test/Analysis/CostModel/ARM/arith-overflow.ll

; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
; RUN: opt -cost-model -analyze -mtriple=thumbv8m.main < %s | FileCheck %s --check-prefix=V8M-RECIP
; RUN: opt -cost-model -analyze -mtriple=armv8a-linux-gnueabihf < %s | FileCheck %s --check-prefix=NEON-RECIP
; RUN: opt -cost-model -analyze -mtriple=armv8.1m.main -mattr=+mve < %s | FileCheck %s --check-prefix=MVE-RECIP
; RUN: opt -cost-model -analyze -cost-kind=code-size -mtriple=thumbv8m.main < %s | FileCheck %s --check-prefix=V8M-SIZE
; RUN: opt -cost-model -analyze -cost-kind=code-size -mtriple=armv8a-linux-gnueabihf < %s | FileCheck %s --check-prefix=NEON-SIZE
; RUN: opt -cost-model -analyze -cost-kind=code-size -mtriple=armv8.1m.main -mattr=+mve < %s | FileCheck %s --check-prefix=MVE-SIZE

declare {i64, i1}              @llvm.sadd.with.overflow.i64(i64, i64)
declare {<2 x i64>, <2 x i1>}  @llvm.sadd.with.overflow.v2i64(<2 x i64>, <2 x i64>)
declare {<4 x i64>, <4 x i1>}  @llvm.sadd.with.overflow.v4i64(<4 x i64>, <4 x i64>)
declare {<8 x i64>, <8 x i1>}  @llvm.sadd.with.overflow.v8i64(<8 x i64>, <8 x i64>)

declare {i32, i1}               @llvm.sadd.with.overflow.i32(i32, i32)
declare {<4 x i32>, <4 x i1>}   @llvm.sadd.with.overflow.v4i32(<4 x i32>, <4 x i32>)
declare {<8 x i32>, <8 x i1>}   @llvm.sadd.with.overflow.v8i32(<8 x i32>, <8 x i32>)
declare {<16 x i32>, <16 x i1>} @llvm.sadd.with.overflow.v16i32(<16 x i32>, <16 x i32>)

declare {i16, i1}               @llvm.sadd.with.overflow.i16(i16, i16)
declare {<8 x i16>,  <8 x i1>}  @llvm.sadd.with.overflow.v8i16(<8 x i16>, <8 x i16>)
declare {<16 x i16>, <16 x i1>} @llvm.sadd.with.overflow.v16i16(<16 x i16>, <16 x i16>)
declare {<32 x i16>, <32 x i1>} @llvm.sadd.with.overflow.v32i16(<32 x i16>, <32 x i16>)

declare {i8, i1}                @llvm.sadd.with.overflow.i8(i8, i8)
declare {<16 x i8>, <16 x i1>}  @llvm.sadd.with.overflow.v16i8(<16 x i8>, <16 x i8>)
declare {<32 x i8>, <32 x i1>}  @llvm.sadd.with.overflow.v32i8(<32 x i8>, <32 x i8>)
declare {<64 x i8>, <64 x i1>}  @llvm.sadd.with.overflow.v64i8(<64 x i8>, <64 x i8>)

define i32 @sadd(i32 %arg) {
; V8M-RECIP-LABEL: 'sadd'
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %I64 = call { i64, i1 } @llvm.sadd.with.overflow.i64(i64 undef, i64 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.sadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.sadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.sadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 undef, i32 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.sadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.sadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.sadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I16 = call { i16, i1 } @llvm.sadd.with.overflow.i16(i16 undef, i16 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.sadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.sadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.sadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I8 = call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 undef, i8 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.sadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.sadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 768 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.sadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-RECIP-LABEL: 'sadd'
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %I64 = call { i64, i1 } @llvm.sadd.with.overflow.i64(i64 undef, i64 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.sadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.sadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.sadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 undef, i32 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.sadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.sadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.sadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I16 = call { i16, i1 } @llvm.sadd.with.overflow.i16(i16 undef, i16 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.sadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.sadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.sadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I8 = call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 undef, i8 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.sadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.sadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.sadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; MVE-RECIP-LABEL: 'sadd'
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %I64 = call { i64, i1 } @llvm.sadd.with.overflow.i64(i64 undef, i64 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 74 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.sadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 242 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.sadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 866 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.sadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 undef, i32 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 90 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.sadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 498 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.sadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1890 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.sadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I16 = call { i16, i1 } @llvm.sadd.with.overflow.i16(i16 undef, i16 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 298 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.sadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1874 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.sadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7332 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.sadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I8 = call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 undef, i8 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1098 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.sadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7316 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.sadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 28968 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.sadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; V8M-SIZE-LABEL: 'sadd'
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.sadd.with.overflow.i64(i64 undef, i64 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.sadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.sadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.sadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I32 = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 undef, i32 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.sadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.sadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.sadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I16 = call { i16, i1 } @llvm.sadd.with.overflow.i16(i16 undef, i16 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.sadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.sadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 69 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.sadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I8 = call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 undef, i8 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.sadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 69 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.sadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.sadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-SIZE-LABEL: 'sadd'
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I64 = call { i64, i1 } @llvm.sadd.with.overflow.i64(i64 undef, i64 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.sadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.sadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.sadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 undef, i32 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.sadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.sadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.sadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I16 = call { i16, i1 } @llvm.sadd.with.overflow.i16(i16 undef, i16 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.sadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.sadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.sadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I8 = call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 undef, i8 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.sadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.sadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.sadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; MVE-SIZE-LABEL: 'sadd'
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.sadd.with.overflow.i64(i64 undef, i64 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.sadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.sadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 150 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.sadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I32 = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 undef, i32 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.sadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 203 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.sadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 787 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.sadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I16 = call { i16, i1 } @llvm.sadd.with.overflow.i16(i16 undef, i16 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.sadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 779 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.sadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 3092 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.sadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I8 = call { i8, i1 } @llvm.sadd.with.overflow.i8(i8 undef, i8 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.sadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 3084 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.sadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 12310 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.sadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
  %I64 = call {i64, i1} @llvm.sadd.with.overflow.i64(i64 undef, i64 undef)
  %V2I64 = call {<2 x i64>, <2 x i1>} @llvm.sadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
  %V4I64 = call {<4 x i64>, <4 x i1>} @llvm.sadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
  %V8I64 = call {<8 x i64>, <8 x i1>} @llvm.sadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)

