[X86] Add ADD/SUB SSAT/USAT vector costs (PR40123)

Costs for real SSE2 instructions

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@350295 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86TargetTransformInfo.cpp b/lib/Target/X86/X86TargetTransformInfo.cpp
index 401ad7979809cc88e8870205a2ed88eb3a7a83e6..89f45dd5461bae86062aa8aae6973fc733fa6bd4 100644 (file)
--- a/lib/Target/X86/X86TargetTransformInfo.cpp
+++ b/lib/Target/X86/X86TargetTransformInfo.cpp
@@ -1742,6 +1742,14 @@ int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
     { ISD::CTTZ,       MVT::v16i32, 14 },
     { ISD::CTTZ,       MVT::v32i16, 12 },
     { ISD::CTTZ,       MVT::v64i8,   9 },
+    { ISD::SADDSAT,    MVT::v32i16,  1 },
+    { ISD::SADDSAT,    MVT::v64i8,   1 },
+    { ISD::SSUBSAT,    MVT::v32i16,  1 },
+    { ISD::SSUBSAT,    MVT::v64i8,   1 },
+    { ISD::UADDSAT,    MVT::v32i16,  1 },
+    { ISD::UADDSAT,    MVT::v64i8,   1 },
+    { ISD::USUBSAT,    MVT::v32i16,  1 },
+    { ISD::USUBSAT,    MVT::v64i8,   1 },
   };
   static const CostTblEntry AVX512CostTbl[] = {
     { ISD::BITREVERSE, MVT::v8i64,  36 },
@@ -1787,6 +1795,14 @@ int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
     { ISD::CTTZ,       MVT::v8i32,  14 },
     { ISD::CTTZ,       MVT::v16i16, 12 },
     { ISD::CTTZ,       MVT::v32i8,   9 },
+    { ISD::SADDSAT,    MVT::v16i16,  1 },
+    { ISD::SADDSAT,    MVT::v32i8,   1 },
+    { ISD::SSUBSAT,    MVT::v16i16,  1 },
+    { ISD::SSUBSAT,    MVT::v32i8,   1 },
+    { ISD::UADDSAT,    MVT::v16i16,  1 },
+    { ISD::UADDSAT,    MVT::v32i8,   1 },
+    { ISD::USUBSAT,    MVT::v16i16,  1 },
+    { ISD::USUBSAT,    MVT::v32i8,   1 },
     { ISD::FSQRT,      MVT::f32,     7 }, // Haswell from http://www.agner.org/
     { ISD::FSQRT,      MVT::v4f32,   7 }, // Haswell from http://www.agner.org/
     { ISD::FSQRT,      MVT::v8f32,  14 }, // Haswell from http://www.agner.org/
@@ -1814,6 +1830,14 @@ int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
     { ISD::CTTZ,       MVT::v8i32,  30 }, // 2 x 128-bit Op + extract/insert
     { ISD::CTTZ,       MVT::v16i16, 26 }, // 2 x 128-bit Op + extract/insert
     { ISD::CTTZ,       MVT::v32i8,  20 }, // 2 x 128-bit Op + extract/insert
+    { ISD::SADDSAT,    MVT::v16i16,  4 }, // 2 x 128-bit Op + extract/insert
+    { ISD::SADDSAT,    MVT::v32i8,   4 }, // 2 x 128-bit Op + extract/insert
+    { ISD::SSUBSAT,    MVT::v16i16,  4 }, // 2 x 128-bit Op + extract/insert
+    { ISD::SSUBSAT,    MVT::v32i8,   4 }, // 2 x 128-bit Op + extract/insert
+    { ISD::UADDSAT,    MVT::v16i16,  4 }, // 2 x 128-bit Op + extract/insert
+    { ISD::UADDSAT,    MVT::v32i8,   4 }, // 2 x 128-bit Op + extract/insert
+    { ISD::USUBSAT,    MVT::v16i16,  4 }, // 2 x 128-bit Op + extract/insert
+    { ISD::USUBSAT,    MVT::v32i8,   4 }, // 2 x 128-bit Op + extract/insert
     { ISD::FSQRT,      MVT::f32,    14 }, // SNB from http://www.agner.org/
     { ISD::FSQRT,      MVT::v4f32,  14 }, // SNB from http://www.agner.org/
     { ISD::FSQRT,      MVT::v8f32,  28 }, // SNB from http://www.agner.org/
@@ -1878,6 +1902,14 @@ int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
     { ISD::CTTZ,       MVT::v4i32,  18 },
     { ISD::CTTZ,       MVT::v8i16,  16 },
     { ISD::CTTZ,       MVT::v16i8,  13 },
+    { ISD::SADDSAT,    MVT::v8i16,   1 },
+    { ISD::SADDSAT,    MVT::v16i8,   1 },
+    { ISD::SSUBSAT,    MVT::v8i16,   1 },
+    { ISD::SSUBSAT,    MVT::v16i8,   1 },
+    { ISD::UADDSAT,    MVT::v8i16,   1 },
+    { ISD::UADDSAT,    MVT::v16i8,   1 },
+    { ISD::USUBSAT,    MVT::v8i16,   1 },
+    { ISD::USUBSAT,    MVT::v16i8,   1 },
     { ISD::FSQRT,      MVT::f64,    32 }, // Nehalem from http://www.agner.org/
     { ISD::FSQRT,      MVT::v2f64,  32 }, // Nehalem from http://www.agner.org/
   };
@@ -1913,6 +1945,18 @@ int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
   case Intrinsic::cttz:
     ISD = ISD::CTTZ;
     break;
+  case Intrinsic::sadd_sat:
+    ISD = ISD::SADDSAT;
+    break;
+  case Intrinsic::ssub_sat:
+    ISD = ISD::SSUBSAT;
+    break;
+  case Intrinsic::uadd_sat:
+    ISD = ISD::UADDSAT;
+    break;
+  case Intrinsic::usub_sat:
+    ISD = ISD::USUBSAT;
+    break;
   case Intrinsic::sqrt:
     ISD = ISD::FSQRT;
     break;

