1 | // Tests the phases generated for a CUDA offloading target for different |
2 | // combinations of: |
3 | // - Number of gpu architectures; |
4 | // - Host/device-only compilation; |
5 | // - User-requested final phase - binary or assembly. |
6 | |
7 | // REQUIRES: clang-driver |
8 | // REQUIRES: powerpc-registered-target |
9 | // REQUIRES: nvptx-registered-target |
10 | // REQUIRES: amdgpu-registered-target |
11 | // |
12 | // Test single gpu architecture with complete compilation. |
13 | // |
14 | // Test CUDA NVPTX phases. |
15 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
16 | // RUN: --cuda-gpu-arch=sm_30 %s 2>&1 \ |
17 | // RUN: | FileCheck -check-prefixes=BIN,BIN_NV %s |
18 | // |
19 | // Test HIP AMDGPU -fgpu-rdc phases. |
20 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
21 | // RUN: --cuda-gpu-arch=gfx803 -fgpu-rdc %s 2>&1 \ |
22 | // RUN: | FileCheck -check-prefixes=BIN,BIN_AMD,BIN_AMD_RDC %s |
23 | // |
24 | // Test HIP AMDGPU -fno-gpu-rdc phases (default). |
25 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
26 | // RUN: --cuda-gpu-arch=gfx803 %s 2>&1 \ |
27 | // RUN: | FileCheck -check-prefixes=BIN,BIN_AMD,BIN_AMD_NRDC %s |
28 | // |
29 | // BIN_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (host-[[T]]) |
30 | // BIN_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (host-[[T]]) |
31 | // BIN-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]]) |
32 | // BIN-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]]) |
33 | // BIN_NV-DAG: [[P3:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (device-[[T]], [[ARCH:sm_30]]) |
34 | // BIN_AMD-DAG: [[P3:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (device-[[T]], [[ARCH:gfx803]]) |
35 | // BIN-DAG: [[P4:[0-9]+]]: preprocessor, {[[P3]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH]]) |
36 | // BIN-DAG: [[P5:[0-9]+]]: compiler, {[[P4]]}, ir, (device-[[T]], [[ARCH]]) |
37 | // BIN_NV-DAG: [[P6:[0-9]+]]: backend, {[[P5]]}, assembler, (device-[[T]], [[ARCH]]) |
38 | // BIN_NV-DAG: [[P7:[0-9]+]]: assembler, {[[P6]]}, object, (device-[[T]], [[ARCH]]) |
39 | // BIN_NV-DAG: [[P8:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE:nvptx64-nvidia-cuda]]:[[ARCH]])" {[[P7]]}, object |
40 | // BIN_NV-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE]]:[[ARCH]])" {[[P6]]}, assembler |
41 | // BIN_NV-DAG: [[P10:[0-9]+]]: linker, {[[P8]], [[P9]]}, cuda-fatbin, (device-[[T]]) |
42 | // BIN_NV-DAG: [[P11:[0-9]+]]: offload, "host-[[T]] (powerpc64le-ibm-linux-gnu)" {[[P2]]}, "device-[[T]] ([[TRIPLE]])" {[[P10]]}, ir |
43 | // BIN_NV-DAG: [[P12:[0-9]+]]: backend, {[[P11]]}, assembler, (host-[[T]]) |
44 | // BIN_AMD_RDC-DAG: [[P12:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]]) |
45 | // BIN_AMD_NRDC-DAG: [[P6:[0-9]+]]: linker, {[[P5]]}, image, (device-hip, [[ARCH]]) |
46 | // BIN_AMD_NRDC-DAG: [[P7:[0-9]+]]: offload, "device-hip (amdgcn-amd-amdhsa:[[ARCH]])" {[[P6]]}, image |
47 | // BIN_AMD_NRDC-DAG: [[P8:[0-9]+]]: linker, {[[P7]]}, hip-fatbin, (device-hip) |
48 | // BIN_AMD_NRDC-DAG: [[P11:[0-9]+]]: offload, "host-hip (powerpc64le-ibm-linux-gnu)" {[[P2]]}, "device-hip (amdgcn-amd-amdhsa)" {[[P8]]}, ir |
49 | // BIN_AMD_NRDC-DAG: [[P12:[0-9]+]]: backend, {[[P11]]}, assembler, (host-[[T]]) |
50 | // BIN-DAG: [[P13:[0-9]+]]: assembler, {[[P12]]}, object, (host-[[T]]) |
51 | // BIN-DAG: [[P14:[0-9]+]]: linker, {[[P13]]}, image, (host-[[T]]) |
52 | // BIN_AMD_RDC-DAG: [[P15:[0-9]+]]: linker, {[[P5]]}, image, (device-[[T]], [[ARCH]]) |