  %I32 = call {i32, i1} @llvm.sadd.with.overflow.i32(i32 undef, i32 undef)
  %V4I32  = call {<4 x i32>, <4 x i1>}  @llvm.sadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
  %V8I32  = call {<8 x i32>, <8 x i1>}  @llvm.sadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
  %V16I32 = call {<16 x i32>, <16 x i1>} @llvm.sadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)

  %I16 = call {i16, i1} @llvm.sadd.with.overflow.i16(i16 undef, i16 undef)
  %V8I16  = call {<8 x i16>, <8 x i1>}  @llvm.sadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
  %V16I16 = call {<16 x i16>, <16 x i1>} @llvm.sadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
  %V32I16 = call {<32 x i16>, <32 x i1>} @llvm.sadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)

  %I8 = call {i8, i1} @llvm.sadd.with.overflow.i8(i8 undef, i8 undef)
  %V16I8 = call {<16 x i8>, <16 x i1>} @llvm.sadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
  %V32I8 = call {<32 x i8>, <32 x i1>} @llvm.sadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
  %V64I8 = call {<64 x i8>, <64 x i1>} @llvm.sadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)

  ret i32 undef
}

declare {i64, i1}              @llvm.uadd.with.overflow.i64(i64, i64)
declare {<2 x i64>, <2 x i1>}  @llvm.uadd.with.overflow.v2i64(<2 x i64>, <2 x i64>)
declare {<4 x i64>, <4 x i1>}  @llvm.uadd.with.overflow.v4i64(<4 x i64>, <4 x i64>)
declare {<8 x i64>, <8 x i1>}  @llvm.uadd.with.overflow.v8i64(<8 x i64>, <8 x i64>)

declare {i32, i1}               @llvm.uadd.with.overflow.i32(i32, i32)
declare {<4 x i32>, <4 x i1>}   @llvm.uadd.with.overflow.v4i32(<4 x i32>, <4 x i32>)
declare {<8 x i32>, <8 x i1>}   @llvm.uadd.with.overflow.v8i32(<8 x i32>, <8 x i32>)
declare {<16 x i32>, <16 x i1>} @llvm.uadd.with.overflow.v16i32(<16 x i32>, <16 x i32>)

declare {i16, i1}               @llvm.uadd.with.overflow.i16(i16, i16)
declare {<8 x i16>,  <8 x i1>}  @llvm.uadd.with.overflow.v8i16(<8 x i16>, <8 x i16>)
declare {<16 x i16>, <16 x i1>} @llvm.uadd.with.overflow.v16i16(<16 x i16>, <16 x i16>)
declare {<32 x i16>, <32 x i1>} @llvm.uadd.with.overflow.v32i16(<32 x i16>, <32 x i16>)

declare {i8, i1}                @llvm.uadd.with.overflow.i8(i8, i8)
declare {<16 x i8>, <16 x i1>}  @llvm.uadd.with.overflow.v16i8(<16 x i8>, <16 x i8>)
declare {<32 x i8>, <32 x i1>}  @llvm.uadd.with.overflow.v32i8(<32 x i8>, <32 x i8>)
declare {<64 x i8>, <64 x i1>}  @llvm.uadd.with.overflow.v64i8(<64 x i8>, <64 x i8>)

