GlobalISel: support selection of G_ICMP on AArch64.

Patch from Ahmed Bougaca again.

llvm-svn: 284072

llvm/lib/Target/AArch64/AArch64InstructionSelector.cpp

@@ -274,6 +274,33 @@ static bool selectCopy(MachineInstr &I, const TargetInstrInfo &TII,
   return true;
 }
 
+static AArch64CC::CondCode changeICMPPredToAArch64CC(CmpInst::Predicate P) {
+  switch (P) {
+  default:
+    llvm_unreachable("Unknown condition code!");
+  case CmpInst::ICMP_NE:
+    return AArch64CC::NE;
+  case CmpInst::ICMP_EQ:
+    return AArch64CC::EQ;
+  case CmpInst::ICMP_SGT:
+    return AArch64CC::GT;
+  case CmpInst::ICMP_SGE:
+    return AArch64CC::GE;
+  case CmpInst::ICMP_SLT:
+    return AArch64CC::LT;
+  case CmpInst::ICMP_SLE:
+    return AArch64CC::LE;
+  case CmpInst::ICMP_UGT:
+    return AArch64CC::HI;
+  case CmpInst::ICMP_UGE:
+    return AArch64CC::HS;
+  case CmpInst::ICMP_ULT:
+    return AArch64CC::LO;
+  case CmpInst::ICMP_ULE:
+    return AArch64CC::LS;
+  }
+}
+
 bool AArch64InstructionSelector::select(MachineInstr &I) const {
   assert(I.getParent() && "Instruction should be in a basic block!");
   assert(I.getParent()->getParent() && "Instruction should be in a function!");
@@ -577,6 +604,50 @@ bool AArch64InstructionSelector::select(MachineInstr &I) const {
   case TargetOpcode::G_PTRTOINT:
   case TargetOpcode::G_BITCAST:
     return selectCopy(I, TII, MRI, TRI, RBI);
+
+  case TargetOpcode::G_ICMP: {
+    if (Ty != LLT::scalar(1)) {
+      DEBUG(dbgs() << "G_ICMP result has type: " << Ty
+                   << ", expected: " << LLT::scalar(1) << '\n');
+      return false;
+    }
+
+    unsigned CmpOpc = 0;
+    unsigned ZReg = 0;
+
+    LLT CmpTy = MRI.getType(I.getOperand(2).getReg());
+    if (CmpTy == LLT::scalar(32)) {
+      CmpOpc = AArch64::SUBSWrr;
+      ZReg = AArch64::WZR;
+    } else if (CmpTy == LLT::scalar(64) || CmpTy.isPointer()) {
+      CmpOpc = AArch64::SUBSXrr;
+      ZReg = AArch64::XZR;
+    } else {
+      return false;
+    }
+
+    const AArch64CC::CondCode CC = changeICMPPredToAArch64CC(
+        (CmpInst::Predicate)I.getOperand(1).getPredicate());
+
+    MachineInstr &CmpMI = *BuildMI(MBB, I, I.getDebugLoc(), TII.get(CmpOpc))
+                               .addDef(ZReg)
+                               .addUse(I.getOperand(2).getReg())
+                               .addUse(I.getOperand(3).getReg());
+
+    MachineInstr &CSetMI =
+        *BuildMI(MBB, I, I.getDebugLoc(), TII.get(AArch64::CSINCWr))
+             .addDef(I.getOperand(0).getReg())
+             .addUse(AArch64::WZR)
+             .addUse(AArch64::WZR)
+             .addImm(CC);
+
+    constrainSelectedInstRegOperands(CmpMI, TII, TRI, RBI);
+    constrainSelectedInstRegOperands(CSetMI, TII, TRI, RBI);
+
+    I.eraseFromParent();
+    return true;
+  }
+
   }
 
   return false;

llvm/test/CodeGen/AArch64/GlobalISel/arm64-instructionselect.mir

@@ -99,6 +99,7 @@
   define void @bitcast_s64_gpr_fpr() { ret void }
   define void @bitcast_s64_fpr_gpr() { ret void }
 
+  define void @icmp() { ret void }
 ...
 
 ---
@@ -1682,3 +1683,48 @@ body:             |
     %0(s64) = COPY %d0
     %1(s64) = G_BITCAST %0
 ...
+
+---
+# CHECK-LABEL: name: icmp
+name:            icmp
+legalized:       true
+regBankSelected: true
+
+# CHECK:      registers:
+# CHECK-NEXT:  - { id: 0, class: gpr32 }
+# CHECK-NEXT:  - { id: 1, class: gpr32 }
+# CHECK-NEXT:  - { id: 2, class: gpr64 }
+# CHECK-NEXT:  - { id: 3, class: gpr32 }
+# CHECK-NEXT:  - { id: 4, class: gpr64 }
+# CHECK-NEXT:  - { id: 5, class: gpr32 }
+registers:
+  - { id: 0, class: gpr }
+  - { id: 1, class: gpr }
+  - { id: 2, class: gpr }
+  - { id: 3, class: gpr }
+  - { id: 4, class: gpr }
+  - { id: 5, class: gpr }
+
+# CHECK:  body:
+# CHECK:    %wzr = SUBSWrr %0, %0, implicit-def %nzcv
+# CHECK:    %1 = CSINCWr %wzr, %wzr, 0, implicit %nzcv
+
+# CHECK:    %xzr = SUBSXrr %2, %2, implicit-def %nzcv
+# CHECK:    %3 = CSINCWr %wzr, %wzr, 2, implicit %nzcv
+
+# CHECK:    %xzr = SUBSXrr %4, %4, implicit-def %nzcv
+# CHECK:    %5 = CSINCWr %wzr, %wzr, 1, implicit %nzcv
+
+body:             |
+  bb.0:
+    liveins: %w0, %x0
+
+    %0(s32) = COPY %w0
+    %1(s1) = G_ICMP intpred(eq), %0, %0
+
+    %2(s64) = COPY %x0
+    %3(s1) = G_ICMP intpred(uge), %2, %2
+
+    %4(p0) = COPY %x0
+    %5(s1) = G_ICMP intpred(ne), %4, %4
+...