1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
|
// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
// RUN: %clang_cc1 -std=c++23 -triple aarch64-none-linux-gnu -target-feature +sve \
// RUN: -target-feature +bf16 -target-feature +sve -target-feature +sme -target-feature +sme2 -target-feature +sve2 -target-feature +neon -Waarch64-sme-attributes -fsyntax-only -verify %s
// REQUIRES: aarch64-registered-target
#include "arm_neon.h"
#include "arm_sme.h"
#include "arm_sve.h"
int16x8_t incompat_neon_sm(int16x8_t splat) __arm_streaming {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return (int16x8_t)__builtin_neon_vqaddq_v((int8x16_t)splat, (int8x16_t)splat, 33);
}
__arm_locally_streaming int16x8_t incompat_neon_ls(int16x8_t splat) {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return (int16x8_t)__builtin_neon_vqaddq_v((int8x16_t)splat, (int8x16_t)splat, 33);
}
int16x8_t incompat_neon_smc(int16x8_t splat) __arm_streaming_compatible {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return (int16x8_t)__builtin_neon_vqaddq_v((int8x16_t)splat, (int8x16_t)splat, 33);
}
void incompat_sme_smc(svbool_t pg, void const *ptr) __arm_streaming_compatible __arm_inout("za") {
// expected-error@+1 {{builtin can only be called from a streaming function}}
return __builtin_sme_svld1_hor_za128(0, 0, pg, ptr);
}
float incomp_sve_sm_fadda_sm(void) __arm_streaming {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return svadda(svptrue_b32(), 0, svdup_f32(1));
}
float incomp_sve_sm_fadda_smc(void) __arm_streaming_compatible {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return svadda(svptrue_b32(), 0, svdup_f32(1));
}
svuint32_t incompat_sve_sm(svbool_t pg, svuint32_t a, int16_t b) __arm_streaming {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return __builtin_sve_svld1_gather_u32base_index_u32(pg, a, b);
}
// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
// expected-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
__arm_locally_streaming svuint32_t incompat_sve_ls(svbool_t pg, svuint32_t a, int64_t b) {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return __builtin_sve_svld1_gather_u32base_index_u32(pg, a, b);
}
svuint32_t incompat_sve_smc(svbool_t pg, svuint32_t a, int64_t b) __arm_streaming_compatible {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return __builtin_sve_svld1_gather_u32base_index_u32(pg, a, b);
}
svuint32_t incompat_sve2_sm(svbool_t pg, svuint32_t a, int64_t b) __arm_streaming {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return __builtin_sve_svldnt1_gather_u32base_index_u32(pg, a, b);
}
// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
// expected-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
__arm_locally_streaming svuint32_t incompat_sve2_ls(svbool_t pg, svuint32_t a, int64_t b) {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return __builtin_sve_svldnt1_gather_u32base_index_u32(pg, a, b);
}
svuint32_t incompat_sve2_smc(svbool_t pg, svuint32_t a, int64_t b) __arm_streaming_compatible {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return __builtin_sve_svldnt1_gather_u32base_index_u32(pg, a, b);
}
void incompat_sme_sm(svbool_t pn, svbool_t pm, svfloat32_t zn, svfloat32_t zm) __arm_inout("za") {
// expected-error@+1 {{builtin can only be called from a streaming function}}
svmops_za32_f32_m(0, pn, pm, zn, zm);
}
svfloat64_t streaming_caller_sve(svbool_t pg, svfloat64_t a, float64_t b) __arm_streaming {
return svadd_n_f64_m(pg, a, b);
}
// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
// expected-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
__arm_locally_streaming svfloat64_t locally_streaming_caller_sve(svbool_t pg, svfloat64_t a, float64_t b) {
return svadd_n_f64_m(pg, a, b);