define i32 @uadd(i32 %arg) {
; V8M-RECIP-LABEL: 'uadd'
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.uadd.with.overflow.i64(i64 undef, i64 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.uadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.uadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.uadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 undef, i32 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.uadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.uadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.uadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.uadd.with.overflow.i16(i16 undef, i16 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.uadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.uadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.uadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.uadd.with.overflow.i8(i8 undef, i8 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.uadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.uadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.uadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-RECIP-LABEL: 'uadd'
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.uadd.with.overflow.i64(i64 undef, i64 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.uadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.uadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.uadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 undef, i32 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.uadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.uadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.uadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.uadd.with.overflow.i16(i16 undef, i16 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.uadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.uadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.uadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.uadd.with.overflow.i8(i8 undef, i8 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.uadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.uadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.uadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; MVE-RECIP-LABEL: 'uadd'
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.uadd.with.overflow.i64(i64 undef, i64 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.uadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.uadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 288 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.uadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 undef, i32 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.uadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.uadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 272 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.uadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.uadd.with.overflow.i16(i16 undef, i16 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.uadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 264 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.uadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1040 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.uadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.uadd.with.overflow.i8(i8 undef, i8 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.uadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1032 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.uadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4112 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.uadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; V8M-SIZE-LABEL: 'uadd'
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.uadd.with.overflow.i64(i64 undef, i64 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.uadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.uadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.uadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 undef, i32 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.uadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.uadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.uadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.uadd.with.overflow.i16(i16 undef, i16 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.uadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.uadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.uadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.uadd.with.overflow.i8(i8 undef, i8 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.uadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.uadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.uadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-SIZE-LABEL: 'uadd'
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I64 = call { i64, i1 } @llvm.uadd.with.overflow.i64(i64 undef, i64 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.uadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.uadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.uadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 undef, i32 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.uadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.uadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.uadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.uadd.with.overflow.i16(i16 undef, i16 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.uadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.uadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.uadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.uadd.with.overflow.i8(i8 undef, i8 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.uadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.uadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.uadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; MVE-SIZE-LABEL: 'uadd'
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.uadd.with.overflow.i64(i64 undef, i64 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.uadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 41 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.uadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 145 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.uadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.uadd.with.overflow.i32(i32 undef, i32 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.uadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.uadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 264 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.uadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.uadd.with.overflow.i16(i16 undef, i16 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.uadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 260 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.uadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1032 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.uadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.uadd.with.overflow.i8(i8 undef, i8 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.uadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1028 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.uadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 4104 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.uadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
  %I64 = call {i64, i1} @llvm.uadd.with.overflow.i64(i64 undef, i64 undef)
  %V2I64 = call {<2 x i64>, <2 x i1>} @llvm.uadd.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
  %V4I64 = call {<4 x i64>, <4 x i1>} @llvm.uadd.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
  %V8I64 = call {<8 x i64>, <8 x i1>} @llvm.uadd.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)

  %I32 = call {i32, i1} @llvm.uadd.with.overflow.i32(i32 undef, i32 undef)
  %V4I32  = call {<4 x i32>, <4 x i1>}  @llvm.uadd.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
  %V8I32  = call {<8 x i32>, <8 x i1>}  @llvm.uadd.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
  %V16I32 = call {<16 x i32>, <16 x i1>} @llvm.uadd.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)

  %I16 = call {i16, i1} @llvm.uadd.with.overflow.i16(i16 undef, i16 undef)
  %V8I16  = call {<8 x i16>, <8 x i1>}  @llvm.uadd.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
  %V16I16 = call {<16 x i16>, <16 x i1>} @llvm.uadd.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
  %V32I16 = call {<32 x i16>, <32 x i1>} @llvm.uadd.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)

  %I8 = call {i8, i1} @llvm.uadd.with.overflow.i8(i8 undef, i8 undef)
  %V16I8 = call {<16 x i8>, <16 x i1>} @llvm.uadd.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
  %V32I8 = call {<32 x i8>, <32 x i1>} @llvm.uadd.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
  %V64I8 = call {<64 x i8>, <64 x i1>} @llvm.uadd.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)

  ret i32 undef
}

declare {i64, i1}              @llvm.ssub.with.overflow.i64(i64, i64)
declare {<2 x i64>, <2 x i1>}  @llvm.ssub.with.overflow.v2i64(<2 x i64>, <2 x i64>)
declare {<4 x i64>, <4 x i1>}  @llvm.ssub.with.overflow.v4i64(<4 x i64>, <4 x i64>)
declare {<8 x i64>, <8 x i1>}  @llvm.ssub.with.overflow.v8i64(<8 x i64>, <8 x i64>)

declare {i32, i1}               @llvm.ssub.with.overflow.i32(i32, i32)
declare {<4 x i32>, <4 x i1>}   @llvm.ssub.with.overflow.v4i32(<4 x i32>, <4 x i32>)
declare {<8 x i32>, <8 x i1>}   @llvm.ssub.with.overflow.v8i32(<8 x i32>, <8 x i32>)
declare {<16 x i32>, <16 x i1>} @llvm.ssub.with.overflow.v16i32(<16 x i32>, <16 x i32>)

declare {i16, i1}               @llvm.ssub.with.overflow.i16(i16, i16)
declare {<8 x i16>,  <8 x i1>}  @llvm.ssub.with.overflow.v8i16(<8 x i16>, <8 x i16>)
declare {<16 x i16>, <16 x i1>} @llvm.ssub.with.overflow.v16i16(<16 x i16>, <16 x i16>)
declare {<32 x i16>, <32 x i1>} @llvm.ssub.with.overflow.v32i16(<32 x i16>, <32 x i16>)