diff --git a/test/Analysis/CostModel/X86/arith-ssat.ll b/test/Analysis/CostModel/X86/arith-ssat.ll
index 828e4bb7e68721b880658812d4dc7520d6e1b86c..457573be8ea85f1914ef56e93b6813dafe7fb6c9 100644 (file)
--- a/test/Analysis/CostModel/X86/arith-ssat.ll
+++ b/test/Analysis/CostModel/X86/arith-ssat.ll
@@ -35,24 +35,119 @@ declare <32 x i8>  @llvm.sadd.sat.v32i8(<32 x i8>, <32 x i8>)
 declare <64 x i8>  @llvm.sadd.sat.v64i8(<64 x i8>, <64 x i8>)
 
 define i32 @add(i32 %arg) {
-; CHECK-LABEL: 'add'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.sadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.sadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.sadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.sadd.sat.i32(i32 undef, i32 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.sadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+; SSE-LABEL: 'add'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.sadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.sadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.sadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.sadd.sat.i32(i32 undef, i32 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.sadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX1-LABEL: 'add'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.sadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.sadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.sadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.sadd.sat.i32(i32 undef, i32 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.sadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX2-LABEL: 'add'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.sadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.sadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.sadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.sadd.sat.i32(i32 undef, i32 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.sadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512F-LABEL: 'add'
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.sadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.sadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.sadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.sadd.sat.i32(i32 undef, i32 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.sadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512BW-LABEL: 'add'
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.sadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.sadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.sadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.sadd.sat.i32(i32 undef, i32 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.sadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512DQ-LABEL: 'add'
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.sadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.sadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.sadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.sadd.sat.i32(i32 undef, i32 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.sadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SLM-LABEL: 'add'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
@@ -64,13 +159,13 @@ define i32 @add(i32 %arg) {
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; GLM-LABEL: 'add'
@@ -83,13 +178,13 @@ define i32 @add(i32 %arg) {
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'add'
@@ -102,13 +197,13 @@ define i32 @add(i32 %arg) {
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.sadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.sadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.sadd.sat.i16(i16 undef, i16 undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call <16 x i16> @llvm.sadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call <32 x i16> @llvm.sadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.sadd.sat.i8(i8 undef, i8 undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.sadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call <32 x i8> @llvm.sadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call <64 x i8> @llvm.sadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
   %I64 = call i64 @llvm.sadd.sat.i64(i64 undef, i64 undef)
@@ -155,24 +250,119 @@ declare <32 x i8>  @llvm.ssub.sat.v32i8(<32 x i8>, <32 x i8>)
 declare <64 x i8>  @llvm.ssub.sat.v64i8(<64 x i8>, <64 x i8>)
 
