You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

249 lines
8.2 KiB

//
// Copyright (c) 2017 The Khronos Group Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
#include "harness/compat.h"
#include <stdio.h>
#include <string.h>
#include <sys/types.h>
#include <sys/stat.h>
#include "procs.h"
static const char *fmax_kernel_code =
"__kernel void test_fmax(__global float *srcA, __global float *srcB, __global float *dst)\n"
"{\n"
" int tid = get_global_id(0);\n"
" dst[tid] = fmax(srcA[tid], srcB[tid]);\n"
"}\n";
static const char *fmax2_kernel_code =
"__kernel void test_fmax2(__global float2 *srcA, __global float *srcB, __global float2 *dst)\n"
"{\n"
" int tid = get_global_id(0);\n"
" dst[tid] = fmax(srcA[tid], srcB[tid]);\n"
"}\n";
static const char *fmax4_kernel_code =
"__kernel void test_fmax4(__global float4 *srcA, __global float *srcB, __global float4 *dst)\n"
"{\n"
" int tid = get_global_id(0);\n"
" dst[tid] = fmax(srcA[tid], srcB[tid]);\n"
"}\n";
static const char *fmax8_kernel_code =
"__kernel void test_fmax8(__global float8 *srcA, __global float *srcB, __global float8 *dst)\n"
"{\n"
" int tid = get_global_id(0);\n"
" dst[tid] = fmax(srcA[tid], srcB[tid]);\n"
"}\n";
static const char *fmax16_kernel_code =
"__kernel void test_fmax16(__global float16 *srcA, __global float *srcB, __global float16 *dst)\n"
"{\n"
" int tid = get_global_id(0);\n"
" dst[tid] = fmax(srcA[tid], srcB[tid]);\n"
"}\n";
static const char *fmax3_kernel_code =
"__kernel void test_fmax3(__global float *srcA, __global float *srcB, __global float *dst)\n"
"{\n"
" int tid = get_global_id(0);\n"
" vstore3(fmax(vload3(tid,srcA), srcB[tid]),tid,dst);\n"
"}\n";
static int
verify_fmax(float *inptrA, float *inptrB, float *outptr, int n, int veclen)
{
float r;
int i, j;
for (i=0; i<n; ) {
int ii = i/veclen;
for (j=0; j<veclen && i<n; ++j, ++i) {
r = (inptrA[i] >= inptrB[ii]) ? inptrA[i] : inptrB[ii];
if (r != outptr[i]) {
log_info("Verify noted discrepancy at %d (of %d) (vec %d, pos %d)\n",
i,n,ii,j);
log_info("SHould be %f, is %f\n", r, outptr[i]);
log_info("Taking max of (%f,%f)\n", inptrA[i], inptrB[i]);
return -1;
}
}
}
return 0;
}
int
test_fmaxf(cl_device_id device, cl_context context, cl_command_queue queue, int n_elems)
{
cl_mem streams[3];
cl_float *input_ptr[2], *output_ptr, *p;
cl_program *program;
cl_kernel *kernel;
void *values[3];
size_t threads[1];
int num_elements;
int err;
int i;
MTdata d;
program = (cl_program*)malloc(sizeof(cl_program)*kTotalVecCount);
kernel = (cl_kernel*)malloc(sizeof(cl_kernel)*kTotalVecCount);
num_elements = n_elems * (1 << (kTotalVecCount-1));
input_ptr[0] = (cl_float*)malloc(sizeof(cl_float) * num_elements);
input_ptr[1] = (cl_float*)malloc(sizeof(cl_float) * num_elements);
output_ptr = (cl_float*)malloc(sizeof(cl_float) * num_elements);
streams[0] = clCreateBuffer(context, CL_MEM_READ_WRITE,
sizeof(cl_float) * num_elements, NULL, NULL);
if (!streams[0])
{
log_error("clCreateBuffer failed\n");
return -1;
}
streams[1] =
clCreateBuffer(context, CL_MEM_READ_WRITE,
sizeof(cl_float) * num_elements, NULL, NULL);
if (!streams[1])