declare {i8, i1}                @llvm.ssub.with.overflow.i8(i8, i8)
declare {<16 x i8>, <16 x i1>}  @llvm.ssub.with.overflow.v16i8(<16 x i8>, <16 x i8>)
declare {<32 x i8>, <32 x i1>}  @llvm.ssub.with.overflow.v32i8(<32 x i8>, <32 x i8>)
declare {<64 x i8>, <64 x i1>}  @llvm.ssub.with.overflow.v64i8(<64 x i8>, <64 x i8>)

define i32 @ssub(i32 %arg) {
; V8M-RECIP-LABEL: 'ssub'
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %I64 = call { i64, i1 } @llvm.ssub.with.overflow.i64(i64 undef, i64 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.ssub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.ssub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.ssub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 undef, i32 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.ssub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.ssub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.ssub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I16 = call { i16, i1 } @llvm.ssub.with.overflow.i16(i16 undef, i16 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.ssub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.ssub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.ssub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I8 = call { i8, i1 } @llvm.ssub.with.overflow.i8(i8 undef, i8 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.ssub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.ssub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 768 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.ssub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-RECIP-LABEL: 'ssub'
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %I64 = call { i64, i1 } @llvm.ssub.with.overflow.i64(i64 undef, i64 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.ssub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.ssub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.ssub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 undef, i32 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.ssub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.ssub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.ssub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I16 = call { i16, i1 } @llvm.ssub.with.overflow.i16(i16 undef, i16 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.ssub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.ssub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.ssub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I8 = call { i8, i1 } @llvm.ssub.with.overflow.i8(i8 undef, i8 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.ssub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.ssub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.ssub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; MVE-RECIP-LABEL: 'ssub'
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %I64 = call { i64, i1 } @llvm.ssub.with.overflow.i64(i64 undef, i64 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 74 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.ssub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 242 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.ssub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 866 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.ssub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 undef, i32 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 90 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.ssub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 498 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.ssub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1890 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.ssub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I16 = call { i16, i1 } @llvm.ssub.with.overflow.i16(i16 undef, i16 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 298 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.ssub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1874 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.ssub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7332 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.ssub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I8 = call { i8, i1 } @llvm.ssub.with.overflow.i8(i8 undef, i8 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1098 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.ssub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7316 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.ssub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 28968 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.ssub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; V8M-SIZE-LABEL: 'ssub'
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.ssub.with.overflow.i64(i64 undef, i64 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.ssub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.ssub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.ssub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I32 = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 undef, i32 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.ssub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.ssub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.ssub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I16 = call { i16, i1 } @llvm.ssub.with.overflow.i16(i16 undef, i16 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.ssub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.ssub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 69 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.ssub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I8 = call { i8, i1 } @llvm.ssub.with.overflow.i8(i8 undef, i8 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.ssub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 69 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.ssub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 133 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.ssub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-SIZE-LABEL: 'ssub'
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I64 = call { i64, i1 } @llvm.ssub.with.overflow.i64(i64 undef, i64 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.ssub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.ssub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.ssub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 undef, i32 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.ssub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.ssub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.ssub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I16 = call { i16, i1 } @llvm.ssub.with.overflow.i16(i16 undef, i16 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.ssub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.ssub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.ssub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I8 = call { i8, i1 } @llvm.ssub.with.overflow.i8(i8 undef, i8 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.ssub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.ssub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.ssub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; MVE-SIZE-LABEL: 'ssub'
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.ssub.with.overflow.i64(i64 undef, i64 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.ssub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.ssub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 150 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.ssub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I32 = call { i32, i1 } @llvm.ssub.with.overflow.i32(i32 undef, i32 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.ssub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 203 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.ssub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 787 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.ssub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I16 = call { i16, i1 } @llvm.ssub.with.overflow.i16(i16 undef, i16 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.ssub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 779 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.ssub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 3092 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.ssub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %I8 = call { i8, i1 } @llvm.ssub.with.overflow.i8(i8 undef, i8 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.ssub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 3084 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.ssub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 12310 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.ssub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
  %I64 = call {i64, i1} @llvm.ssub.with.overflow.i64(i64 undef, i64 undef)
  %V2I64 = call {<2 x i64>, <2 x i1>} @llvm.ssub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
  %V4I64 = call {<4 x i64>, <4 x i1>} @llvm.ssub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
  %V8I64 = call {<8 x i64>, <8 x i1>} @llvm.ssub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)

  %I32 = call {i32, i1} @llvm.ssub.with.overflow.i32(i32 undef, i32 undef)
  %V4I32  = call {<4 x i32>, <4 x i1>}  @llvm.ssub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
  %V8I32  = call {<8 x i32>, <8 x i1>}  @llvm.ssub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
  %V16I32 = call {<16 x i32>, <16 x i1>} @llvm.ssub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)

