1 //
2 // Copyright (c) 2017 The Khronos Group Inc.
3 //
4 // Licensed under the Apache License, Version 2.0 (the "License");
5 // you may not use this file except in compliance with the License.
6 // You may obtain a copy of the License at
7 //
8 // http://www.apache.org/licenses/LICENSE-2.0
9 //
10 // Unless required by applicable law or agreed to in writing, software
11 // distributed under the License is distributed on an "AS IS" BASIS,
12 // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 // See the License for the specific language governing permissions and
14 // limitations under the License.
15 //
16 #include "harness/compat.h"
17
18 #include <stdio.h>
19 #include <string.h>
20 #include <sys/types.h>
21 #include <sys/stat.h>
22
23 #include "procs.h"
24
25 static const char *fmin_kernel_code =
26 "__kernel void test_fmin(__global float *srcA, __global float *srcB, __global float *dst)\n"
27 "{\n"
28 " int tid = get_global_id(0);\n"
29 "\n"
30 " dst[tid] = fmin(srcA[tid], srcB[tid]);\n"
31 "}\n";
32
33 static const char *fmin2_kernel_code =
34 "__kernel void test_fmin2(__global float2 *srcA, __global float2 *srcB, __global float2 *dst)\n"
35 "{\n"
36 " int tid = get_global_id(0);\n"
37 "\n"
38 " dst[tid] = fmin(srcA[tid], srcB[tid]);\n"
39 "}\n";
40
41 static const char *fmin4_kernel_code =
42 "__kernel void test_fmin4(__global float4 *srcA, __global float4 *srcB, __global float4 *dst)\n"
43 "{\n"
44 " int tid = get_global_id(0);\n"
45 "\n"
46 " dst[tid] = fmin(srcA[tid], srcB[tid]);\n"
47 "}\n";
48
49 static const char *fmin8_kernel_code =
50 "__kernel void test_fmin8(__global float8 *srcA, __global float8 *srcB, __global float8 *dst)\n"
51 "{\n"
52 " int tid = get_global_id(0);\n"
53 "\n"
54 " dst[tid] = fmin(srcA[tid], srcB[tid]);\n"
55 "}\n";
56
57 static const char *fmin16_kernel_code =
58 "__kernel void test_fmin16(__global float16 *srcA, __global float16 *srcB, __global float16 *dst)\n"
59 "{\n"
60 " int tid = get_global_id(0);\n"
61 "\n"
62 " dst[tid] = fmin(srcA[tid], srcB[tid]);\n"
63 "}\n";
64
65
66 static const char *fmin3_kernel_code =
67 "__kernel void test_fmin3(__global float *srcA, __global float *srcB, __global float *dst)\n"
68 "{\n"
69 " int tid = get_global_id(0);\n"
70 " vstore3(fmin(vload3(tid,srcA), vload3(tid,srcB)),tid,dst);\n"
71 "}\n";
72
73 int
verify_fmin(float * inptrA,float * inptrB,float * outptr,int n)74 verify_fmin(float *inptrA, float *inptrB, float *outptr, int n)
75 {
76 float r;
77 int i;
78
79 for (i=0; i<n; i++)
80 {
81 r = (inptrA[i] > inptrB[i]) ? inptrB[i] : inptrA[i];
82 if (r != outptr[i])
83 return -1;
84 }
85
86 return 0;
87 }
88
89 int
test_fmin(cl_device_id device,cl_context context,cl_command_queue queue,int n_elems)90 test_fmin(cl_device_id device, cl_context context, cl_command_queue queue, int n_elems)
91 {
92 cl_mem streams[3];
93 cl_float *input_ptr[2], *output_ptr, *p;
94 cl_program *program;
95 cl_kernel *kernel;
96 void *values[3];
97 size_t threads[1];
98 int num_elements;
99 int err;
100 int i;
101 MTdata d;
102
103 program = (cl_program*)malloc(sizeof(cl_program)*kTotalVecCount);
104 kernel = (cl_kernel*)malloc(sizeof(cl_kernel)*kTotalVecCount);
105
106 num_elements = n_elems * (1 << (kTotalVecCount-1));;
107
108 input_ptr[0] = (cl_float*)malloc(sizeof(cl_float) * num_elements);
109 input_ptr[1] = (cl_float*)malloc(sizeof(cl_float) * num_elements);
110 output_ptr = (cl_float*)malloc(sizeof(cl_float) * num_elements);
111 streams[0] = clCreateBuffer(context, CL_MEM_READ_WRITE,
112 sizeof(cl_float) * num_elements, NULL, NULL);
113 if (!streams[0])
114 {
115 log_error("clCreateBuffer failed\n");
116 return -1;
117 }
