[PowerPC] Eliminate compares - add i64 sext/zext handling for equality

[nemanjai]

Summary

This patch adds the SETEQ patterns for i64 values.

[inouehrs]

lib/Target/PowerPC/PPCISelDAGToDAG.cpp

2860 ↗

(On Diff #99597)

Do you mean CA bit by addc.1?
IMO, {addc.reg, addc.CA} = (addcarry (xor %a, %b), -1) or similar is clearer.
Someone may misunderstand that we need to execute addcarry twice with the current pseudo-code.

[nemanjai]

lib/Target/PowerPC/PPCISelDAGToDAG.cpp

2860 ↗

(On Diff #99597)

The idea was to use similar notation to that of the SDAG (i.e. result 0, result 1). But I do like your notation better. I'll change it as you suggested.

[echristo]

A few cosmetic changes in addition to the notation change that was mentioned (I also like that).

Otherwise LGTM.

-eric

lib/Target/PowerPC/PPCISelDAGToDAG.cpp

2833 ↗

(On Diff #99597)

Seems to be unused at the moment?

2855 ↗

(On Diff #99597)

Ditto.

2928 ↗

(On Diff #99597)

Can you move this down to just before the first use?

[nemanjai]

lib/Target/PowerPC/PPCISelDAGToDAG.cpp

2833 ↗

(On Diff #99597)

It's actually used in deciding whether we use the LHS or (xor LSH, RHS) as the operand to count leading zeros.

2855 ↗

(On Diff #99597)

Same as above, but as input to add-immediate-with-carry instead of count leading zeros.

llvm/trunk/lib/Target/PowerPC/PPCISelDAGToDAG.cpp

private:
     SDValue computeLogicOpInGPR(SDValue LogicOp);
     SDValue signExtendInputIfNeeded(SDValue Input);
     SDValue zeroExtendInputIfNeeded(SDValue Input);
     SDValue addExtOrTrunc(SDValue NatWidthRes, ExtOrTruncConversion Conv);
     SDValue get32BitZExtCompare(SDValue LHS, SDValue RHS, ISD::CondCode CC,
                                 int64_t RHSValue, SDLoc dl);
     SDValue get32BitSExtCompare(SDValue LHS, SDValue RHS, ISD::CondCode CC,
                                 int64_t RHSValue, SDLoc dl);
+    SDValue get64BitZExtCompare(SDValue LHS, SDValue RHS, ISD::CondCode CC,
+                                int64_t RHSValue, SDLoc dl);
+    SDValue get64BitSExtCompare(SDValue LHS, SDValue RHS, ISD::CondCode CC,
+                                int64_t RHSValue, SDLoc dl);
     SDValue getSETCCInGPR(SDValue Compare, SetccInGPROpts ConvOpts);
 
     void PeepholePPC64();
     void PeepholePPC64ZExt();
     void PeepholeCROps();
 
     SDValue combineToCMPB(SDNode *N);
     void foldBoolExts(SDValue &Res, SDNode *&N);
case ISD::SETEQ: {
     SDValue Sldi =
       SDValue(CurDAG->getMachineNode(PPC::RLDICR_32, dl, MVT::i32, SHLOps), 0);
     return SDValue(CurDAG->getMachineNode(PPC::SRADI_32, dl, MVT::i32, Sldi,
                                           getI32Imm(63, dl)), 0);
   }
   }
 }
 