  %I16 = call {i16, i1} @llvm.ssub.with.overflow.i16(i16 undef, i16 undef)
  %V8I16  = call {<8 x i16>, <8 x i1>}  @llvm.ssub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
  %V16I16 = call {<16 x i16>, <16 x i1>} @llvm.ssub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
  %V32I16 = call {<32 x i16>, <32 x i1>} @llvm.ssub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)

  %I8 = call {i8, i1} @llvm.ssub.with.overflow.i8(i8 undef, i8 undef)
  %V16I8 = call {<16 x i8>, <16 x i1>} @llvm.ssub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
  %V32I8 = call {<32 x i8>, <32 x i1>} @llvm.ssub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
  %V64I8 = call {<64 x i8>, <64 x i1>} @llvm.ssub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)

  ret i32 undef
}

declare {i64, i1}              @llvm.usub.with.overflow.i64(i64, i64)
declare {<2 x i64>, <2 x i1>}  @llvm.usub.with.overflow.v2i64(<2 x i64>, <2 x i64>)
declare {<4 x i64>, <4 x i1>}  @llvm.usub.with.overflow.v4i64(<4 x i64>, <4 x i64>)
declare {<8 x i64>, <8 x i1>}  @llvm.usub.with.overflow.v8i64(<8 x i64>, <8 x i64>)

declare {i32, i1}               @llvm.usub.with.overflow.i32(i32, i32)
declare {<4 x i32>, <4 x i1>}   @llvm.usub.with.overflow.v4i32(<4 x i32>, <4 x i32>)
declare {<8 x i32>, <8 x i1>}   @llvm.usub.with.overflow.v8i32(<8 x i32>, <8 x i32>)
declare {<16 x i32>, <16 x i1>} @llvm.usub.with.overflow.v16i32(<16 x i32>, <16 x i32>)

declare {i16, i1}               @llvm.usub.with.overflow.i16(i16, i16)
declare {<8 x i16>,  <8 x i1>}  @llvm.usub.with.overflow.v8i16(<8 x i16>, <8 x i16>)
declare {<16 x i16>, <16 x i1>} @llvm.usub.with.overflow.v16i16(<16 x i16>, <16 x i16>)
declare {<32 x i16>, <32 x i1>} @llvm.usub.with.overflow.v32i16(<32 x i16>, <32 x i16>)

declare {i8, i1}                @llvm.usub.with.overflow.i8(i8, i8)
declare {<16 x i8>, <16 x i1>}  @llvm.usub.with.overflow.v16i8(<16 x i8>, <16 x i8>)
declare {<32 x i8>, <32 x i1>}  @llvm.usub.with.overflow.v32i8(<32 x i8>, <32 x i8>)
declare {<64 x i8>, <64 x i1>}  @llvm.usub.with.overflow.v64i8(<64 x i8>, <64 x i8>)

define i32 @usub(i32 %arg) {
; V8M-RECIP-LABEL: 'usub'
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.usub.with.overflow.i64(i64 undef, i64 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 10 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.usub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 20 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.usub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 40 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.usub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 undef, i32 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.usub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.usub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.usub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.usub.with.overflow.i16(i16 undef, i16 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.usub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.usub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.usub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.usub.with.overflow.i8(i8 undef, i8 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.usub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.usub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.usub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-RECIP-LABEL: 'usub'
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.usub.with.overflow.i64(i64 undef, i64 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.usub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.usub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.usub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 undef, i32 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.usub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.usub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.usub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.usub.with.overflow.i16(i16 undef, i16 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.usub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.usub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.usub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.usub.with.overflow.i8(i8 undef, i8 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.usub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.usub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.usub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; MVE-RECIP-LABEL: 'usub'
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.usub.with.overflow.i64(i64 undef, i64 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.usub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.usub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 288 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.usub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 undef, i32 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.usub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 72 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.usub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 272 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.usub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.usub.with.overflow.i16(i16 undef, i16 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.usub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 264 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.usub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1040 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.usub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.usub.with.overflow.i8(i8 undef, i8 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.usub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1032 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.usub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4112 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.usub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; V8M-SIZE-LABEL: 'usub'
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.usub.with.overflow.i64(i64 undef, i64 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.usub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.usub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.usub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 undef, i32 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.usub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.usub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.usub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.usub.with.overflow.i16(i16 undef, i16 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.usub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.usub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.usub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.usub.with.overflow.i8(i8 undef, i8 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.usub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 33 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.usub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 65 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.usub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-SIZE-LABEL: 'usub'
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I64 = call { i64, i1 } @llvm.usub.with.overflow.i64(i64 undef, i64 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.usub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.usub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.usub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 undef, i32 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.usub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.usub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.usub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.usub.with.overflow.i16(i16 undef, i16 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.usub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.usub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.usub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.usub.with.overflow.i8(i8 undef, i8 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.usub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.usub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.usub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; MVE-SIZE-LABEL: 'usub'
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %I64 = call { i64, i1 } @llvm.usub.with.overflow.i64(i64 undef, i64 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.usub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 41 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.usub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 145 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.usub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I32 = call { i32, i1 } @llvm.usub.with.overflow.i32(i32 undef, i32 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.usub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 68 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.usub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 264 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.usub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I16 = call { i16, i1 } @llvm.usub.with.overflow.i16(i16 undef, i16 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.usub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 260 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.usub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1032 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.usub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %I8 = call { i8, i1 } @llvm.usub.with.overflow.i8(i8 undef, i8 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.usub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1028 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.usub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 4104 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.usub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
  %I64 = call {i64, i1} @llvm.usub.with.overflow.i64(i64 undef, i64 undef)
  %V2I64 = call {<2 x i64>, <2 x i1>} @llvm.usub.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
  %V4I64 = call {<4 x i64>, <4 x i1>} @llvm.usub.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
  %V8I64 = call {<8 x i64>, <8 x i1>} @llvm.usub.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)