53 | // BIN_AMD_RDC-DAG: [[P16:[0-9]+]]: offload, "host-[[T]] (powerpc64le-ibm-linux-gnu)" {[[P14]]}, |
54 | // BIN_AMD_RDC-DAG-SAME: "device-[[T]] ([[TRIPLE:amdgcn-amd-amdhsa]]:[[ARCH]])" {[[P15]]}, object |
55 | |
56 | // |
57 | // Test single gpu architecture up to the assemble phase. |
58 | // |
59 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
60 | // RUN: --cuda-gpu-arch=sm_30 %s -S 2>&1 \ |
61 | // RUN: | FileCheck -check-prefixes=ASM,ASM_NV %s |
62 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
63 | // RUN: --cuda-gpu-arch=gfx803 -fgpu-rdc %s -S 2>&1 \ |
64 | // RUN: | FileCheck -check-prefixes=ASM,ASM_AMD %s |
65 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
66 | // RUN: --cuda-gpu-arch=gfx803 -fcuda-rdc %s -S 2>&1 \ |
67 | // RUN: | FileCheck -check-prefixes=ASM,ASM_AMD %s |
68 | // ASM_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (device-[[T]], [[ARCH:sm_30]]) |
69 | // ASM_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (device-[[T]], [[ARCH:gfx803]]) |
70 | // ASM-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH]]) |
71 | // ASM-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (device-[[T]], [[ARCH]]) |
72 | // ASM_NV-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (device-[[T]], [[ARCH]]) |
73 | // ASM_NV-DAG: [[P4:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE:nvptx64-nvidia-cuda|amdgcn-amd-amdhsa]]:[[ARCH]])" {[[P3]]}, assembler |
74 | // ASM-DAG: [[P5:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (host-[[T]]) |
75 | // ASM-DAG: [[P6:[0-9]+]]: preprocessor, {[[P5]]}, [[T]]-cpp-output, (host-[[T]]) |
76 | // ASM-DAG: [[P7:[0-9]+]]: compiler, {[[P6]]}, ir, (host-[[T]]) |
77 | // ASM-DAG: [[P8:[0-9]+]]: backend, {[[P7]]}, assembler, (host-[[T]]) |
78 | |
79 | // |
80 | // Test two gpu architectures with complete compilation. |
81 | // |
82 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
83 | // RUN: --cuda-gpu-arch=sm_30 --cuda-gpu-arch=sm_35 %s 2>&1 \ |
84 | // RUN: | FileCheck -check-prefixes=BIN2,BIN2_NV %s |
85 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
86 | // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 -fgpu-rdc %s 2>&1 \ |
87 | // RUN: | FileCheck -check-prefixes=BIN2,BIN2_AMD %s |
88 | // BIN2_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (host-[[T]]) |
89 | // BIN2_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (host-[[T]]) |
90 | // BIN2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]]) |
91 | // BIN2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]]) |
92 | // BIN2-DAG: [[P3:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (device-[[T]], [[ARCH1:sm_30|gfx803]]) |
93 | // BIN2-DAG: [[P4:[0-9]+]]: preprocessor, {[[P3]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH1]]) |
94 | // BIN2-DAG: [[P5:[0-9]+]]: compiler, {[[P4]]}, ir, (device-[[T]], [[ARCH1]]) |
95 | // BIN2_NV-DAG: [[P6:[0-9]+]]: backend, {[[P5]]}, assembler, (device-[[T]], [[ARCH1]]) |
96 | // BIN2_NV-DAG: [[P7:[0-9]+]]: assembler, {[[P6]]}, object, (device-[[T]], [[ARCH1]]) |
97 | // BIN2_NV-DAG: [[P8:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE:nvptx64-nvidia-cuda]]:[[ARCH1]])" {[[P7]]}, object |
98 | // BIN2_NV-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE]]:[[ARCH1]])" {[[P6]]}, assembler |
99 | // BIN2-DAG: [[P10:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (device-[[T]], [[ARCH2:sm_35|gfx900]]) |