+/// Produces a zero-extended result of comparing two 64-bit values according to
+/// the passed condition code.
+SDValue PPCDAGToDAGISel::get64BitZExtCompare(SDValue LHS, SDValue RHS,
+                                             ISD::CondCode CC,
+                                             int64_t RHSValue, SDLoc dl) {
+  bool IsRHSZero = RHSValue == 0;
+  switch (CC) {
+  default: return SDValue();
+  case ISD::SETEQ: {
+    // (zext (setcc %a, %b, seteq)) -> (lshr (ctlz (xor %a, %b)), 6)
+    // (zext (setcc %a, 0, seteq)) ->  (lshr (ctlz %a), 6)
+    SDValue Xor = IsRHSZero ? LHS :
+      SDValue(CurDAG->getMachineNode(PPC::XOR8, dl, MVT::i64, LHS, RHS), 0);
+    SDValue Clz =
+      SDValue(CurDAG->getMachineNode(PPC::CNTLZD, dl, MVT::i64, Xor), 0);
+    return SDValue(CurDAG->getMachineNode(PPC::RLDICL, dl, MVT::i64, Clz,
+                                          getI64Imm(58, dl), getI64Imm(63, dl)),
+                   0);
+  }
+  }
+}
+
+/// Produces a sign-extended result of comparing two 64-bit values according to
+/// the passed condition code.
+SDValue PPCDAGToDAGISel::get64BitSExtCompare(SDValue LHS, SDValue RHS,
+                                             ISD::CondCode CC,
+                                             int64_t RHSValue, SDLoc dl) {
+  bool IsRHSZero = RHSValue == 0;
+  switch (CC) {
+  default: return SDValue();
+  case ISD::SETEQ: {
+    // {addc.reg, addc.CA} = (addcarry (xor %a, %b), -1)
+    // (sext (setcc %a, %b, seteq)) -> (sube addc.reg, addc.reg, addc.CA)
+    // {addcz.reg, addcz.CA} = (addcarry %a, -1)
+    // (sext (setcc %a, 0, seteq)) -> (sube addcz.reg, addcz.reg, addcz.CA)
+    SDValue AddInput = IsRHSZero ? LHS :
+      SDValue(CurDAG->getMachineNode(PPC::XOR8, dl, MVT::i64, LHS, RHS), 0);
+    SDValue Addic =
+      SDValue(CurDAG->getMachineNode(PPC::ADDIC8, dl, MVT::i64, MVT::Glue,
+                                     AddInput, getI32Imm(~0U, dl)), 0);
+    return SDValue(CurDAG->getMachineNode(PPC::SUBFE8, dl, MVT::i64, Addic,
+                                          Addic, Addic.getValue(1)), 0);
+  }
+  }
+}
+
 /// Does this SDValue have any uses for which keeping the value in a GPR is
 /// appropriate. This is meant to be used on values that have type i1 since
 /// it is somewhat meaningless to ask if values of other types can be kept in
 /// GPR's.
 static bool allUsesExtend(SDValue Compare, SelectionDAG *CurDAG) {
   assert(Compare.getOpcode() == ISD::SETCC &&
          "An ISD::SETCC node required here.");
 
SDValue PPCDAGToDAGISel::getSETCCInGPR(SDValue Compare,
   SDValue LHS = Compare.getOperand(0);
   SDValue RHS = Compare.getOperand(1);
 
   // The condition code is operand 2 for SETCC and operand 4 for SELECT_CC.
   int CCOpNum = Compare.getOpcode() == ISD::SELECT_CC ? 4 : 2;
   ISD::CondCode CC =
     cast<CondCodeSDNode>(Compare.getOperand(CCOpNum))->get();
   EVT InputVT = LHS.getValueType();
-  if (InputVT != MVT::i32)
+  if (InputVT != MVT::i32 && InputVT != MVT::i64)
     return SDValue();
 
-  SDLoc dl(Compare);
-  ConstantSDNode *RHSConst = dyn_cast<ConstantSDNode>(RHS);
-  int64_t RHSValue = RHSConst ? RHSConst->getSExtValue() : INT64_MAX;
-
   if (ConvOpts == SetccInGPROpts::ZExtInvert ||
       ConvOpts == SetccInGPROpts::SExtInvert)
     CC = ISD::getSetCCInverse(CC, true);
 
-  if (ISD::isSignedIntSetCC(CC)) {
+  bool Inputs32Bit = InputVT == MVT::i32;
+  if (ISD::isSignedIntSetCC(CC) && Inputs32Bit) {
     LHS = signExtendInputIfNeeded(LHS);
     RHS = signExtendInputIfNeeded(RHS);
-  } else if (ISD::isUnsignedIntSetCC(CC)) {
+  } else if (ISD::isUnsignedIntSetCC(CC) && Inputs32Bit) {
     LHS = zeroExtendInputIfNeeded(LHS);
     RHS = zeroExtendInputIfNeeded(RHS);
   }
 