{
log_error("clCreateBuffer failed\n");
return -1;
}
streams[2] =
clCreateBuffer(context, CL_MEM_READ_WRITE,
sizeof(cl_float) * num_elements, NULL, NULL);
if (!streams[2])
{
log_error("clCreateBuffer failed\n");
return -1;
}
d = init_genrand( gRandomSeed );
p = input_ptr[0];
for (i=0; i<num_elements; i++)
{
p[i] = get_random_float(-0x20000000, 0x20000000, d);
}
p = input_ptr[1];
for (i=0; i<num_elements; i++)
{
p[i] = get_random_float(-0x20000000, 0x20000000, d);
}
free_mtdata(d); d = NULL;
err = clEnqueueWriteBuffer( queue, streams[0], true, 0, sizeof(cl_float)*num_elements,
(void *)input_ptr[0], 0, NULL, NULL );
if (err != CL_SUCCESS)
{
log_error("clWriteArray failed\n");
return -1;
}
err = clEnqueueWriteBuffer( queue, streams[1], true, 0, sizeof(cl_float)*num_elements,
(void *)input_ptr[1], 0, NULL, NULL );
if (err != CL_SUCCESS)
{
log_error("clWriteArray failed\n");
return -1;
}
err = create_single_kernel_helper( context, &program[0], &kernel[0], 1, &fmax_kernel_code, "test_fmax" );
if (err)
return -1;
err = create_single_kernel_helper( context, &program[1], &kernel[1], 1, &fmax2_kernel_code, "test_fmax2" );
if (err)
return -1;
err = create_single_kernel_helper( context, &program[2], &kernel[2], 1, &fmax4_kernel_code, "test_fmax4" );
if (err)
return -1;
err = create_single_kernel_helper( context, &program[3], &kernel[3], 1, &fmax8_kernel_code, "test_fmax8" );
if (err)
return -1;
err = create_single_kernel_helper( context, &program[4], &kernel[4], 1, &fmax16_kernel_code, "test_fmax16" );
if (err)
return -1;
err = create_single_kernel_helper( context, &program[5], &kernel[5], 1, &fmax3_kernel_code, "test_fmax3" );
if (err)
return -1;
values[0] = streams[0];
values[1] = streams[1];
values[2] = streams[2];
for (i=0; i < kTotalVecCount; i++)
{
err = clSetKernelArg(kernel[i], 0, sizeof streams[0], &streams[0] );
err |= clSetKernelArg(kernel[i], 1, sizeof streams[1], &streams[1] );
err |= clSetKernelArg(kernel[i], 2, sizeof streams[2], &streams[2] );
if (err != CL_SUCCESS)
{
log_error("clSetKernelArgs failed\n");
return -1;
}
}
threads[0] = (size_t)n_elems;
for (i=0; i < kTotalVecCount; i++)
{
err = clEnqueueNDRangeKernel( queue, kernel[i], 1, NULL, threads, NULL, 0, NULL, NULL );
if (err != CL_SUCCESS)
{
log_error("clEnqueueNDRangeKernel failed\n");
return -1;
}
err = clEnqueueReadBuffer(queue, streams[2], true, 0, sizeof(cl_float)*num_elements,
output_ptr, 0, NULL, NULL);
if (err != CL_SUCCESS)
{
log_error("clEnqueueReadBuffer failed\n");
return -1;
}
if (verify_fmax(input_ptr[0], input_ptr[1], output_ptr, n_elems*((g_arrVecSizes[i])), (g_arrVecSizes[i])))
{
log_error("FMAX float%d,float test failed\n", (g_arrVecSizes[i]));
err = -1;
}
else
{
log_info("FMAX float%d,float test passed\n", (g_arrVecSizes[i]));
err = 0;
}
if (err)
break;
}
clReleaseMemObject(streams[0]);
clReleaseMemObject(streams[1]);
clReleaseMemObject(streams[2]);
for (i=0; i < kTotalVecCount; i++)
{
clReleaseKernel(kernel[i]);
clReleaseProgram(program[i]);
}
free(program);
free(kernel);
free(input_ptr[0]);
free(input_ptr[1]);
free(output_ptr);
return err;
}