100 | // BIN2-DAG: [[P11:[0-9]+]]: preprocessor, {[[P10]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH2]]) |
101 | // BIN2-DAG: [[P12:[0-9]+]]: compiler, {[[P11]]}, ir, (device-[[T]], [[ARCH2]]) |
102 | // BIN2_NV-DAG: [[P13:[0-9]+]]: backend, {[[P12]]}, assembler, (device-[[T]], [[ARCH2]]) |
103 | // BIN2_NV-DAG: [[P14:[0-9]+]]: assembler, {[[P13]]}, object, (device-[[T]], [[ARCH2]]) |
104 | // BIN2_NV-DAG: [[P15:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE]]:[[ARCH2]])" {[[P14]]}, object |
105 | // BIN2_NV-DAG: [[P16:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE]]:[[ARCH2]])" {[[P13]]}, assembler |
106 | // BIN2_NV-DAG: [[P17:[0-9]+]]: linker, {[[P8]], [[P9]], [[P15]], [[P16]]}, cuda-fatbin, (device-[[T]]) |
107 | // BIN2_NV-DAG: [[P18:[0-9]+]]: offload, "host-[[T]] (powerpc64le-ibm-linux-gnu)" {[[P2]]}, "device-[[T]] ([[TRIPLE]])" {[[P17]]}, ir |
108 | // BIN2_NV-DAG: [[P19:[0-9]+]]: backend, {[[P18]]}, assembler, (host-[[T]]) |
109 | // BIN2_AMD-DAG: [[P19:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]]) |
110 | // BIN2-DAG: [[P20:[0-9]+]]: assembler, {[[P19]]}, object, (host-[[T]]) |
111 | // BIN2-DAG: [[P21:[0-9]+]]: linker, {[[P20]]}, image, (host-[[T]]) |
112 | // BIN2_AMD-DAG: [[P22:[0-9]+]]: linker, {[[P5]]}, image, (device-[[T]], [[ARCH1]]) |
113 | // BIN2_AMD-DAG: [[P23:[0-9]+]]: linker, {[[P12]]}, image, (device-[[T]], [[ARCH2]]) |
114 | // BIN2_AMD-DAG: [[P24:[0-9]+]]: offload, "host-[[T]] (powerpc64le-ibm-linux-gnu)" {[[P21]]}, |
115 | // BIN2_AMD-DAG-SAME: "device-[[T]] ([[TRIPLE:amdgcn-amd-amdhsa]]:[[ARCH1]])" {[[P22]]}, |
116 | // BIN2_AMD-DAG-SAME: "device-[[T]] ([[TRIPLE:amdgcn-amd-amdhsa]]:[[ARCH2]])" {[[P23]]}, object |
117 | |
118 | // |
119 | // Test two gpu architecturess up to the assemble phase. |
120 | // |
121 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
122 | // RUN: --cuda-gpu-arch=sm_30 --cuda-gpu-arch=sm_35 %s -S 2>&1 \ |
123 | // RUN: | FileCheck -check-prefixes=ASM2,ASM2_NV %s |
124 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
125 | // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 -fgpu-rdc %s -S 2>&1 \ |
126 | // RUN: | FileCheck -check-prefixes=ASM2,ASM2_AMD %s |
127 | // ASM2_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (device-[[T]], [[ARCH1:sm_30]]) |
128 | // ASM2_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (device-[[T]], [[ARCH1:gfx803]]) |
129 | // ASM2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH1]]) |
130 | // ASM2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (device-[[T]], [[ARCH1]]) |
131 | // ASM2_NV-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (device-[[T]], [[ARCH1]]) |
132 | // ASM2_NV-DAG: [[P4:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE:nvptx64-nvidia-cuda|amdgcn-amd-amdhsa]]:[[ARCH1]])" {[[P3]]}, assembler |
133 | // ASM2-DAG: [[P5:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (device-[[T]], [[ARCH2:sm_35|gfx900]]) |
134 | // ASM2-DAG: [[P6:[0-9]+]]: preprocessor, {[[P5]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH2]]) |
135 | // ASM2-DAG: [[P7:[0-9]+]]: compiler, {[[P6]]}, ir, (device-[[T]], [[ARCH2]]) |
136 | // ASM2_NV-DAG: [[P8:[0-9]+]]: backend, {[[P7]]}, assembler, (device-[[T]], [[ARCH2]]) |
137 | // ASM2_NV-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE]]:[[ARCH2]])" {[[P8]]}, assembler |
138 | // ASM2-DAG: [[P10:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (host-[[T]]) |
139 | // ASM2-DAG: [[P11:[0-9]+]]: preprocessor, {[[P10]]}, [[T]]-cpp-output, (host-[[T]]) |