+  SDLoc dl(Compare);
+  ConstantSDNode *RHSConst = dyn_cast<ConstantSDNode>(RHS);
+  int64_t RHSValue = RHSConst ? RHSConst->getSExtValue() : INT64_MAX;
   bool IsSext = ConvOpts == SetccInGPROpts::SExtOrig ||
     ConvOpts == SetccInGPROpts::SExtInvert;
-  if (IsSext)
+
+  if (IsSext && Inputs32Bit)
     return get32BitSExtCompare(LHS, RHS, CC, RHSValue, dl);
+  else if (Inputs32Bit)
     return get32BitZExtCompare(LHS, RHS, CC, RHSValue, dl);
+  else if (IsSext)
+    return get64BitSExtCompare(LHS, RHS, CC, RHSValue, dl);
+  return get64BitZExtCompare(LHS, RHS, CC, RHSValue, dl);
 }
 
 void PPCDAGToDAGISel::transferMemOperands(SDNode *N, SDNode *Result) {
   // Transfer memoperands.
   MachineSDNode::mmo_iterator MemOp = MF->allocateMemRefsArray(1);
   MemOp[0] = cast<MemSDNode>(N)->getMemOperand();
   cast<MachineSDNode>(Result)->setMemRefs(MemOp, MemOp + 1);
 }

llvm/trunk/test/CodeGen/PowerPC/expand-isel.ll

if.end:
   %conv3 = zext i1 %cmp to i32
   br label %cleanup
 
 cleanup:
   %retval.0 = phi i32 [ %conv3, %if.end ], [ 1, %entry ]
   ret i32 %retval.0
 
 ; CHECK-LABEL: @testComplexISEL
-; CHECK: bc 12, 2, [[TRUE:.LBB[0-9]+]]
-; CHECK-NEXT: b [[SUCCESSOR:.LBB[0-9]+]]
-; CHECK-NEXT:  [[TRUE]]
-; CHECK-NEXT: addi r3, r12, 0
-; CHECK-NEXT: [[SUCCESSOR]]
-; CHECK-NEXT: clrldi r3, r3, 32
-; CHECK-NEXT: blr
+; CHECK-DAG: [[LI:r[0-9]+]], 1
+; CHECK-DAG: cmplwi [[LD:r[0-9]+]], 0
+; CHECK: beq cr0, [[EQ:.LBB[0-9_]+]]
+; CHECK: blr
+; CHECK: [[EQ]]
+; CHECK: xor [[XOR:r[0-9]+]]
+; CHECK: cntlzd [[CZ:r[0-9]+]], [[XOR]]
+; CHECK: rldicl [[SH:r[0-9]+]], [[CZ]], 58, 63
 }
 
 !1 = !{!2, !2, i64 0}
 !2 = !{!"int", !3, i64 0}
 !3 = !{!"omnipotent char", !4, i64 0}
 !4 = !{!"Simple C/C++ TBAA"}

llvm/trunk/test/CodeGen/PowerPC/logic-ops-on-compares.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu -O2 \
 ; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
 ; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
 ; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu -O2 \
 ; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
 ; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
 
 ; Function Attrs: nounwind
-define signext i32 @test(i32 signext %a, i32 signext %b, i32 signext %c) {
-; CHECK-LABEL: test:
+define signext i32 @logic_ne_32(i32 signext %a, i32 signext %b, i32 signext %c) {
+; CHECK-LABEL: logic_ne_32:
 ; CHECK:    xor r7, r3, r4
 ; CHECK-NEXT:    li r6, 55
 ; CHECK-NEXT:    xor r5, r5, r6
 ; CHECK-NEXT:    or r7, r7, r4
 ; CHECK-NEXT:    cntlzw r5, r5
 ; CHECK-NEXT:    cntlzw r6, r7
 ; CHECK-NEXT:    srwi r6, r6, 5
 ; CHECK-NEXT:    srwi r5, r5, 5

 if.end29.thread136:                               ; preds = %if.else
   unreachable
 