 define i32 @sub(i32 %arg) {
-; CHECK-LABEL: 'sub'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.ssub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.ssub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.ssub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.ssub.sat.i32(i32 undef, i32 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+; SSE-LABEL: 'sub'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.ssub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.ssub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.ssub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.ssub.sat.i32(i32 undef, i32 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX1-LABEL: 'sub'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.ssub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.ssub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.ssub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.ssub.sat.i32(i32 undef, i32 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX2-LABEL: 'sub'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.ssub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.ssub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.ssub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.ssub.sat.i32(i32 undef, i32 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512F-LABEL: 'sub'
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.ssub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.ssub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.ssub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.ssub.sat.i32(i32 undef, i32 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512BW-LABEL: 'sub'
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.ssub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.ssub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.ssub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.ssub.sat.i32(i32 undef, i32 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512DQ-LABEL: 'sub'
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.ssub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.ssub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.ssub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.ssub.sat.i32(i32 undef, i32 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.ssub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SLM-LABEL: 'sub'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)
@@ -184,13 +374,13 @@ define i32 @sub(i32 %arg) {
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; GLM-LABEL: 'sub'
@@ -203,13 +393,13 @@ define i32 @sub(i32 %arg) {
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'sub'
@@ -222,13 +412,13 @@ define i32 @sub(i32 %arg) {
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.ssub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.ssub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.ssub.sat.i16(i16 undef, i16 undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.ssub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call <16 x i16> @llvm.ssub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call <32 x i16> @llvm.ssub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.ssub.sat.i8(i8 undef, i8 undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.ssub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call <32 x i8> @llvm.ssub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call <64 x i8> @llvm.ssub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
   %I64 = call i64 @llvm.ssub.sat.i64(i64 undef, i64 undef)

diff --git a/test/Analysis/CostModel/X86/arith-usat.ll b/test/Analysis/CostModel/X86/arith-usat.ll
index a9bbb095c13e0ece201196a7119306a15af75e8f..03bd5b09e2bebe669a6c95e8c8e2014b874a68ab 100644 (file)
--- a/test/Analysis/CostModel/X86/arith-usat.ll
+++ b/test/Analysis/CostModel/X86/arith-usat.ll
@@ -35,24 +35,119 @@ declare <32 x i8>  @llvm.uadd.sat.v32i8(<32 x i8>, <32 x i8>)
 declare <64 x i8>  @llvm.uadd.sat.v64i8(<64 x i8>, <64 x i8>)
 