140 | // ASM2-DAG: [[P12:[0-9]+]]: compiler, {[[P11]]}, ir, (host-[[T]]) |
141 | // ASM2-DAG: [[P13:[0-9]+]]: backend, {[[P12]]}, assembler, (host-[[T]]) |
142 | |
143 | // |
144 | // Test single gpu architecture with complete compilation in host-only |
145 | // compilation mode. |
146 | // |
147 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
148 | // RUN: --cuda-gpu-arch=sm_30 %s --cuda-host-only 2>&1 \ |
149 | // RUN: | FileCheck -check-prefixes=HBIN,HBIN_NV %s |
150 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
151 | // RUN: --cuda-gpu-arch=gfx803 %s --cuda-host-only 2>&1 \ |
152 | // RUN: | FileCheck -check-prefixes=HBIN,HBIN_AMD %s |
153 | // HBIN_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (host-[[T]]) |
154 | // HBIN_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (host-[[T]]) |
155 | // HBIN-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]]) |
156 | // HBIN-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]]) |
157 | // HBIN-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]]) |
158 | // HBIN-DAG: [[P4:[0-9]+]]: assembler, {[[P3]]}, object, (host-[[T]]) |
159 | // HBIN-DAG: [[P5:[0-9]+]]: linker, {[[P4]]}, image, (host-[[T]]) |
160 | // HBIN-NOT: device |
161 | // |
162 | // Test single gpu architecture up to the assemble phase in host-only |
163 | // compilation mode. |
164 | // |
165 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
166 | // RUN: --cuda-gpu-arch=sm_30 %s --cuda-host-only -S 2>&1 \ |
167 | // RUN: | FileCheck -check-prefixes=HASM,HASM_NV %s |
168 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
169 | // RUN: --cuda-gpu-arch=gfx803 %s --cuda-host-only -S 2>&1 \ |
170 | // RUN: | FileCheck -check-prefixes=HASM,HASM_AMD %s |
171 | // HASM_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (host-[[T]]) |
172 | // HASM_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (host-[[T]]) |
173 | // HASM-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]]) |
174 | // HASM-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]]) |
175 | // HASM-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]]) |
176 | // HASM-NOT: device |
177 | |
178 | // |
179 | // Test two gpu architectures with complete compilation in host-only |
180 | // compilation mode. |
181 | // |
182 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
183 | // RUN: --cuda-gpu-arch=sm_30 --cuda-gpu-arch=sm_35 %s --cuda-host-only 2>&1 \ |
184 | // RUN: | FileCheck -check-prefixes=HBIN2,HBIN2_NV %s |
185 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
186 | // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s --cuda-host-only 2>&1 \ |
187 | // RUN: | FileCheck -check-prefixes=HBIN2,HBIN2_AMD %s |
188 | // HBIN2_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (host-[[T]]) |
189 | // HBIN2_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (host-[[T]]) |
190 | // HBIN2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]]) |
191 | // HBIN2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]]) |
192 | // HBIN2-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]]) |
193 | // HBIN2-DAG: [[P4:[0-9]+]]: assembler, {[[P3]]}, object, (host-[[T]]) |
194 | // HBIN2-DAG: [[P5:[0-9]+]]: linker, {[[P4]]}, image, (host-[[T]]) |
195 | // HBIN2-NOT: device |
196 | |
197 | // |
198 | // Test two gpu architectures up to the assemble phase in host-only |
199 | // compilation mode. |
200 | // |
201 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
