Debian Codegraph @

clang/test/CodeGen/AArch64/sve2p1-intrinsics/acle_sve2p1_sclamp.c

raw
1 // NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
2 // REQUIRES: aarch64-registered-target
3 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 \
4 // RUN: -Werror -emit-llvm -disable-O0-optnone -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
5 // RUN: %clang_cc1 -fclang-abi-compat=latest -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 \
6 // RUN: -Werror -emit-llvm -disable-O0-optnone -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
7 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 \
8 // RUN: -Werror -emit-llvm -disable-O0-optnone -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
9 // RUN: %clang_cc1 -fclang-abi-compat=latest -DSVE_OVERLOADED_FORMS -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 \
10 // RUN: -Werror -emit-llvm -disable-O0-optnone -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
11 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 \
12 // RUN: -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
13 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sme \
14 // RUN: -Werror -emit-llvm -disable-O0-optnone -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
15 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sme -target-feature +sve \
16 // RUN: -Werror -emit-llvm -disable-O0-optnone -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
17 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sme \
18 // RUN: -Werror -emit-llvm -disable-O0-optnone -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
19 // RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64-none-linux-gnu -target-feature +sme \
20 // RUN: -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
21
22 #include <arm_sve.h>
23
24 #if defined(__ARM_FEATURE_SME) && defined(__ARM_FEATURE_SVE)
25 #define ATTR __arm_streaming_compatible
26 #elif defined(__ARM_FEATURE_SME)
27 #define ATTR __arm_streaming
28 #else
29 #define ATTR
30 #endif
31
32 #ifdef SVE_OVERLOADED_FORMS
33 // A simple used,unused... macro, long enough to represent any SVE builtin.
34 #define SVE_ACLE_FUNC(A1, A2_UNUSED, A3, A4_UNUSED) A1##A3
35 #else
36 #define SVE_ACLE_FUNC(A1, A2, A3, A4) A1##A2##A3##A4
37 #endif
38
39 // CHECK-LABEL: @test_svclamp_s8(
40 // CHECK-NEXT: entry:
41 // CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sclamp.nxv16i8(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
42 // CHECK-NEXT: ret <vscale x 16 x i8> [[TMP0]]
43 //
44 // CPP-CHECK-LABEL: @_Z15test_svclamp_s8u10__SVInt8_tS_S_(
45 // CPP-CHECK-NEXT: entry:
46 // CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 16 x i8> @llvm.aarch64.sve.sclamp.nxv16i8(<vscale x 16 x i8> [[OP1:%.*]], <vscale x 16 x i8> [[OP2:%.*]], <vscale x 16 x i8> [[OP3:%.*]])
47 // CPP-CHECK-NEXT: ret <vscale x 16 x i8> [[TMP0]]
48 //
49 svint8_t test_svclamp_s8(svint8_t op1, svint8_t op2, svint8_t op3) ATTR {
50 return SVE_ACLE_FUNC(svclamp, _s8, , )(op1, op2, op3);
51 }
52
53 // CHECK-LABEL: @test_svclamp_s16(
54 // CHECK-NEXT: entry:
55 // CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.sclamp.nxv8i16(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
56 // CHECK-NEXT: ret <vscale x 8 x i16> [[TMP0]]
57 //
58 // CPP-CHECK-LABEL: @_Z16test_svclamp_s16u11__SVInt16_tS_S_(
59 // CPP-CHECK-NEXT: entry:
60 // CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 8 x i16> @llvm.aarch64.sve.sclamp.nxv8i16(<vscale x 8 x i16> [[OP1:%.*]], <vscale x 8 x i16> [[OP2:%.*]], <vscale x 8 x i16> [[OP3:%.*]])
61 // CPP-CHECK-NEXT: ret <vscale x 8 x i16> [[TMP0]]
62 //
63 svint16_t test_svclamp_s16(svint16_t op1, svint16_t op2, svint16_t op3) ATTR {
64 return SVE_ACLE_FUNC(svclamp, _s16, , )(op1, op2, op3);
65 }
66
67 // CHECK-LABEL: @test_svclamp_s32(
68 // CHECK-NEXT: entry:
69 // CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.sclamp.nxv4i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
70 // CHECK-NEXT: ret <vscale x 4 x i32> [[TMP0]]
71 //
72 // CPP-CHECK-LABEL: @_Z16test_svclamp_s32u11__SVInt32_tS_S_(
73 // CPP-CHECK-NEXT: entry:
74 // CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 4 x i32> @llvm.aarch64.sve.sclamp.nxv4i32(<vscale x 4 x i32> [[OP1:%.*]], <vscale x 4 x i32> [[OP2:%.*]], <vscale x 4 x i32> [[OP3:%.*]])
75 // CPP-CHECK-NEXT: ret <vscale x 4 x i32> [[TMP0]]
76 //
77 svint32_t test_svclamp_s32(svint32_t op1, svint32_t op2, svint32_t op3) ATTR {
78 return SVE_ACLE_FUNC(svclamp, _s32, , )(op1, op2, op3);
79 }
80
81 // CHECK-LABEL: @test_svclamp_s64(
82 // CHECK-NEXT: entry:
83 // CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.sclamp.nxv2i64(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
84 // CHECK-NEXT: ret <vscale x 2 x i64> [[TMP0]]
85 //
86 // CPP-CHECK-LABEL: @_Z16test_svclamp_s64u11__SVInt64_tS_S_(
87 // CPP-CHECK-NEXT: entry:
88 // CPP-CHECK-NEXT: [[TMP0:%.*]] = tail call <vscale x 2 x i64> @llvm.aarch64.sve.sclamp.nxv2i64(<vscale x 2 x i64> [[OP1:%.*]], <vscale x 2 x i64> [[OP2:%.*]], <vscale x 2 x i64> [[OP3:%.*]])
89 // CPP-CHECK-NEXT: ret <vscale x 2 x i64> [[TMP0]]
90 //
91 svint64_t test_svclamp_s64(svint64_t op1, svint64_t op2, svint64_t op3) ATTR {
92 return SVE_ACLE_FUNC(svclamp, _s64, , )(op1, op2, op3);
93 }
94