Run DCE after a LoopFlatten test to reduce spurious output [nfc]
[llvm-project.git] / clang / test / CodeGen / aarch64-sve-intrinsics / acle_sve_dupq-bfloat.c
blob5d238ffeca94a419d76f7aa69af193dd3d195e0b
1 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
2 // REQUIRES: aarch64-registered-target
3 // RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s
4 // RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
5 // RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s
6 // RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -passes=mem2reg,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
8 // RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +bf16 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
9 #include <arm_sve.h>
11 #ifdef SVE_OVERLOADED_FORMS
12 // A simple used,unused... macro, long enough to represent any SVE builtin.
13 #define SVE_ACLE_FUNC(A1, A2_UNUSED, A3, A4_UNUSED) A1##A3
14 #else
15 #define SVE_ACLE_FUNC(A1, A2, A3, A4) A1##A2##A3##A4
16 #endif
18 // CHECK-LABEL: @test_svdupq_lane_bf16(
19 // CHECK-NEXT: entry:
20 // CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x bfloat> @llvm.aarch64.sve.dupq.lane.nxv8bf16(<vscale x 8 x bfloat> [[DATA:%.*]], i64 [[INDEX:%.*]])
21 // CHECK-NEXT: ret <vscale x 8 x bfloat> [[TMP0]]
23 // CPP-CHECK-LABEL: @_Z21test_svdupq_lane_bf16u14__SVBfloat16_tm(
24 // CPP-CHECK-NEXT: entry:
25 // CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x bfloat> @llvm.aarch64.sve.dupq.lane.nxv8bf16(<vscale x 8 x bfloat> [[DATA:%.*]], i64 [[INDEX:%.*]])
26 // CPP-CHECK-NEXT: ret <vscale x 8 x bfloat> [[TMP0]]
28 svbfloat16_t test_svdupq_lane_bf16(svbfloat16_t data, uint64_t index) {
29 // expected-warning@+1 {{implicit declaration of function 'svdupq_lane_bf16'}}
30 return SVE_ACLE_FUNC(svdupq_lane, _bf16, , )(data, index);
32 // CHECK-LABEL: @test_svdupq_n_bf16(
33 // CHECK-NEXT: entry:
34 // CHECK-NEXT: [[TMP0:%.*]] = insertelement <8 x bfloat> poison, bfloat [[X0:%.*]], i64 0
35 // CHECK-NEXT: [[TMP1:%.*]] = insertelement <8 x bfloat> [[TMP0]], bfloat [[X1:%.*]], i64 1
36 // CHECK-NEXT: [[TMP2:%.*]] = insertelement <8 x bfloat> [[TMP1]], bfloat [[X2:%.*]], i64 2
37 // CHECK-NEXT: [[TMP3:%.*]] = insertelement <8 x bfloat> [[TMP2]], bfloat [[X3:%.*]], i64 3
38 // CHECK-NEXT: [[TMP4:%.*]] = insertelement <8 x bfloat> [[TMP3]], bfloat [[X4:%.*]], i64 4
39 // CHECK-NEXT: [[TMP5:%.*]] = insertelement <8 x bfloat> [[TMP4]], bfloat [[X5:%.*]], i64 5
40 // CHECK-NEXT: [[TMP6:%.*]] = insertelement <8 x bfloat> [[TMP5]], bfloat [[X6:%.*]], i64 6
41 // CHECK-NEXT: [[TMP7:%.*]] = insertelement <8 x bfloat> [[TMP6]], bfloat [[X7:%.*]], i64 7
42 // CHECK-NEXT: [[TMP8:%.*]] = tail call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.v8bf16(<vscale x 8 x bfloat> poison, <8 x bfloat> [[TMP7]], i64 0)
43 // CHECK-NEXT: [[TMP9:%.*]] = tail call <vscale x 8 x bfloat> @llvm.aarch64.sve.dupq.lane.nxv8bf16(<vscale x 8 x bfloat> [[TMP8]], i64 0)
44 // CHECK-NEXT: ret <vscale x 8 x bfloat> [[TMP9]]
46 // CPP-CHECK-LABEL: @_Z18test_svdupq_n_bf16u6__bf16u6__bf16u6__bf16u6__bf16u6__bf16u6__bf16u6__bf16u6__bf16(
47 // CPP-CHECK-NEXT: entry:
48 // CPP-CHECK-NEXT: [[TMP0:%.*]] = insertelement <8 x bfloat> poison, bfloat [[X0:%.*]], i64 0
49 // CPP-CHECK-NEXT: [[TMP1:%.*]] = insertelement <8 x bfloat> [[TMP0]], bfloat [[X1:%.*]], i64 1
50 // CPP-CHECK-NEXT: [[TMP2:%.*]] = insertelement <8 x bfloat> [[TMP1]], bfloat [[X2:%.*]], i64 2
51 // CPP-CHECK-NEXT: [[TMP3:%.*]] = insertelement <8 x bfloat> [[TMP2]], bfloat [[X3:%.*]], i64 3
52 // CPP-CHECK-NEXT: [[TMP4:%.*]] = insertelement <8 x bfloat> [[TMP3]], bfloat [[X4:%.*]], i64 4
53 // CPP-CHECK-NEXT: [[TMP5:%.*]] = insertelement <8 x bfloat> [[TMP4]], bfloat [[X5:%.*]], i64 5
54 // CPP-CHECK-NEXT: [[TMP6:%.*]] = insertelement <8 x bfloat> [[TMP5]], bfloat [[X6:%.*]], i64 6
55 // CPP-CHECK-NEXT: [[TMP7:%.*]] = insertelement <8 x bfloat> [[TMP6]], bfloat [[X7:%.*]], i64 7
56 // CPP-CHECK-NEXT: [[TMP8:%.*]] = tail call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.v8bf16(<vscale x 8 x bfloat> poison, <8 x bfloat> [[TMP7]], i64 0)
57 // CPP-CHECK-NEXT: [[TMP9:%.*]] = tail call <vscale x 8 x bfloat> @llvm.aarch64.sve.dupq.lane.nxv8bf16(<vscale x 8 x bfloat> [[TMP8]], i64 0)
58 // CPP-CHECK-NEXT: ret <vscale x 8 x bfloat> [[TMP9]]
60 svbfloat16_t test_svdupq_n_bf16(bfloat16_t x0, bfloat16_t x1, bfloat16_t x2, bfloat16_t x3,
61 bfloat16_t x4, bfloat16_t x5, bfloat16_t x6, bfloat16_t x7) {
62 // <assume other insertelement>
63 // expected-warning@+1 {{implicit declaration of function 'svdupq_n_bf16'}}
64 return SVE_ACLE_FUNC(svdupq, _n, _bf16, )(x0, x1, x2, x3, x4, x5, x6, x7);