118 streams[1] = clCreateBuffer(context, CL_MEM_READ_WRITE,
119 sizeof(cl_float) * num_elements, NULL, NULL);
120 if (!streams[1])
121 {
122 log_error("clCreateBuffer failed\n");
123 return -1;
124 }
125
126 streams[2] = clCreateBuffer(context, CL_MEM_READ_WRITE,
127 sizeof(cl_float) * num_elements, NULL, NULL);
128 if (!streams[2])
129 {
130 log_error("clCreateBuffer failed\n");
131 return -1;
132 }
133
134 d = init_genrand( gRandomSeed );
135 p = input_ptr[0];
136 for (i=0; i<num_elements; i++)
137 {
138 p[i] = get_random_float(-0x20000000, 0x20000000, d);
139 }
140 p = input_ptr[1];
141 for (i=0; i<num_elements; i++)
142 {
143 p[i] = get_random_float(-0x20000000, 0x20000000, d);
144 }
145 free_mtdata(d); d = NULL;
146
147 err = clEnqueueWriteBuffer( queue, streams[0], true, 0, sizeof(cl_float)*num_elements,
148 (void *)input_ptr[0], 0, NULL, NULL );
149 if (err != CL_SUCCESS)
150 {
151 log_error("clWriteArray failed\n");
152 return -1;
153 }
154 err = clEnqueueWriteBuffer( queue, streams[1], true, 0, sizeof(cl_float)*num_elements,
155 (void *)input_ptr[1], 0, NULL, NULL );
156 if (err != CL_SUCCESS)
157 {
158 log_error("clWriteArray failed\n");
159 return -1;
160 }
161
162 err = create_single_kernel_helper( context, &program[0], &kernel[0], 1, &fmin_kernel_code, "test_fmin" );
163 if (err)
164 return -1;
165 err = create_single_kernel_helper( context, &program[1], &kernel[1], 1, &fmin2_kernel_code, "test_fmin2" );
166 if (err)
167 return -1;
168 err = create_single_kernel_helper( context, &program[2], &kernel[2], 1, &fmin4_kernel_code, "test_fmin4" );
169 if (err)
170 return -1;
171 err = create_single_kernel_helper( context, &program[3], &kernel[3], 1, &fmin8_kernel_code, "test_fmin8" );
172 if (err)
173 return -1;
174 err = create_single_kernel_helper( context, &program[4], &kernel[4], 1, &fmin16_kernel_code, "test_fmin16" );
175 if (err)
176 return -1;
177 err = create_single_kernel_helper( context, &program[5], &kernel[5], 1, &fmin3_kernel_code, "test_fmin3" );
178 if (err)
179 return -1;
180
181 values[0] = streams[0];
182 values[1] = streams[1];
183 values[2] = streams[2];
184 for (i=0; i<kTotalVecCount; i++)
185 {
186 err = clSetKernelArg(kernel[i], 0, sizeof streams[0], &streams[0] );
187 err |= clSetKernelArg(kernel[i], 1, sizeof streams[1], &streams[1] );
188 err |= clSetKernelArg(kernel[i], 2, sizeof streams[2], &streams[2] );
189 if (err != CL_SUCCESS)
190 {
191 log_error("clSetKernelArgs failed\n");
192 return -1;
193 }
194 }
195
196 threads[0] = (size_t)n_elems;
197 for (i=0; i<kTotalVecCount; i++)
198 {
199 err = clEnqueueNDRangeKernel( queue, kernel[i], 1, NULL, threads, NULL, 0, NULL, NULL );
200 if (err != CL_SUCCESS)
201 {
202 log_error("clEnqueueNDRangeKernel failed\n");
203 return -1;
204 }
205
206 err = clEnqueueReadBuffer( queue, streams[2], true, 0, sizeof(cl_float)*num_elements, (void *)output_ptr, 0, NULL, NULL );
207 if (err != CL_SUCCESS)
208 {
209 log_error("clEnqueueReadBuffer failed\n");
210 return -1;
211 }
212
213 if (verify_fmin(input_ptr[0], input_ptr[1], output_ptr, n_elems*((g_arrVecSizes[i]))))
214 {
215 log_error("FMIN float%d test failed\n", (g_arrVecSizes[i]));
216 err = -1;
217 }
218 else
219 {
220 log_info("FMIN float%d test passed\n", (g_arrVecSizes[i]));
221 err = 0;
222 }
223 }
224
225 clReleaseMemObject(streams[0]);
226 clReleaseMemObject(streams[1]);
227 clReleaseMemObject(streams[2]);
228 for (i=0; i<kTotalVecCount; i++)
229 {
230 clReleaseKernel(kernel[i]);
231 clReleaseProgram(program[i]);
232 }
233 free(program);
234 free(kernel);
235 free(input_ptr[0]);
236 free(input_ptr[1]);
237 free(output_ptr);
238
239 return err;
240 }
241
242
243