AMDGPU/SI: Prevent the DAGCombiner from creating setcc with i1 inputs

Reviewers: arsenm

Subscribers: arsenm, llvm-commits

Differential Revision: http://reviews.llvm.org/D15035

llvm-svn: 258256

llvm/lib/Target/AMDGPU/SIISelLowering.cpp

@@ -533,6 +533,16 @@ bool SITargetLowering::shouldConvertConstantLoadToIntImm(const APInt &Imm,
   return TII->isInlineConstant(Imm);
 }
 
+bool SITargetLowering::isTypeDesirableForOp(unsigned Op, EVT VT) const {
+
+  // SimplifySetCC uses this function to determine whether or not it should
+  // create setcc with i1 operands.  We don't have instructions for i1 setcc.
+  if (VT == MVT::i1 && Op == ISD::SETCC)
+    return false;
+
+  return TargetLowering::isTypeDesirableForOp(Op, VT);
+}
+
 SDValue SITargetLowering::LowerParameter(SelectionDAG &DAG, EVT VT, EVT MemVT,
                                          SDLoc SL, SDValue Chain,
                                          unsigned Offset, bool Signed) const {

llvm/lib/Target/AMDGPU/SIISelLowering.h

@@ -89,6 +89,8 @@ public:
   bool shouldConvertConstantLoadToIntImm(const APInt &Imm,
                                         Type *Ty) const override;
 
+  bool isTypeDesirableForOp(unsigned Op, EVT VT) const override;
+
   SDValue LowerFormalArguments(SDValue Chain, CallingConv::ID CallConv,
                                bool isVarArg,
                                const SmallVectorImpl<ISD::InputArg> &Ins,

llvm/test/CodeGen/AMDGPU/setcc-opt.ll

@@ -68,6 +68,34 @@ define void @sext_bool_icmp_ne_1(i1 addrspace(1)* %out, i32 %a, i32 %b) nounwind
   ret void
 }
 
+; FUNC-LABEL: {{^}}sext_bool_icmp_eq_neg1:
+; GCN-NOT: v_cmp
+; GCN: v_cmp_eq_i32_e32 vcc,
+; GCN-NEXT: v_cndmask_b32_e64 [[RESULT:v[0-9]+]], 0, 1, vcc
+; GCN-NEXT: buffer_store_byte [[RESULT]]
+; GCN-NEXT: s_endpgm
+define void @sext_bool_icmp_eq_neg1(i1 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
+  %icmp0 = icmp eq i32 %a, %b
+  %ext = sext i1 %icmp0 to i32
+  %icmp1 = icmp eq i32 %ext, -1
+  store i1 %icmp1, i1 addrspace(1)* %out
+  ret void
+}
+
+; FUNC-LABEL: {{^}}sext_bool_icmp_ne_neg1:
+; GCN-NOT: v_cmp
+; GCN: v_cmp_eq_i32_e32 vcc,
+; GCN-NEXT: v_cndmask_b32_e64 [[RESULT:v[0-9]+]], 0, 1, vcc
+; GCN-NEXT: buffer_store_byte [[RESULT]]
+; GCN-NEXT: s_endpgm
+define void @sext_bool_icmp_ne_neg1(i1 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
+  %icmp0 = icmp ne i32 %a, %b
+  %ext = sext i1 %icmp0 to i32
+  %icmp1 = icmp ne i32 %ext, -1
+  store i1 %icmp1, i1 addrspace(1)* %out
+  ret void
+}
+
 ; FUNC-LABEL: {{^}}zext_bool_icmp_eq_0:
 ; GCN-NOT: v_cmp
 ; GCN: v_cmp_ne_i32_e32 vcc,
@@ -123,6 +151,32 @@ define void @zext_bool_icmp_ne_1(i1 addrspace(1)* %out, i32 %a, i32 %b) nounwind
   ret void
 }
 
+; Reduces to false:
+; FUNC-LABEL: {{^}}zext_bool_icmp_eq_neg1:
+; GCN: v_mov_b32_e32 [[TMP:v[0-9]+]], 0{{$}}
+; GCN: buffer_store_byte [[TMP]]
+; GCN-NEXT: s_endpgm
+define void @zext_bool_icmp_eq_neg1(i1 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
+  %icmp0 = icmp eq i32 %a, %b
+  %ext = zext i1 %icmp0 to i32
+  %icmp1 = icmp eq i32 %ext, -1
+  store i1 %icmp1, i1 addrspace(1)* %out
+  ret void
+}
+
+; Reduces to true:
+; FUNC-LABEL: {{^}}zext_bool_icmp_ne_neg1:
+; GCN: v_mov_b32_e32 [[TMP:v[0-9]+]], 1{{$}}
+; GCN: buffer_store_byte [[TMP]]
+; GCN-NEXT: s_endpgm
+define void @zext_bool_icmp_ne_neg1(i1 addrspace(1)* %out, i32 %a, i32 %b) nounwind {
+  %icmp0 = icmp ne i32 %a, %b
+  %ext = zext i1 %icmp0 to i32
+  %icmp1 = icmp ne i32 %ext, -1
+  store i1 %icmp1, i1 addrspace(1)* %out
+  ret void
+}
+
 ; FUNC-LABEL: {{^}}sext_bool_icmp_ne_k:
 ; SI-DAG: s_load_dword [[A:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xb
 ; SI-DAG: s_load_dword [[B:s[0-9]+]], s{{\[[0-9]+:[0-9]+\]}}, 0xc

llvm/test/CodeGen/AMDGPU/setcc.ll

@@ -375,3 +375,17 @@ define void @v3i8_eq(<3 x i8> addrspace(1)* %out, <3 x i8> addrspace(1)* %ptra,
   store <3 x i8> %ext, <3 x i8> addrspace(1)* %gep.out
   ret void
 }
+
+; Make sure we don't try to emit i1 setcc ops
+; FUNC-LABEL: setcc-i1
+; SI: s_and_b32 [[AND:s[0-9]+]], s{{[0-9]+}}, 1
+; SI: v_cmp_eq_i32_e64 s[0:1], 0, [[AND]]
+define void @setcc-i1(i32 %in) {
+  %and = and i32 %in, 1
+  %cmp = icmp eq i32 %and, 0
+  br i1 %cmp, label %endif, label %if
+if:
+  unreachable
+endif:
+  ret void
+}

llvm/test/CodeGen/AMDGPU/trunc-cmp-constant.ll

@@ -4,8 +4,7 @@
 ; FUNC-LABEL {{^}}sextload_i1_to_i32_trunc_cmp_eq_0:
 ; SI: buffer_load_ubyte [[LOAD:v[0-9]+]]
 ; SI: v_and_b32_e32 [[TMP:v[0-9]+]], 1, [[LOAD]]
-; SI: v_cmp_eq_i32_e32 vcc, 1, [[TMP]]{{$}}
-; SI: s_xor_b64 s{{\[[0-9]+:[0-9]+\]}}, vcc, -1{{$}}
+; SI: v_cmp_eq_i32_e32 vcc, 0, [[TMP]]{{$}}
 ; SI: v_cndmask_b32_e64
 ; SI: buffer_store_byte
 define void @sextload_i1_to_i32_trunc_cmp_eq_0(i1 addrspace(1)* %out, i1 addrspace(1)* %in) nounwind {