202 | // RUN: --cuda-gpu-arch=sm_30 --cuda-gpu-arch=sm_35 %s --cuda-host-only -S \ |
203 | // RUN: 2>&1 | FileCheck -check-prefixes=HASM2,HASM2_NV %s |
204 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
205 | // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s --cuda-host-only -S \ |
206 | // RUN: 2>&1 | FileCheck -check-prefixes=HASM2,HASM2_AMD %s |
207 | // HASM2_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (host-[[T]]) |
208 | // HASM2_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (host-[[T]]) |
209 | // HASM2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (host-[[T]]) |
210 | // HASM2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (host-[[T]]) |
211 | // HASM2-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (host-[[T]]) |
212 | // HASM2-NOT: device |
213 | |
214 | // |
215 | // Test single gpu architecture with complete compilation in device-only |
216 | // compilation mode. |
217 | // |
218 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
219 | // RUN: --cuda-gpu-arch=sm_30 %s --cuda-device-only 2>&1 \ |
220 | // RUN: | FileCheck -check-prefixes=DBIN,DBIN_NV %s |
221 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
222 | // RUN: --cuda-gpu-arch=gfx803 %s --cuda-device-only 2>&1 \ |
223 | // RUN: | FileCheck -check-prefixes=DBIN,DBIN_AMD %s |
224 | // DBIN_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (device-[[T]], [[ARCH:sm_30]]) |
225 | // DBIN_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (device-[[T]], [[ARCH:gfx803]]) |
226 | // DBIN-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH]]) |
227 | // DBIN-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (device-[[T]], [[ARCH]]) |
228 | // DBIN_NV-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (device-[[T]], [[ARCH]]) |
229 | // DBIN_NV-DAG: [[P4:[0-9]+]]: assembler, {[[P3]]}, object, (device-[[T]], [[ARCH]]) |
230 | // DBIN_NV-DAG: [[P5:[0-9]+]]: offload, "device-[[T]] (nvptx64-nvidia-cuda:[[ARCH]])" {[[P4]]}, object |
231 | // DBIN-NOT: host |
232 | // |
233 | // Test single gpu architecture up to the assemble phase in device-only |
234 | // compilation mode. |
235 | // |
236 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
237 | // RUN: --cuda-gpu-arch=sm_30 %s --cuda-device-only -S 2>&1 \ |
238 | // RUN: | FileCheck -check-prefixes=DASM,DASM_NV %s |
239 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
240 | // RUN: --cuda-gpu-arch=gfx803 %s --cuda-device-only -S 2>&1 \ |
241 | // RUN: | FileCheck -check-prefixes=DASM,DASM_AMD %s |
242 | // DASM_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (device-[[T]], [[ARCH:sm_30]]) |
243 | // DASM_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (device-[[T]], [[ARCH:gfx803]]) |
244 | // DASM-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH]]) |
245 | // DASM-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (device-[[T]], [[ARCH]]) |
246 | // DASM_NV-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (device-[[T]], [[ARCH]]) |
247 | // DASM_NV-DAG: [[P4:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE:nvptx64-nvidia-cuda|amdgcn-amd-amdhsa]]:[[ARCH]])" {[[P3]]}, assembler |
248 | // DASM-NOT: host |
249 | |
250 | // |
251 | // Test two gpu architectures with complete compilation in device-only |
252 | // compilation mode. |
253 | // |
254 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
255 | // RUN: --cuda-gpu-arch=sm_30 --cuda-gpu-arch=sm_35 %s --cuda-device-only 2>&1 \ |
