidth <= DstWidth && SatWidth <= TmpWidth &&
          "Expected saturation width smaller than result width");
 
+  // AVX512 provides VCVTPS/D2UD/QQ which return INT_MAX/LONG_MAX when overflow
+  // happens. X86ISD::FMAX makes sure negative value and NaN return 0.
+  if (Subtarget.hasAVX512() && !IsSigned && SatWidth == DstWidth &&
+      (DstVT == MVT::i32 || (Subtarget.is64Bit() && DstVT == MVT::i64))) {
+    SDValue MinFloatNode = DAG.getConstantFP(0.0, dl, SrcVT);
+    SDValue Clamped = DAG.getNode(X86ISD::FMAX, dl, SrcVT, Src, MinFloatNode);
+    return DAG.getNode(ISD::FP_TO_UINT, dl, DstVT, Clamped);
+  }
+
   // Promote result of FP_TO_*INT to at least 32 bits.
   if (TmpWidth < 32) {
     TmpVT = MVT::i32;
@@ -21912,9 +21921,10 @@ X86TargetLowering::LowerFP_TO_INT_SAT(SDValue Op, SelectionDAG &DAG) const {
       return DAG.getNode(ISD::TRUNCATE, dl, DstVT, FpToInt);
     }
 
+    SDValue MinClamped = Src;
     // Clamp by MinFloat from below. If Src is NaN, the result is MinFloat.
-    SDValue MinClamped = DAG.getNode(
-      X86ISD::FMAX, dl, SrcVT, Src, MinFloatNode);
+    if (!IsSigned || SatWidth != DstWidth)
+      MinClamped = DAG.getNode(X86ISD::FMAX, dl, SrcVT, Src, MinFloatNode);
     // Clamp by MaxFloat from above. NaN cannot occur.
     SDValue BothClamped = DAG.getNode(
       X86ISD::FMINC, dl, SrcVT, MinClamped, MaxFloatNode);

llvm/test/CodeGen/X86/fpclamptosat.ll

@@ -6,11 +6,11 @@
 define i32 @stest_f64i32(double %x) nounwind {
 ; CHECK-LABEL: stest_f64i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    movsd {{.*#+}} xmm1 = [2.147483647E+9,0.0E+0]
+; CHECK-NEXT:    minsd %xmm0, %xmm1
+; CHECK-NEXT:    cvttsd2si %xmm1, %ecx
 ; CHECK-NEXT:    xorl %eax, %eax
 ; CHECK-NEXT:    ucomisd %xmm0, %xmm0
-; CHECK-NEXT:    maxsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    minsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    cvttsd2si %xmm0, %ecx
 ; CHECK-NEXT:    cmovnpl %ecx, %eax
 ; CHECK-NEXT:    retq
 entry:
@@ -619,11 +619,11 @@ entry:
 define i32 @stest_f64i32_mm(double %x) nounwind {
 ; CHECK-LABEL: stest_f64i32_mm:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    movsd {{.*#+}} xmm1 = [2.147483647E+9,0.0E+0]
+; CHECK-NEXT:    minsd %xmm0, %xmm1
+; CHECK-NEXT:    cvttsd2si %xmm1, %ecx
 ; CHECK-NEXT:    xorl %eax, %eax
 ; CHECK-NEXT:    ucomisd %xmm0, %xmm0
-; CHECK-NEXT:    maxsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    minsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    cvttsd2si %xmm0, %ecx
 ; CHECK-NEXT:    cmovnpl %ecx, %eax
 ; CHECK-NEXT:    retq
 entry:

llvm/test/CodeGen/X86/fptosi-sat-scalar.ll

@@ -1453,21 +1453,21 @@ define i32 @test_signed_i32_f64(double %f) nounwind {
 ; X86-SSE-LABEL: test_signed_i32_f64:
 ; X86-SSE:       # %bb.0:
 ; X86-SSE-NEXT:    movsd {{.*#+}} xmm0 = mem[0],zero
+; X86-SSE-NEXT:    movsd {{.*#+}} xmm1 = [2.147483647E+9,0.0E+0]
+; X86-SSE-NEXT:    minsd %xmm0, %xmm1
+; X86-SSE-NEXT:    cvttsd2si %xmm1, %ecx
 ; X86-SSE-NEXT:    xorl %eax, %eax
 ; X86-SSE-NEXT:    ucomisd %xmm0, %xmm0
-; X86-SSE-NEXT:    maxsd {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0
-; X86-SSE-NEXT:    minsd {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0
-; X86-SSE-NEXT:    cvttsd2si %xmm0, %ecx
 ; X86-SSE-NEXT:    cmovnpl %ecx, %eax
 ; X86-SSE-NEXT:    retl
 ;
 ; X64-LABEL: test_signed_i32_f64:
 ; X64:       # %bb.0:
+; X64-NEXT:    movsd {{.*#+}} xmm1 = [2.147483647E+9,0.0E+0]
+; X64-NEXT:    minsd %xmm0, %xmm1
+; X64-NEXT:    cvttsd2si %xmm1, %ecx
 ; X64-NEXT:    xorl %eax, %eax
 ; X64-NEXT:    ucomisd %xmm0, %xmm0
-; X64-NEXT:    maxsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; X64-NEXT:    minsd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; X64-NEXT:    cvttsd2si %xmm0, %ecx
 ; X64-NEXT:    cmovnpl %ecx, %eax
 ; X64-NEXT:    retq
     %x = call i32 @llvm.fptosi.sat.i32.f64(double %f)

llvm/test/CodeGen/X86/fptosi-sat-vector-128.ll

@@ -418,21 +418,18 @@ define <2 x i16> @test_signed_v2i16_v2f64(<2 x double> %f) nounwind {
 define <2 x i32> @test_signed_v2i32_v2f64(<2 x double> %f) nounwind {
 ; CHECK-LABEL: test_signed_v2i32_v2f64:
 ; CHECK:       # %bb.0:
-; CHECK-NEXT:    movsd {{.*#+}} xmm2 = [-2.147483648E+9,0.0E+0]
+; CHECK-NEXT:    movsd {{.*#+}} xmm2 = [2.147483647E+9,0.0E+0]
 ; CHECK-NEXT:    movapd %xmm0, %xmm1
-; CHECK-NEXT:    maxsd %xmm2, %xmm1
-; CHECK-NEXT:    movsd {{.*#+}} xmm3 = [2.147483647E+9,0.0E+0]
-; CHECK-NEXT:    minsd %xmm3, %xmm1
+; CHECK-NEXT:    minsd %xmm2, %xmm1
 ; CHECK-NEXT:    cvttsd2si %xmm1, %eax
 ; CHECK-NEXT:    xorl %ecx, %ecx
 ; CHECK-NEXT:    ucomisd %xmm0, %xmm0
 ; CHECK-NEXT:    cmovpl %ecx, %eax
 ; CHECK-NEXT:    movd %eax, %xmm1
 ; CHECK-NEXT:    unpckhpd {{.*#+}} xmm0 = xmm0[1,1]
+; CHECK-NEXT:    minsd %xmm0, %xmm2
+; CHECK-NEXT:    cvttsd2si %xmm2, %eax
 ; CHECK-NEXT:    ucomisd %xmm0, %xmm0
-; CHECK-NEXT:    maxsd %xmm2, %xmm0
-; CHECK-NEXT:    minsd %xmm3, %xmm0
-; CHECK-NEXT:    cvttsd2si %xmm0, %eax
 ; CHECK-NEXT:    cmovpl %ecx, %eax
 ; CHECK-NEXT:    movd %eax, %xmm0
 ; CHECK-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1]