; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
; RUN: opt -passes="print<cost-model>" -cost-kind=all 2>&1 -disable-output -mtriple=thumbv8m.main < %s | FileCheck %s --check-prefix=V8M
; RUN: opt -passes="print<cost-model>" -cost-kind=all 2>&1 -disable-output -mtriple=armv8a-linux-gnueabihf < %s | FileCheck %s --check-prefix=NEON
; RUN: opt -passes="print<cost-model>" -cost-kind=all 2>&1 -disable-output -mtriple=armv8.1m.main -mattr=+mve < %s | FileCheck %s --check-prefix=MVE

target datalayout = "e-m:e-p:32:32-Fi8-i64:64-v128:64:128-a:0:32-n32-S64"

define i32 @reduce_i64(i32 %arg) {
; V8M-LABEL: 'reduce_i64'
; V8M-NEXT:  Cost Model: Found costs of 2 for: %V1 = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> undef)
; V8M-NEXT:  Cost Model: Found costs of 8 for: %V2 = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
; V8M-NEXT:  Cost Model: Found costs of 20 for: %V4 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
; V8M-NEXT:  Cost Model: Found costs of 44 for: %V8 = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
; V8M-NEXT:  Cost Model: Found costs of 92 for: %V16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
; V8M-NEXT:  Cost Model: Found costs of 1 for: ret i32 undef
;
; NEON-LABEL: 'reduce_i64'
; NEON-NEXT:  Cost Model: Found costs of 3 for: %V1 = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> undef)
; NEON-NEXT:  Cost Model: Found costs of 16 for: %V2 = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
; NEON-NEXT:  Cost Model: Found costs of 29 for: %V4 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
; NEON-NEXT:  Cost Model: Found costs of RThru:55 CodeSize:54 Lat:54 SizeLat:54 for: %V8 = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
; NEON-NEXT:  Cost Model: Found costs of RThru:107 CodeSize:103 Lat:103 SizeLat:103 for: %V16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
; NEON-NEXT:  Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret i32 undef
;
; MVE-LABEL: 'reduce_i64'
; MVE-NEXT:  Cost Model: Found costs of 8 for: %V1 = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:108 CodeSize:76 Lat:108 SizeLat:108 for: %V2 = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:208 CodeSize:144 Lat:208 SizeLat:208 for: %V4 = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:408 CodeSize:280 Lat:408 SizeLat:408 for: %V8 = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:808 CodeSize:552 Lat:808 SizeLat:808 for: %V16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret i32 undef
;
  %V1  = call i64 @llvm.vector.reduce.add.v1i64(<1 x i64> undef)
  %V2  = call i64 @llvm.vector.reduce.add.v2i64(<2 x i64> undef)
  %V4  = call i64 @llvm.vector.reduce.add.v4i64(<4 x i64> undef)
  %V8  = call i64 @llvm.vector.reduce.add.v8i64(<8 x i64> undef)
  %V16 = call i64 @llvm.vector.reduce.add.v16i64(<16 x i64> undef)
  ret i32 undef
}

define i32 @reduce_i32(i32 %arg) {
; V8M-LABEL: 'reduce_i32'
; V8M-NEXT:  Cost Model: Found costs of 4 for: %V2 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
; V8M-NEXT:  Cost Model: Found costs of 10 for: %V4 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
; V8M-NEXT:  Cost Model: Found costs of 22 for: %V8 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
; V8M-NEXT:  Cost Model: Found costs of 46 for: %V16 = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
; V8M-NEXT:  Cost Model: Found costs of 94 for: %V32 = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> undef)
; V8M-NEXT:  Cost Model: Found costs of 190 for: %V64 = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> undef)
; V8M-NEXT:  Cost Model: Found costs of 382 for: %V128 = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> undef)
; V8M-NEXT:  Cost Model: Found costs of 1 for: ret i32 undef
;
; NEON-LABEL: 'reduce_i32'
; NEON-NEXT:  Cost Model: Found costs of 16 for: %V2 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
; NEON-NEXT:  Cost Model: Found costs of 53 for: %V4 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
; NEON-NEXT:  Cost Model: Found costs of 150 for: %V8 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
; NEON-NEXT:  Cost Model: Found costs of 391 for: %V16 = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
; NEON-NEXT:  Cost Model: Found costs of 488 for: %V32 = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> undef)
; NEON-NEXT:  Cost Model: Found costs of RThru:682 CodeSize:681 Lat:681 SizeLat:681 for: %V64 = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> undef)
; NEON-NEXT:  Cost Model: Found costs of RThru:1070 CodeSize:1066 Lat:1066 SizeLat:1066 for: %V128 = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> undef)
; NEON-NEXT:  Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret i32 undef
;
; MVE-LABEL: 'reduce_i32'
; MVE-NEXT:  Cost Model: Found costs of RThru:54 CodeSize:38 Lat:54 SizeLat:54 for: %V2 = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:2 CodeSize:1 Lat:2 SizeLat:2 for: %V4 = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:2 CodeSize:1 Lat:2 SizeLat:2 for: %V8 = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:2 CodeSize:1 Lat:2 SizeLat:2 for: %V16 = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:4 CodeSize:2 Lat:4 SizeLat:4 for: %V32 = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:8 CodeSize:4 Lat:8 SizeLat:8 for: %V64 = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:16 CodeSize:8 Lat:16 SizeLat:16 for: %V128 = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> undef)
; MVE-NEXT:  Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret i32 undef
;
  %V2   = call i8 @llvm.vector.reduce.add.v2i8(<2 x i8> undef)
  %V4   = call i8 @llvm.vector.reduce.add.v4i8(<4 x i8> undef)
  %V8   = call i8 @llvm.vector.reduce.add.v8i8(<8 x i8> undef)
  %V16  = call i8 @llvm.vector.reduce.add.v16i8(<16 x i8> undef)
  %V32  = call i8 @llvm.vector.reduce.add.v32i8(<32 x i8> undef)
  %V64  = call i8 @llvm.vector.reduce.add.v64i8(<64 x i8> undef)
  %V128 = call i8 @llvm.vector.reduce.add.v128i8(<128 x i8> undef)
  ret i32 undef
}