  %I32 = call {i32, i1} @llvm.usub.with.overflow.i32(i32 undef, i32 undef)
  %V4I32  = call {<4 x i32>, <4 x i1>}  @llvm.usub.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
  %V8I32  = call {<8 x i32>, <8 x i1>}  @llvm.usub.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
  %V16I32 = call {<16 x i32>, <16 x i1>} @llvm.usub.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)

  %I16 = call {i16, i1} @llvm.usub.with.overflow.i16(i16 undef, i16 undef)
  %V8I16  = call {<8 x i16>, <8 x i1>}  @llvm.usub.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
  %V16I16 = call {<16 x i16>, <16 x i1>} @llvm.usub.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
  %V32I16 = call {<32 x i16>, <32 x i1>} @llvm.usub.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)

  %I8 = call {i8, i1} @llvm.usub.with.overflow.i8(i8 undef, i8 undef)
  %V16I8 = call {<16 x i8>, <16 x i1>} @llvm.usub.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
  %V32I8 = call {<32 x i8>, <32 x i1>} @llvm.usub.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
  %V64I8 = call {<64 x i8>, <64 x i1>} @llvm.usub.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)

  ret i32 undef
}

declare {i64, i1}              @llvm.smul.with.overflow.i64(i64, i64)
declare {<2 x i64>, <2 x i1>}  @llvm.smul.with.overflow.v2i64(<2 x i64>, <2 x i64>)
declare {<4 x i64>, <4 x i1>}  @llvm.smul.with.overflow.v4i64(<4 x i64>, <4 x i64>)
declare {<8 x i64>, <8 x i1>}  @llvm.smul.with.overflow.v8i64(<8 x i64>, <8 x i64>)

declare {i32, i1}               @llvm.smul.with.overflow.i32(i32, i32)
declare {<4 x i32>, <4 x i1>}   @llvm.smul.with.overflow.v4i32(<4 x i32>, <4 x i32>)
declare {<8 x i32>, <8 x i1>}   @llvm.smul.with.overflow.v8i32(<8 x i32>, <8 x i32>)
declare {<16 x i32>, <16 x i1>} @llvm.smul.with.overflow.v16i32(<16 x i32>, <16 x i32>)

declare {i16, i1}               @llvm.smul.with.overflow.i16(i16, i16)
declare {<8 x i16>,  <8 x i1>}  @llvm.smul.with.overflow.v8i16(<8 x i16>, <8 x i16>)
declare {<16 x i16>, <16 x i1>} @llvm.smul.with.overflow.v16i16(<16 x i16>, <16 x i16>)
declare {<32 x i16>, <32 x i1>} @llvm.smul.with.overflow.v32i16(<32 x i16>, <32 x i16>)

declare {i8, i1}                @llvm.smul.with.overflow.i8(i8, i8)
declare {<16 x i8>, <16 x i1>}  @llvm.smul.with.overflow.v16i8(<16 x i8>, <16 x i8>)
declare {<32 x i8>, <32 x i1>}  @llvm.smul.with.overflow.v32i8(<32 x i8>, <32 x i8>)
declare {<64 x i8>, <64 x i1>}  @llvm.smul.with.overflow.v64i8(<64 x i8>, <64 x i8>)

