Neko 1.99.2
A portable framework for high-order spectral element flow simulations
Loading...
Searching...
No Matches
compressible_ops_update.c
Go to the documentation of this file.
1/*
2 Copyright (c) 2025, 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 <stdio.h>
42#include <device/device_config.h>
43#include <device/opencl/jit.h>
44#include <device/opencl/prgm_lib.h>
45#include <device/opencl/check.h>
46#include "compressible_ops_update_kernel.cl.h"
47
48void opencl_update_uvw(void *u, void *v, void *w,
49 void *m_x, void *m_y, void *m_z,
50 void *rho, int n) {
51
52 cl_int err;
53
54 if (compressible_ops_update_program == NULL)
55 opencl_kernel_jit(compressible_ops_update_kernel,
56 (cl_program *) &compressible_ops_update_program);
57
58 cl_kernel kernel =
59 clCreateKernel((cl_program) compressible_ops_update_program,
60 "update_uvw_kernel", &err);
61 CL_CHECK(err);
62
63 CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), (void *) &u));
64 CL_CHECK(clSetKernelArg(kernel, 1, sizeof(cl_mem), (void *) &v));
65 CL_CHECK(clSetKernelArg(kernel, 2, sizeof(cl_mem), (void *) &w));
66 CL_CHECK(clSetKernelArg(kernel, 3, sizeof(cl_mem), (void *) &m_x));
67 CL_CHECK(clSetKernelArg(kernel, 4, sizeof(cl_mem), (void *) &m_y));
68 CL_CHECK(clSetKernelArg(kernel, 5, sizeof(cl_mem), (void *) &m_z));
69 CL_CHECK(clSetKernelArg(kernel, 6, sizeof(cl_mem), (void *) &rho));
70 CL_CHECK(clSetKernelArg(kernel, 7, sizeof(int), &n));
71
72 const int nb = (n + 256 - 1) / 256;
73 const size_t global_item_size = 256 * nb;
74 const size_t local_item_size = 256;
75
76 CL_CHECK(clEnqueueNDRangeKernel((cl_command_queue) glb_cmd_queue, kernel, 1,
77 NULL, &global_item_size, &local_item_size,
78 0, NULL, NULL));
79 CL_CHECK(clReleaseKernel(kernel));
80}
81
82void opencl_update_mxyz_p_ruvw(void *m_x, void *m_y, void *m_z,
83 void *p, void *ruvw,
84 void *u, void *v, void *w, void *E,
85 void *rho, real gamma, int n) {
86 cl_int err;
87
88 if (compressible_ops_update_program == NULL)
89 opencl_kernel_jit(compressible_ops_update_kernel,
90 (cl_program *) &compressible_ops_update_program);
91
92 cl_kernel kernel =
93 clCreateKernel((cl_program) compressible_ops_update_program,
94 "update_mxyz_p_ruvw_kernel", &err);
95 CL_CHECK(err);
96
97 CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), (void *) &m_x));
98 CL_CHECK(clSetKernelArg(kernel, 1, sizeof(cl_mem), (void *) &m_y));
99 CL_CHECK(clSetKernelArg(kernel, 2, sizeof(cl_mem), (void *) &m_z));
100 CL_CHECK(clSetKernelArg(kernel, 3, sizeof(cl_mem), (void *) &p));
101 CL_CHECK(clSetKernelArg(kernel, 4, sizeof(cl_mem), (void *) &ruvw));
102 CL_CHECK(clSetKernelArg(kernel, 5, sizeof(cl_mem), (void *) &u));
103 CL_CHECK(clSetKernelArg(kernel, 6, sizeof(cl_mem), (void *) &v));
104 CL_CHECK(clSetKernelArg(kernel, 7, sizeof(cl_mem), (void *) &w));
105 CL_CHECK(clSetKernelArg(kernel, 8, sizeof(cl_mem), (void *) &E));
106 CL_CHECK(clSetKernelArg(kernel, 9, sizeof(cl_mem), (void *) &rho));
107 CL_CHECK(clSetKernelArg(kernel, 10, sizeof(real), &gamma));
108 CL_CHECK(clSetKernelArg(kernel, 11, sizeof(int), &n));
109
110 const int nb = (n + 256 - 1) / 256;
111 const size_t global_item_size = 256 * nb;
112 const size_t local_item_size = 256;
113
114 CL_CHECK(clEnqueueNDRangeKernel((cl_command_queue) glb_cmd_queue, kernel, 1,
115 NULL, &global_item_size, &local_item_size,
116 0, NULL, NULL));
117 CL_CHECK(clReleaseKernel(kernel));
118}
119
120void opencl_update_e(void *E, void *p, void *ruvw, real gamma, int n) {
121
122 cl_int err;
123
124 if (compressible_ops_update_program == NULL)
125 opencl_kernel_jit(compressible_ops_update_kernel,
126 (cl_program *) &compressible_ops_update_program);
127
128 cl_kernel kernel =
129 clCreateKernel((cl_program) compressible_ops_update_program,
130 "update_e_kernel", &err);
131 CL_CHECK(err);
132
133 CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), (void *) &E));
134 CL_CHECK(clSetKernelArg(kernel, 1, sizeof(cl_mem), (void *) &p));
135 CL_CHECK(clSetKernelArg(kernel, 2, sizeof(cl_mem), (void *) &ruvw));
136 CL_CHECK(clSetKernelArg(kernel, 3, sizeof(real), &gamma));
137 CL_CHECK(clSetKernelArg(kernel, 4, sizeof(int), &n));
138
139 const int nb = (n + 256 - 1) / 256;
140 const size_t global_item_size = 256 * nb;
141 const size_t local_item_size = 256;
142
143 CL_CHECK(clEnqueueNDRangeKernel((cl_command_queue) glb_cmd_queue, kernel, 1,
144 NULL, &global_item_size, &local_item_size,
145 0, NULL, NULL));
146 CL_CHECK(clReleaseKernel(kernel));
147}
opencl_update_uvw
void opencl_update_uvw(void *u, void *v, void *w, void *m_x, void *m_y, void *m_z, void *rho, int n)
Definition compressible_ops_update.c:48
opencl_update_e
void opencl_update_e(void *E, void *p, void *ruvw, real gamma, int n)
Definition compressible_ops_update.c:120
opencl_update_mxyz_p_ruvw
void opencl_update_mxyz_p_ruvw(void *m_x, void *m_y, void *m_z, void *p, void *ruvw, void *u, void *v, void *w, void *E, void *rho, real gamma, int n)
Definition compressible_ops_update.c:82
w
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ w
Definition ax_helm_full_kernel.h:48
u
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ u
Definition ax_helm_full_kernel.h:46
v
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ v
Definition ax_helm_full_kernel.h:47
dirichlet_apply_scalar_kernel
__global__ void dirichlet_apply_scalar_kernel(const int *__restrict__ msk, T *__restrict__ x, const T g, const int m)
Definition dirichlet_kernel.h:42
device_config.h
real
double real
Definition device_config.h:12
opencl_kernel_jit
void opencl_kernel_jit(const char *kernel, cl_program *program)
Definition jit.c:50
CL_CHECK
#define CL_CHECK(err)
Definition check.h:12
prgm_lib.h
compressible_ops_update_program
void * compressible_ops_update_program