Neko  0.9.99
A portable framework for high-order spectral element flow simulations
filters.c
Go to the documentation of this file.
1 /*
2  Copyright (c) 2021-2024, The Neko Authors
3  All rights reserved.
4 
5  Redistribution and use in source and binary forms, with or without
6  modification, are permitted provided that the following conditions
7  are met:
8 
9  * Redistributions of source code must retain the above copyright
10  notice, this list of conditions and the following disclaimer.
11 
12  * Redistributions in binary form must reproduce the above
13  copyright notice, this list of conditions and the following
14  disclaimer in the documentation and/or other materials provided
15  with the distribution.
16 
17  * Neither the name of the authors nor the names of its
18  contributors may be used to endorse or promote products derived
19  from this software without specific prior written permission.
20 
21  THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
22  "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
23  LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
24  FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
25  COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
26  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
27  BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
28  LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
29  CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
30  LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
31  ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
32  POSSIBILITY OF SUCH DAMAGE.
33 */
34 
35 #ifdef __APPLE__
36  #include <OpenCL/cl.h>
37 #else
38  #include <CL/cl.h>
39 #endif
40 
41 #include <device/device_config.h>
42 #include <device/opencl/check.h>
43 #include <device/opencl/jit.h>
44 #include <device/opencl/prgm_lib.h>
45 #include <stdio.h>
46 #include <stdlib.h>
47 
48 #include "filter_kernels.cl.h"
49 
56 void opencl_smooth_step(void* x, real* edge0, real* edge1, int* n) {
57  cl_int err;
58 
59  if (filter_program == NULL)
60  opencl_kernel_jit(filter_kernels, (cl_program*)&filter_program);
61 
62  cl_kernel kernel =
63  clCreateKernel(filter_program, "smooth_step_kernel", &err);
64  CL_CHECK(err);
65 
66  CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), (void*)&x));
67  CL_CHECK(clSetKernelArg(kernel, 1, sizeof(real), edge0));
68  CL_CHECK(clSetKernelArg(kernel, 2, sizeof(real), edge1));
69  CL_CHECK(clSetKernelArg(kernel, 3, sizeof(int), n));
70 
71  const int nb = ((*n) + 256 - 1) / 256;
72  const size_t global_item_size = 256 * nb;
73  const size_t local_item_size = 256;
74 
75  CL_CHECK(clEnqueueNDRangeKernel(
76  (cl_command_queue)glb_cmd_queue, kernel, 1, NULL, &global_item_size,
77  &local_item_size, 0, NULL, NULL));
78 }
79 
93  void* x, real* edge, real* left, real* right, int* n) {
94  cl_int err;
95 
96  if (filter_program == NULL)
97  opencl_kernel_jit(filter_kernels, (cl_program*)&filter_program);
98 
99  cl_kernel kernel =
100  clCreateKernel(filter_program, "step_function_kernel", &err);
101  CL_CHECK(err);
102 
103  CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), (void*)&x));
104  CL_CHECK(clSetKernelArg(kernel, 1, sizeof(real), edge));
105  CL_CHECK(clSetKernelArg(kernel, 2, sizeof(real), left));
106  CL_CHECK(clSetKernelArg(kernel, 3, sizeof(real), right));
107  CL_CHECK(clSetKernelArg(kernel, 4, sizeof(int), n));
108 
109  const int nb = ((*n) + 256 - 1) / 256;
110  const size_t global_item_size = 256 * nb;
111  const size_t local_item_size = 256;
112 
113  CL_CHECK(clEnqueueNDRangeKernel(
114  (cl_command_queue)glb_cmd_queue, kernel, 1, NULL, &global_item_size,
115  &local_item_size, 0, NULL, NULL));
116 }
117 
126 void opencl_permeability(void* x, real* k_0, real* k_1, real* q, int* n) {
127  cl_int err;
128 
129  if (filter_program == NULL)
130  opencl_kernel_jit(filter_kernels, (cl_program*)&filter_program);
131 
132  cl_kernel kernel =
133  clCreateKernel(filter_program, "permeability_kernel", &err);
134  CL_CHECK(err);
135 
136  CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), (void*)&x));
137  CL_CHECK(clSetKernelArg(kernel, 1, sizeof(real), k_0));
138  CL_CHECK(clSetKernelArg(kernel, 2, sizeof(real), k_1));
139  CL_CHECK(clSetKernelArg(kernel, 3, sizeof(real), q));
140  CL_CHECK(clSetKernelArg(kernel, 4, sizeof(int), n));
141 
142  const int nb = ((*n) + 256 - 1) / 256;
143  const size_t global_item_size = 256 * nb;
144  const size_t local_item_size = 256;
145 
146  CL_CHECK(clEnqueueNDRangeKernel(
147  (cl_command_queue)glb_cmd_queue, kernel, 1, NULL, &global_item_size,
148  &local_item_size, 0, NULL, NULL));
149 }
__global__ void const T *__restrict__ x
Definition: cdtp_kernel.h:106
double real
Definition: device_config.h:12
void * glb_cmd_queue
void opencl_step_function(void *x, real *edge, real *left, real *right, int *n)
Definition: filters.c:92
void opencl_permeability(void *x, real *k_0, real *k_1, real *q, int *n)
Definition: filters.c:126
void opencl_smooth_step(void *x, real *edge0, real *edge1, int *n)
Definition: filters.c:56
void opencl_kernel_jit(const char *kernel, cl_program *program)
Definition: jit.c:50
#define CL_CHECK(err)
Definition: check.h:12
void * filter_program