 if.end29:                                         ; preds = %if.else
   ret void
 
 }
 
+; Function Attrs: nounwind
+define i64 @logic_ne_64(i64 %a, i64 %b, i64 %c) {
+; CHECK-LABEL: logic_ne_64:
+; CHECK:    xor r7, r3, r4
+; CHECK-NEXT:    li r6, 55
+; CHECK-NEXT:    xor r5, r5, r6
+; CHECK-NEXT:    or r7, r7, r4
+; CHECK-NEXT:    cntlzd r6, r7
+; CHECK-NEXT:    cntlzd r5, r5
+; CHECK-NEXT:    rldicl r6, r6, 58, 63
+; CHECK-NEXT:    rldicl r5, r5, 58, 63
+; CHECK-NEXT:    or. r5, r6, r5
+; CHECK-NEXT:    bc 4, 1
+entry:
+  %tobool = icmp eq i64 %a, %b
+  %tobool1 = icmp eq i64 %b, 0
+  %or.cond = and i1 %tobool, %tobool1
+  %tobool3 = icmp eq i64 %c, 55
+  %or.cond5 = or i1 %or.cond, %tobool3
+  br i1 %or.cond5, label %if.end, label %if.then
+
+if.then:                                          ; preds = %entry
+  %call = tail call i64 @foo64(i64 %a) #2
+  br label %return
+
+if.end:                                           ; preds = %entry
+  %call4 = tail call i64 @bar64(i64 %b) #2
+  br label %return
+
+return:                                           ; preds = %if.end, %if.then
+  %retval.0 = phi i64 [ %call4, %if.end ], [ %call, %if.then ]
+  ret i64 %retval.0
+}
+
+define void @neg_truncate_i64(i64 *%ptr) {
+; CHECK-LABEL: neg_truncate_i64:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    ld r3, 0(r3)
+; CHECK-NEXT:    rldicl. r3, r3, 0, 63
+; CHECK-NEXT:    bclr 12, 2, 0
+; CHECK-NEXT:  # BB#1: # %if.end29.thread136
+; CHECK-NEXT:  .LBB3_2: # %if.end29
+entry:
+  %0 = load i64, i64* %ptr, align 4
+  %rem17127 = and i64 %0, 1
+  %cmp18 = icmp eq i64 %rem17127, 0
+  br label %if.else
+
+if.else:                                          ; preds = %entry
+  br i1 %cmp18, label %if.end29, label %if.end29.thread136
+
+if.end29.thread136:                               ; preds = %if.else
+  unreachable
+
+if.end29:                                         ; preds = %if.else
+  ret void
+
+}
+
 declare signext i32 @foo(i32 signext)
 declare signext i32 @bar(i32 signext)
+declare i64 @foo64(i64)
+declare i64 @bar64(i64)

llvm/trunk/test/CodeGen/PowerPC/testComparesieqsll.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu -O2 \
+; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
+; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu -O2 \
+; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
+; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
+; ModuleID = 'ComparisonTestCases/testComparesieqsll.c'
+
+@glob = common local_unnamed_addr global i64 0, align 8
+
+; Function Attrs: norecurse nounwind readnone
+define signext i32 @test_ieqsll(i64 %a, i64 %b) {
+; CHECK-LABEL: test_ieqsll:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv = zext i1 %cmp to i32
+  ret i32 %conv
+}
+
+; Function Attrs: norecurse nounwind readnone
+define signext i32 @test_ieqsll_sext(i64 %a, i64 %b) {
+; CHECK-LABEL: test_ieqsll_sext:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %sub = sext i1 %cmp to i32
+  ret i32 %sub
+}
+
+; Function Attrs: norecurse nounwind readnone
+define signext i32 @test_ieqsll_z(i64 %a) {
+; CHECK-LABEL: test_ieqsll_z:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv = zext i1 %cmp to i32
+  ret i32 %conv
+}
+
+; Function Attrs: norecurse nounwind readnone
+define signext i32 @test_ieqsll_sext_z(i64 %a) {
+; CHECK-LABEL: test_ieqsll_sext_z:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %sub = sext i1 %cmp to i32
+  ret i32 %sub
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_ieqsll_store(i64 %a, i64 %b) {
+; CHECK-LABEL: test_ieqsll_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r5, r2, .LC0@toc@ha
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    ld r12, .LC0@toc@l(r5)
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    std r3, 0(r12)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = zext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_ieqsll_sext_store(i64 %a, i64 %b) {
+; CHECK-LABEL: test_ieqsll_sext_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r5, r2, .LC0@toc@ha
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    ld r12, .LC0@toc@l(r5)
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    std r3, 0(r12)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = sext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_ieqsll_z_store(i64 %a) {
+; CHECK-LABEL: test_ieqsll_z_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r4, r2, .LC0@toc@ha
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    ld r4, .LC0@toc@l(r4)
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    std r3, 0(r4)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = zext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_ieqsll_sext_z_store(i64 %a) {
+; CHECK-LABEL: test_ieqsll_sext_z_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r4, r2, .LC0@toc@ha
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    ld r4, .LC0@toc@l(r4)
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    std r3, 0(r4)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = sext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}