256 | // RUN: | FileCheck -check-prefixes=DBIN2,DBIN2_NV %s |
257 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
258 | // RUN: --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s --cuda-device-only \ |
259 | // RUN: 2>&1 | FileCheck -check-prefixes=DBIN2,DBIN2_AMD %s |
260 | // DBIN2_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (device-[[T]], [[ARCH:sm_30]]) |
261 | // DBIN2_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (device-[[T]], [[ARCH:gfx803]]) |
262 | // DBIN2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH]]) |
263 | // DBIN2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (device-[[T]], [[ARCH]]) |
264 | // DBIN2_NV-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (device-[[T]], [[ARCH]]) |
265 | // DBIN2_NV-DAG: [[P4:[0-9]+]]: assembler, {[[P3]]}, object, (device-[[T]], [[ARCH]]) |
266 | // DBIN2_NV-DAG: [[P5:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE:nvptx64-nvidia-cuda]]:[[ARCH]])" {[[P4]]}, object |
267 | // DBIN2-DAG: [[P6:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (device-[[T]], [[ARCH2:sm_35|gfx900]]) |
268 | // DBIN2-DAG: [[P7:[0-9]+]]: preprocessor, {[[P6]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH2]]) |
269 | // DBIN2-DAG: [[P8:[0-9]+]]: compiler, {[[P7]]}, ir, (device-[[T]], [[ARCH2]]) |
270 | // DBIN2_NV-DAG: [[P9:[0-9]+]]: backend, {[[P8]]}, assembler, (device-[[T]], [[ARCH2]]) |
271 | // DBIN2_NV-DAG: [[P10:[0-9]+]]: assembler, {[[P9]]}, object, (device-[[T]], [[ARCH2]]) |
272 | // DBIN2_NV-DAG: [[P11:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE]]:[[ARCH2]])" {[[P10]]}, object |
273 | // DBIN2-NOT: host |
274 | // |
275 | // Test two gpu architectures up to the assemble phase in device-only |
276 | // compilation mode. |
277 | // |
278 | // RUN: %clang -target powerpc64le-ibm-linux-gnu -ccc-print-phases \ |
279 | // RUN: --cuda-gpu-arch=sm_30 --cuda-gpu-arch=sm_35 %s --cuda-device-only -S \ |
280 | // RUN: 2>&1 | FileCheck -check-prefixes=DASM2,DASM2_NV %s |
281 | // RUN: %clang -x hip -target powerpc64le-ibm-linux-gnu \ |
282 | // RUN: -ccc-print-phases --cuda-gpu-arch=gfx803 --cuda-gpu-arch=gfx900 %s \ |
283 | // RUN: --cuda-device-only -S 2>&1 \ |
284 | // RUN: | FileCheck -check-prefixes=DASM2,DASM2_AMD %s |
285 | // DASM2_NV-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:cuda]], (device-[[T]], [[ARCH:sm_30]]) |
286 | // DASM2_AMD-DAG: [[P0:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T:hip]], (device-[[T]], [[ARCH:gfx803]]) |
287 | // DASM2-DAG: [[P1:[0-9]+]]: preprocessor, {[[P0]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH]]) |
288 | // DASM2-DAG: [[P2:[0-9]+]]: compiler, {[[P1]]}, ir, (device-[[T]], [[ARCH]]) |
289 | // DASM2_NV-DAG: [[P3:[0-9]+]]: backend, {[[P2]]}, assembler, (device-[[T]], [[ARCH]]) |
290 | // DASM2_NV-DAG: [[P4:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE:nvptx64-nvidia-cuda|amdgcn-amd-amdhsa]]:[[ARCH]])" {[[P3]]}, assembler |
291 | // DASM2-DAG: [[P5:[0-9]+]]: input, "{{.*}}cuda-phases.cu", [[T]], (device-[[T]], [[ARCH2:sm_35|gfx900]]) |
292 | // DASM2-DAG: [[P6:[0-9]+]]: preprocessor, {[[P5]]}, [[T]]-cpp-output, (device-[[T]], [[ARCH2]]) |
293 | // DASM2-DAG: [[P7:[0-9]+]]: compiler, {[[P6]]}, ir, (device-[[T]], [[ARCH2]]) |
294 | // DASM2_NV-DAG: [[P8:[0-9]+]]: backend, {[[P7]]}, assembler, (device-[[T]], [[ARCH2]]) |
295 | // DASM2_NV-DAG: [[P9:[0-9]+]]: offload, "device-[[T]] ([[TRIPLE]]:[[ARCH2]])" {[[P8]]}, assembler |
296 | // DASM2-NOT: host |
297 | |