 define i32 @add(i32 %arg) {
-; CHECK-LABEL: 'add'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.uadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.uadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.uadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.uadd.sat.i32(i32 undef, i32 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+; SSE-LABEL: 'add'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.uadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.uadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.uadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.uadd.sat.i32(i32 undef, i32 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX1-LABEL: 'add'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.uadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.uadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.uadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.uadd.sat.i32(i32 undef, i32 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX2-LABEL: 'add'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.uadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.uadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.uadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.uadd.sat.i32(i32 undef, i32 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512F-LABEL: 'add'
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.uadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.uadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.uadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.uadd.sat.i32(i32 undef, i32 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512BW-LABEL: 'add'
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.uadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.uadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.uadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.uadd.sat.i32(i32 undef, i32 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512DQ-LABEL: 'add'
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.uadd.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.uadd.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.uadd.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.uadd.sat.i32(i32 undef, i32 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.uadd.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SLM-LABEL: 'add'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
@@ -64,13 +159,13 @@ define i32 @add(i32 %arg) {
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; GLM-LABEL: 'add'
@@ -83,13 +178,13 @@ define i32 @add(i32 %arg) {
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'add'
@@ -102,13 +197,13 @@ define i32 @add(i32 %arg) {
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.uadd.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.uadd.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.uadd.sat.i16(i16 undef, i16 undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.uadd.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call <16 x i16> @llvm.uadd.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call <32 x i16> @llvm.uadd.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.uadd.sat.i8(i8 undef, i8 undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.uadd.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call <32 x i8> @llvm.uadd.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call <64 x i8> @llvm.uadd.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
   %I64 = call i64 @llvm.uadd.sat.i64(i64 undef, i64 undef)
@@ -155,24 +250,119 @@ declare <32 x i8>  @llvm.usub.sat.v32i8(<32 x i8>, <32 x i8>)
 declare <64 x i8>  @llvm.usub.sat.v64i8(<64 x i8>, <64 x i8>)
 
 define i32 @sub(i32 %arg) {
-; CHECK-LABEL: 'sub'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.usub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.usub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.usub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.usub.sat.i32(i32 undef, i32 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+; SSE-LABEL: 'sub'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.usub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.usub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.usub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.usub.sat.i32(i32 undef, i32 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX1-LABEL: 'sub'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.usub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.usub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.usub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.usub.sat.i32(i32 undef, i32 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX2-LABEL: 'sub'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.usub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.usub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.usub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.usub.sat.i32(i32 undef, i32 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512F-LABEL: 'sub'
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.usub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.usub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.usub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.usub.sat.i32(i32 undef, i32 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512F-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512BW-LABEL: 'sub'
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.usub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.usub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.usub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.usub.sat.i32(i32 undef, i32 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512BW-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
+;
+; AVX512DQ-LABEL: 'sub'
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2I64 = call <2 x i64> @llvm.usub.sat.v2i64(<2 x i64> undef, <2 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I64 = call <4 x i64> @llvm.usub.sat.v4i64(<4 x i64> undef, <4 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call <8 x i64> @llvm.usub.sat.v8i64(<8 x i64> undef, <8 x i64> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I32 = call i32 @llvm.usub.sat.i32(i32 undef, i32 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4I32 = call <4 x i32> @llvm.usub.sat.v4i32(<4 x i32> undef, <4 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; AVX512DQ-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; SLM-LABEL: 'sub'
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)
@@ -184,13 +374,13 @@ define i32 @sub(i32 %arg) {
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; SLM-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; SLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; SLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; GLM-LABEL: 'sub'
@@ -203,13 +393,13 @@ define i32 @sub(i32 %arg) {
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; GLM-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; GLM-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; GLM-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
 ; BTVER2-LABEL: 'sub'
@@ -222,13 +412,13 @@ define i32 @sub(i32 %arg) {
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I32 = call <8 x i32> @llvm.usub.sat.v8i32(<8 x i32> undef, <8 x i32> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I32 = call <16 x i32> @llvm.usub.sat.v16i32(<16 x i32> undef, <16 x i32> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I16 = call i16 @llvm.usub.sat.i16(i16 undef, i16 undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V8I16 = call <8 x i16> @llvm.usub.sat.v8i16(<8 x i16> undef, <8 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call <16 x i16> @llvm.usub.sat.v16i16(<16 x i16> undef, <16 x i16> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call <32 x i16> @llvm.usub.sat.v32i16(<32 x i16> undef, <32 x i16> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %I8 = call i8 @llvm.usub.sat.i8(i8 undef, i8 undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
-; BTVER2-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %V16I8 = call <16 x i8> @llvm.usub.sat.v16i8(<16 x i8> undef, <16 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call <32 x i8> @llvm.usub.sat.v32i8(<32 x i8> undef, <32 x i8> undef)
+; BTVER2-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call <64 x i8> @llvm.usub.sat.v64i8(<64 x i8> undef, <64 x i8> undef)
 ; BTVER2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
   %I64 = call i64 @llvm.usub.sat.i64(i64 undef, i64 undef)