define i32 @smul(i32 %arg) {
; V8M-RECIP-LABEL: 'smul'
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.smul.with.overflow.i64(i64 undef, i64 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 78 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.smul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 156 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.smul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 312 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.smul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.smul.with.overflow.i32(i32 undef, i32 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 80 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.smul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 160 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.smul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 320 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 112 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 224 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 112 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 224 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 448 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.smul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-RECIP-LABEL: 'smul'
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.smul.with.overflow.i64(i64 undef, i64 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 81 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.smul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 154 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.smul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 300 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.smul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.smul.with.overflow.i32(i32 undef, i32 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.smul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 30 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.smul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 60 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 42 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 84 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 42 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 84 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.smul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; MVE-RECIP-LABEL: 'smul'
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.smul.with.overflow.i64(i64 undef, i64 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 120 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.smul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 288 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.smul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 768 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.smul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.smul.with.overflow.i32(i32 undef, i32 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 86 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.smul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 380 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.smul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1464 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 62 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 420 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1512 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 94 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1252 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4712 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.smul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; V8M-SIZE-LABEL: 'smul'
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.smul.with.overflow.i64(i64 undef, i64 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.smul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.smul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 69 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.smul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.smul.with.overflow.i32(i32 undef, i32 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 21 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.smul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.smul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 69 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 51 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 99 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 195 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.smul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-SIZE-LABEL: 'smul'
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %I64 = call { i64, i1 } @llvm.smul.with.overflow.i64(i64 undef, i64 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.smul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.smul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.smul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I32 = call { i32, i1 } @llvm.smul.with.overflow.i32(i32 undef, i32 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.smul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.smul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.smul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; MVE-SIZE-LABEL: 'smul'
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %I64 = call { i64, i1 } @llvm.smul.with.overflow.i64(i64 undef, i64 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 37 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.smul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 101 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.smul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 325 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.smul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I32 = call { i32, i1 } @llvm.smul.with.overflow.i32(i32 undef, i32 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 47 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.smul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 218 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.smul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 816 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I16 = call { i16, i1 } @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 39 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 332 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1174 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I8 = call { i8, i1 } @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 71 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1164 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 4374 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.smul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
  %I64 = call {i64, i1} @llvm.smul.with.overflow.i64(i64 undef, i64 undef)
  %V2I64 = call {<2 x i64>, <2 x i1>} @llvm.smul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
  %V4I64 = call {<4 x i64>, <4 x i1>} @llvm.smul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
  %V8I64 = call {<8 x i64>, <8 x i1>} @llvm.smul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)

  %I32 = call {i32, i1} @llvm.smul.with.overflow.i32(i32 undef, i32 undef)
  %V4I32  = call {<4 x i32>, <4 x i1>}  @llvm.smul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
  %V8I32  = call {<8 x i32>, <8 x i1>}  @llvm.smul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
  %V16I32 = call {<16 x i32>, <16 x i1>} @llvm.smul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)

  %I16 = call {i16, i1} @llvm.smul.with.overflow.i16(i16 undef, i16 undef)
  %V8I16  = call {<8 x i16>, <8 x i1>}  @llvm.smul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
  %V16I16 = call {<16 x i16>, <16 x i1>} @llvm.smul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
  %V32I16 = call {<32 x i16>, <32 x i1>} @llvm.smul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)

  %I8 = call {i8, i1} @llvm.smul.with.overflow.i8(i8 undef, i8 undef)
  %V16I8 = call {<16 x i8>, <16 x i1>} @llvm.smul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
  %V32I8 = call {<32 x i8>, <32 x i1>} @llvm.smul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
  %V64I8 = call {<64 x i8>, <64 x i1>} @llvm.smul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)

  ret i32 undef
}

declare {i64, i1}              @llvm.umul.with.overflow.i64(i64, i64)
declare {<2 x i64>, <2 x i1>}  @llvm.umul.with.overflow.v2i64(<2 x i64>, <2 x i64>)
declare {<4 x i64>, <4 x i1>}  @llvm.umul.with.overflow.v4i64(<4 x i64>, <4 x i64>)
declare {<8 x i64>, <8 x i1>}  @llvm.umul.with.overflow.v8i64(<8 x i64>, <8 x i64>)

declare {i32, i1}               @llvm.umul.with.overflow.i32(i32, i32)
declare {<4 x i32>, <4 x i1>}   @llvm.umul.with.overflow.v4i32(<4 x i32>, <4 x i32>)
declare {<8 x i32>, <8 x i1>}   @llvm.umul.with.overflow.v8i32(<8 x i32>, <8 x i32>)
declare {<16 x i32>, <16 x i1>} @llvm.umul.with.overflow.v16i32(<16 x i32>, <16 x i32>)

declare {i16, i1}               @llvm.umul.with.overflow.i16(i16, i16)
declare {<8 x i16>,  <8 x i1>}  @llvm.umul.with.overflow.v8i16(<8 x i16>, <8 x i16>)
declare {<16 x i16>, <16 x i1>} @llvm.umul.with.overflow.v16i16(<16 x i16>, <16 x i16>)
declare {<32 x i16>, <32 x i1>} @llvm.umul.with.overflow.v32i16(<32 x i16>, <32 x i16>)

declare {i8, i1}                @llvm.umul.with.overflow.i8(i8, i8)
declare {<16 x i8>, <16 x i1>}  @llvm.umul.with.overflow.v16i8(<16 x i8>, <16 x i8>)
declare {<32 x i8>, <32 x i1>}  @llvm.umul.with.overflow.v32i8(<32 x i8>, <32 x i8>)
declare {<64 x i8>, <64 x i1>}  @llvm.umul.with.overflow.v64i8(<64 x i8>, <64 x i8>)