}
svfloat64_t streaming_compatible_caller_sve(svbool_t pg, svfloat64_t a, float64_t b) __arm_streaming_compatible {
return svadd_n_f64_m(pg, a, b);
}
svint16_t streaming_caller_sve2(svint16_t op1, svint16_t op2) __arm_streaming {
return svmul_lane_s16(op1, op2, 0);
}
// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
// expected-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
__arm_locally_streaming svint16_t locally_streaming_caller_sve2(svint16_t op1, svint16_t op2) {
return svmul_lane_s16(op1, op2, 0);
}
svint16_t streaming_compatible_caller_sve2(svint16_t op1, svint16_t op2) __arm_streaming_compatible {
return svmul_lane_s16(op1, op2, 0);
}
svbool_t streaming_caller_ptrue(void) __arm_streaming {
return svand_z(svptrue_b16(), svptrue_pat_b16(SV_ALL), svptrue_pat_b16(SV_VL4));
}
svint8_t missing_za(svint8_t zd, svbool_t pg, uint32_t slice_base) __arm_streaming {
// expected-warning@+1 {{builtin call is not valid when calling from a function without active ZA state}}
return svread_hor_za8_s8_m(zd, pg, 0, slice_base);
}
__arm_new("za")
svint8_t new_za(svint8_t zd, svbool_t pg, uint32_t slice_base) __arm_streaming {
return svread_hor_za8_s8_m(zd, pg, 0, slice_base);
}
void missing_zt0(void) __arm_streaming {
// expected-warning@+1 {{builtin call is not valid when calling from a function without active ZT0 state}}
svzero_zt(0);
}
__arm_new("zt0")
void new_zt0(void) __arm_streaming { svzero_zt(0); }
/// C++ lambda tests:
void use_streaming_builtin_in_lambda(uint32_t slice_base, svbool_t pg, const void *ptr) __arm_streaming __arm_out("za")
{
[&]{
/// The lambda is its own function and does not inherit the SME attributes (so this should error).
// expected-error@+1 {{builtin can only be called from a streaming function}}
svld1_hor_za64(0, slice_base, pg, ptr);
}();
}
void use_streaming_builtin(uint32_t slice_base, svbool_t pg, const void *ptr) __arm_streaming __arm_out("za")
{
/// Without the lambda the same builtin is okay (as the SME attributes apply).
svld1_hor_za64(0, slice_base, pg, ptr);
}
int16x8_t use_neon_builtin_sm(int16x8_t splat) __arm_streaming_compatible {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return (int16x8_t)__builtin_neon_vqaddq_v((int8x16_t)splat, (int8x16_t)splat, 33);
}
int16x8_t use_neon_builtin_sm_in_lambda(int16x8_t splat) __arm_streaming_compatible {
return [&]{
/// This should not error (as we switch out of streaming mode to execute the lambda).
/// Note: The result int16x8_t is spilled and reloaded as a q-register.
return (int16x8_t)__builtin_neon_vqaddq_v((int8x16_t)splat, (int8x16_t)splat, 33);
}();
}
float use_incomp_sve_builtin_sm() __arm_streaming {
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return svadda(svptrue_b32(), 0, svdup_f32(1));
}
float incomp_sve_sm_fadda_sm_in_lambda(void) __arm_streaming {
return [&]{
/// This should work like the Neon builtin.
return svadda(svptrue_b32(), 0, svdup_f32(1));
}();
}
void use_streaming_builtin_in_streaming_lambda(uint32_t slice_base, const void *ptr)
{
[&] __arm_new("za") () __arm_streaming {
// Here the lambda is streaming with ZA state, so this is okay.
svld1_hor_za64(0, slice_base, svptrue_b64(), ptr);
}();
}
int16x8_t use_neon_builtin_in_streaming_lambda(int16x8_t splat) {
return [&]() __arm_streaming_compatible {
/// This should error as the lambda is streaming-compatible.
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return (int16x8_t)__builtin_neon_vqaddq_v((int8x16_t)splat, (int8x16_t)splat, 33);
}();
}
float incomp_sve_fadda_in_streaming_lambda(void) {
return [&]() __arm_streaming {
// Should error (like the Neon case above).
// expected-error@+1 {{builtin can only be called from a non-streaming function}}
return svadda(svptrue_b32(), 0, svdup_f32(1));
}();
}
|