llvm/trunk/test/CodeGen/PowerPC/testComparesiequll.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu -O2 \
+; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
+; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu -O2 \
+; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
+; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
+; ModuleID = 'ComparisonTestCases/testComparesiequll.c'
+
+@glob = common local_unnamed_addr global i64 0, align 8
+
+; Function Attrs: norecurse nounwind readnone
+define signext i32 @test_iequll(i64 %a, i64 %b) {
+; CHECK-LABEL: test_iequll:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv = zext i1 %cmp to i32
+  ret i32 %conv
+}
+
+; Function Attrs: norecurse nounwind readnone
+define signext i32 @test_iequll_sext(i64 %a, i64 %b) {
+; CHECK-LABEL: test_iequll_sext:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %sub = sext i1 %cmp to i32
+  ret i32 %sub
+}
+
+; Function Attrs: norecurse nounwind readnone
+define signext i32 @test_iequll_z(i64 %a) {
+; CHECK-LABEL: test_iequll_z:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv = zext i1 %cmp to i32
+  ret i32 %conv
+}
+
+; Function Attrs: norecurse nounwind readnone
+define signext i32 @test_iequll_sext_z(i64 %a) {
+; CHECK-LABEL: test_iequll_sext_z:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %sub = sext i1 %cmp to i32
+  ret i32 %sub
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_iequll_store(i64 %a, i64 %b) {
+; CHECK-LABEL: test_iequll_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r5, r2, .LC0@toc@ha
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    ld r12, .LC0@toc@l(r5)
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    std r3, 0(r12)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = zext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_iequll_sext_store(i64 %a, i64 %b) {
+; CHECK-LABEL: test_iequll_sext_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r5, r2, .LC0@toc@ha
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    ld r12, .LC0@toc@l(r5)
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    std r3, 0(r12)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = sext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_iequll_z_store(i64 %a) {
+; CHECK-LABEL: test_iequll_z_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r4, r2, .LC0@toc@ha
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    ld r4, .LC0@toc@l(r4)
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    std r3, 0(r4)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = zext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_iequll_sext_z_store(i64 %a) {
+; CHECK-LABEL: test_iequll_sext_z_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r4, r2, .LC0@toc@ha
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    ld r4, .LC0@toc@l(r4)
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    std r3, 0(r4)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = sext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}

llvm/trunk/test/CodeGen/PowerPC/testCompareslleqsll.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu -O2 \
+; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
+; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu -O2 \
+; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
+; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
+
+@glob = common local_unnamed_addr global i64 0, align 8
+
+; Function Attrs: norecurse nounwind readnone
+define i64 @test_lleqsll(i64 %a, i64 %b) {
+; CHECK-LABEL: test_lleqsll:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = zext i1 %cmp to i64
+  ret i64 %conv1
+}
+
+; Function Attrs: norecurse nounwind readnone
+define i64 @test_lleqsll_sext(i64 %a, i64 %b) {
+; CHECK-LABEL: test_lleqsll_sext:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = sext i1 %cmp to i64
+  ret i64 %conv1
+}
+
+; Function Attrs: norecurse nounwind readnone
+define i64 @test_lleqsll_z(i64 %a) {
+; CHECK-LABEL: test_lleqsll_z:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = zext i1 %cmp to i64
+  ret i64 %conv1
+}
+
+; Function Attrs: norecurse nounwind readnone
+define i64 @test_lleqsll_sext_z(i64 %a) {
+; CHECK-LABEL: test_lleqsll_sext_z:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = sext i1 %cmp to i64
+  ret i64 %conv1
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_lleqsll_store(i64 %a, i64 %b) {
+; CHECK-LABEL: test_lleqsll_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r5, r2, .LC0@toc@ha
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    ld r12, .LC0@toc@l(r5)
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    std r3, 0(r12)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = zext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_lleqsll_sext_store(i64 %a, i64 %b) {
+; CHECK-LABEL: test_lleqsll_sext_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r5, r2, .LC0@toc@ha
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    ld r12, .LC0@toc@l(r5)
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    std r3, 0(r12)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = sext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_lleqsll_z_store(i64 %a) {
+; CHECK-LABEL: test_lleqsll_z_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r4, r2, .LC0@toc@ha
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    ld r4, .LC0@toc@l(r4)
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    std r3, 0(r4)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = zext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_lleqsll_sext_z_store(i64 %a) {
+; CHECK-LABEL: test_lleqsll_sext_z_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r4, r2, .LC0@toc@ha
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    ld r4, .LC0@toc@l(r4)
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    std r3, 0(r4)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = sext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}

