[SelectionDAG] ComputeKnownBits - merge getValidMinimumShiftAmountConstant() and generic ISD::SRL handling.

As mentioned by @nikic on rGef5debac4302 (although that was just about SHL), we can merge the guaranteed top zero bits from the shifted value, and then, if a min shift amount is known, zero out the top bits as well.

SHL tests / handling will be added in a follow up patch.

Author: RKSimon

llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp

@@ -2885,23 +2885,25 @@ KnownBits SelectionDAG::computeKnownBits(SDValue Op, const APInt &DemandedElts,
     }
     break;
   case ISD::SRL:
+    Known = computeKnownBits(Op.getOperand(0), DemandedElts, Depth + 1);
+
     if (const APInt *ShAmt = getValidShiftAmountConstant(Op, DemandedElts)) {
-      Known = computeKnownBits(Op.getOperand(0), DemandedElts, Depth + 1);
       unsigned Shift = ShAmt->getZExtValue();
       Known.Zero.lshrInPlace(Shift);
       Known.One.lshrInPlace(Shift);
       // High bits are known zero.
       Known.Zero.setHighBits(Shift);
-    } else if (const APInt *ShMinAmt =
-                   getValidMinimumShiftAmountConstant(Op, DemandedElts)) {
-      // Minimum shift high bits are known zero.
-      Known.Zero.setHighBits(ShMinAmt->getZExtValue());
-    } else {
-      // No matter the shift amount, the leading zeros will stay zero.
-      Known = computeKnownBits(Op.getOperand(0), DemandedElts, Depth + 1);
-      Known.Zero = APInt::getHighBitsSet(BitWidth, Known.countMinLeadingZeros());
-      Known.One.clearAllBits();
+      break;
     }
+
+    // No matter the shift amount, the leading zeros will stay zero.
+    Known.Zero = APInt::getHighBitsSet(BitWidth, Known.countMinLeadingZeros());
+    Known.One.clearAllBits();
+
+    // Minimum shift high bits are known zero.
+    if (const APInt *ShMinAmt =
+            getValidMinimumShiftAmountConstant(Op, DemandedElts))
+      Known.Zero.setHighBits(ShMinAmt->getZExtValue());
     break;
   case ISD::SRA:
     if (const APInt *ShAmt = getValidShiftAmountConstant(Op, DemandedElts)) {

llvm/test/CodeGen/X86/vector-idiv-udiv-128.ll

@@ -405,7 +405,6 @@ define <16 x i8> @test_divconstant_16i8(<16 x i8> %a) nounwind {
 ; AVX512BW-NEXT:    vmovdqa {{.*#+}} ymm1 = [0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0]
 ; AVX512BW-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX512BW-NEXT:    vpsrlvw %zmm1, %zmm2, %zmm1
-; AVX512BW-NEXT:    vpand {{.*}}(%rip), %ymm1, %ymm1
 ; AVX512BW-NEXT:    vpmullw {{.*}}(%rip), %ymm1, %ymm1
 ; AVX512BW-NEXT:    vpsrlw $8, %ymm1, %ymm1
 ; AVX512BW-NEXT:    vpmovwb %zmm1, %ymm1
@@ -923,7 +922,6 @@ define <16 x i8> @test_remconstant_16i8(<16 x i8> %a) nounwind {
 ; AVX512BW-NEXT:    vmovdqa {{.*#+}} ymm1 = [0,0,0,0,0,0,0,1,1,0,0,0,0,0,0,0]
 ; AVX512BW-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
 ; AVX512BW-NEXT:    vpsrlvw %zmm1, %zmm2, %zmm1
-; AVX512BW-NEXT:    vpand {{.*}}(%rip), %ymm1, %ymm1
 ; AVX512BW-NEXT:    vpmullw {{.*}}(%rip), %ymm1, %ymm1
 ; AVX512BW-NEXT:    vpsrlw $8, %ymm1, %ymm1
 ; AVX512BW-NEXT:    vpmovwb %zmm1, %ymm1

llvm/test/CodeGen/X86/vector-idiv-udiv-256.ll

@@ -345,7 +345,6 @@ define <32 x i8> @test_divconstant_32i8(<32 x i8> %a) nounwind {
 ; AVX512BW:       # %bb.0:
 ; AVX512BW-NEXT:    vpmovzxbw {{.*#+}} zmm1 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero,ymm0[16],zero,ymm0[17],zero,ymm0[18],zero,ymm0[19],zero,ymm0[20],zero,ymm0[21],zero,ymm0[22],zero,ymm0[23],zero,ymm0[24],zero,ymm0[25],zero,ymm0[26],zero,ymm0[27],zero,ymm0[28],zero,ymm0[29],zero,ymm0[30],zero,ymm0[31],zero
 ; AVX512BW-NEXT:    vpsrlvw {{.*}}(%rip), %zmm1, %zmm1
-; AVX512BW-NEXT:    vpandq {{.*}}(%rip), %zmm1, %zmm1
 ; AVX512BW-NEXT:    vpmullw {{.*}}(%rip), %zmm1, %zmm1
 ; AVX512BW-NEXT:    vpsrlw $8, %zmm1, %zmm1
 ; AVX512BW-NEXT:    vpmovwb %zmm1, %ymm1
@@ -793,7 +792,6 @@ define <32 x i8> @test_remconstant_32i8(<32 x i8> %a) nounwind {
 ; AVX512BW:       # %bb.0:
 ; AVX512BW-NEXT:    vpmovzxbw {{.*#+}} zmm1 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero,ymm0[16],zero,ymm0[17],zero,ymm0[18],zero,ymm0[19],zero,ymm0[20],zero,ymm0[21],zero,ymm0[22],zero,ymm0[23],zero,ymm0[24],zero,ymm0[25],zero,ymm0[26],zero,ymm0[27],zero,ymm0[28],zero,ymm0[29],zero,ymm0[30],zero,ymm0[31],zero
 ; AVX512BW-NEXT:    vpsrlvw {{.*}}(%rip), %zmm1, %zmm1
-; AVX512BW-NEXT:    vpandq {{.*}}(%rip), %zmm1, %zmm1
 ; AVX512BW-NEXT:    vpmullw {{.*}}(%rip), %zmm1, %zmm1
 ; AVX512BW-NEXT:    vpsrlw $8, %zmm1, %zmm1
 ; AVX512BW-NEXT:    vpmovwb %zmm1, %ymm1