define i32 @umul(i32 %arg) {
; V8M-RECIP-LABEL: 'umul'
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %I64 = call { i64, i1 } @llvm.umul.with.overflow.i64(i64 undef, i64 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 74 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.umul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 148 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.umul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 296 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.umul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I32 = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 undef, i32 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 76 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.umul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 152 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.umul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 304 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 192 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 384 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.umul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-RECIP-LABEL: 'umul'
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %I64 = call { i64, i1 } @llvm.umul.with.overflow.i64(i64 undef, i64 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 75 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.umul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 146 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.umul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 288 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.umul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I32 = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 undef, i32 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 13 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.umul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 26 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.umul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 52 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 76 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 76 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.umul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; MVE-RECIP-LABEL: 'umul'
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %I64 = call { i64, i1 } @llvm.umul.with.overflow.i64(i64 undef, i64 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 108 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.umul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 248 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.umul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 624 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.umul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I32 = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 undef, i32 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 84 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.umul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 376 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.umul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1456 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 60 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 416 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1504 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 92 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 1248 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 4704 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.umul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
;
; V8M-SIZE-LABEL: 'umul'
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %I64 = call { i64, i1 } @llvm.umul.with.overflow.i64(i64 undef, i64 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.umul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.umul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.umul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I32 = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 undef, i32 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 17 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.umul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 29 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.umul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 53 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 19 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 35 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 67 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 131 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.umul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; V8M-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; NEON-SIZE-LABEL: 'umul'
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %I64 = call { i64, i1 } @llvm.umul.with.overflow.i64(i64 undef, i64 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.umul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.umul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.umul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I32 = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 undef, i32 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.umul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.umul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.umul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; NEON-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
; MVE-SIZE-LABEL: 'umul'
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %I64 = call { i64, i1 } @llvm.umul.with.overflow.i64(i64 undef, i64 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 25 for instruction: %V2I64 = call { <2 x i64>, <2 x i1> } @llvm.umul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 61 for instruction: %V4I64 = call { <4 x i64>, <4 x i1> } @llvm.umul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 181 for instruction: %V8I64 = call { <8 x i64>, <8 x i1> } @llvm.umul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %I32 = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 undef, i32 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 46 for instruction: %V4I32 = call { <4 x i32>, <4 x i1> } @llvm.umul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 216 for instruction: %V8I32 = call { <8 x i32>, <8 x i1> } @llvm.umul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 812 for instruction: %V16I32 = call { <16 x i32>, <16 x i1> } @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I16 = call { i16, i1 } @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 38 for instruction: %V8I16 = call { <8 x i16>, <8 x i1> } @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 330 for instruction: %V16I16 = call { <16 x i16>, <16 x i1> } @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1170 for instruction: %V32I16 = call { <32 x i16>, <32 x i1> } @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %I8 = call { i8, i1 } @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 70 for instruction: %V16I8 = call { <16 x i8>, <16 x i1> } @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1162 for instruction: %V32I8 = call { <32 x i8>, <32 x i1> } @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 4370 for instruction: %V64I8 = call { <64 x i8>, <64 x i1> } @llvm.umul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)
; MVE-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret i32 undef
;
  %I64 = call {i64, i1} @llvm.umul.with.overflow.i64(i64 undef, i64 undef)
  %V2I64 = call {<2 x i64>, <2 x i1>} @llvm.umul.with.overflow.v2i64(<2 x i64> undef, <2 x i64> undef)
  %V4I64 = call {<4 x i64>, <4 x i1>} @llvm.umul.with.overflow.v4i64(<4 x i64> undef, <4 x i64> undef)
  %V8I64 = call {<8 x i64>, <8 x i1>} @llvm.umul.with.overflow.v8i64(<8 x i64> undef, <8 x i64> undef)

  %I32 = call {i32, i1} @llvm.umul.with.overflow.i32(i32 undef, i32 undef)
  %V4I32  = call {<4 x i32>, <4 x i1>}  @llvm.umul.with.overflow.v4i32(<4 x i32> undef, <4 x i32> undef)
  %V8I32  = call {<8 x i32>, <8 x i1>}  @llvm.umul.with.overflow.v8i32(<8 x i32> undef, <8 x i32> undef)
  %V16I32 = call {<16 x i32>, <16 x i1>} @llvm.umul.with.overflow.v16i32(<16 x i32> undef, <16 x i32> undef)

  %I16 = call {i16, i1} @llvm.umul.with.overflow.i16(i16 undef, i16 undef)
  %V8I16  = call {<8 x i16>, <8 x i1>}  @llvm.umul.with.overflow.v8i16(<8 x i16> undef, <8 x i16> undef)
  %V16I16 = call {<16 x i16>, <16 x i1>} @llvm.umul.with.overflow.v16i16(<16 x i16> undef, <16 x i16> undef)
  %V32I16 = call {<32 x i16>, <32 x i1>} @llvm.umul.with.overflow.v32i16(<32 x i16> undef, <32 x i16> undef)

  %I8 = call {i8, i1} @llvm.umul.with.overflow.i8(i8 undef, i8 undef)
  %V16I8 = call {<16 x i8>, <16 x i1>} @llvm.umul.with.overflow.v16i8(<16 x i8> undef, <16 x i8> undef)
  %V32I8 = call {<32 x i8>, <32 x i1>} @llvm.umul.with.overflow.v32i8(<32 x i8> undef, <32 x i8> undef)
  %V64I8 = call {<64 x i8>, <64 x i1>} @llvm.umul.with.overflow.v64i8(<64 x i8> undef, <64 x i8> undef)

  ret i32 undef
}