llvm/trunk/test/CodeGen/PowerPC/testComparesllequll.ll

+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu -O2 \
+; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
+; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu -O2 \
+; RUN:   -ppc-asm-full-reg-names -mcpu=pwr8 < %s | FileCheck %s \
+; RUN:  --implicit-check-not cmpw --implicit-check-not cmpd --implicit-check-not cmpl
+
+@glob = common local_unnamed_addr global i64 0, align 8
+
+; Function Attrs: norecurse nounwind readnone
+define i64 @test_llequll(i64 %a, i64 %b) {
+; CHECK-LABEL: test_llequll:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = zext i1 %cmp to i64
+  ret i64 %conv1
+}
+
+; Function Attrs: norecurse nounwind readnone
+define i64 @test_llequll_sext(i64 %a, i64 %b) {
+; CHECK-LABEL: test_llequll_sext:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = sext i1 %cmp to i64
+  ret i64 %conv1
+}
+
+; Function Attrs: norecurse nounwind readnone
+define i64 @test_llequll_z(i64 %a) {
+; CHECK-LABEL: test_llequll_z:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = zext i1 %cmp to i64
+  ret i64 %conv1
+}
+
+; Function Attrs: norecurse nounwind readnone
+define i64 @test_llequll_sext_z(i64 %a) {
+; CHECK-LABEL: test_llequll_sext_z:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = sext i1 %cmp to i64
+  ret i64 %conv1
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_llequll_store(i64 %a, i64 %b) {
+; CHECK-LABEL: test_llequll_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r5, r2, .LC0@toc@ha
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    ld r12, .LC0@toc@l(r5)
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    std r3, 0(r12)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = zext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_llequll_sext_store(i64 %a, i64 %b) {
+; CHECK-LABEL: test_llequll_sext_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r5, r2, .LC0@toc@ha
+; CHECK-NEXT:    xor r3, r3, r4
+; CHECK-NEXT:    ld r12, .LC0@toc@l(r5)
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    std r3, 0(r12)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, %b
+  %conv1 = sext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_llequll_z_store(i64 %a) {
+; CHECK-LABEL: test_llequll_z_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r4, r2, .LC0@toc@ha
+; CHECK-NEXT:    cntlzd r3, r3
+; CHECK-NEXT:    ld r4, .LC0@toc@l(r4)
+; CHECK-NEXT:    rldicl r3, r3, 58, 63
+; CHECK-NEXT:    std r3, 0(r4)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = zext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}
+
+; Function Attrs: norecurse nounwind
+define void @test_llequll_sext_z_store(i64 %a) {
+; CHECK-LABEL: test_llequll_sext_z_store:
+; CHECK:       # BB#0: # %entry
+; CHECK-NEXT:    addis r4, r2, .LC0@toc@ha
+; CHECK-NEXT:    addic r3, r3, -1
+; CHECK-NEXT:    ld r4, .LC0@toc@l(r4)
+; CHECK-NEXT:    subfe r3, r3, r3
+; CHECK-NEXT:    std r3, 0(r4)
+; CHECK-NEXT:    blr
+entry:
+  %cmp = icmp eq i64 %a, 0
+  %conv1 = sext i1 %cmp to i64
+  store i64 %conv1, i64* @glob, align 8
+  ret void
+}