Neko  0.9.99
A portable framework for high-order spectral element flow simulations
opr_conv1.c
Go to the documentation of this file.
1 /*
2  Copyright (c) 2021-2022, 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 
47 #include "conv1_kernel.cl.h"
48 
52 void opencl_conv1(void *du, void *u,
53  void *vx, void *vy, void *vz,
54  void *dx, void *dy, void *dz,
55  void *drdx, void *dsdx, void *dtdx,
56  void *drdy, void *dsdy, void *dtdy,
57  void *drdz, void *dsdz, void *dtdz,
58  void *jacinv, int *nel, int *gdim, int *lx) {
59  cl_int err;
60 
61  if (conv1_program == NULL)
62  opencl_kernel_jit(conv1_kernel, (cl_program *) &conv1_program);
63 
64  const size_t global_item_size = 256 * (*nel);
65  const size_t local_item_size = 256;
66 
67 #define STR(X) #X
68 #define CASE(LX) \
69  case LX: \
70  { \
71  cl_kernel kernel = clCreateKernel(conv1_program, \
72  STR(conv1_kernel_lx##LX), &err); \
73  CL_CHECK(err); \
74  \
75  CL_CHECK(clSetKernelArg(kernel, 0, sizeof(cl_mem), (void *) &du)); \
76  CL_CHECK(clSetKernelArg(kernel, 1, sizeof(cl_mem), (void *) &u)); \
77  CL_CHECK(clSetKernelArg(kernel, 2, sizeof(cl_mem), (void *) &vx)); \
78  CL_CHECK(clSetKernelArg(kernel, 3, sizeof(cl_mem), (void *) &vy)); \
79  CL_CHECK(clSetKernelArg(kernel, 4, sizeof(cl_mem), (void *) &vz)); \
80  CL_CHECK(clSetKernelArg(kernel, 5, sizeof(cl_mem), (void *) &dx)); \
81  CL_CHECK(clSetKernelArg(kernel, 6, sizeof(cl_mem), (void *) &dy)); \
82  CL_CHECK(clSetKernelArg(kernel, 7, sizeof(cl_mem), (void *) &dz)); \
83  CL_CHECK(clSetKernelArg(kernel, 8, sizeof(cl_mem), (void *) &drdx)); \
84  CL_CHECK(clSetKernelArg(kernel, 9, sizeof(cl_mem), (void *) &dsdx)); \
85  CL_CHECK(clSetKernelArg(kernel, 10, sizeof(cl_mem), (void *) &dtdx)); \
86  CL_CHECK(clSetKernelArg(kernel, 11, sizeof(cl_mem), (void *) &drdy)); \
87  CL_CHECK(clSetKernelArg(kernel, 12, sizeof(cl_mem), (void *) &dsdy)); \
88  CL_CHECK(clSetKernelArg(kernel, 13, sizeof(cl_mem), (void *) &dtdy)); \
89  CL_CHECK(clSetKernelArg(kernel, 14, sizeof(cl_mem), (void *) &drdz)); \
90  CL_CHECK(clSetKernelArg(kernel, 15, sizeof(cl_mem), (void *) &dsdz)); \
91  CL_CHECK(clSetKernelArg(kernel, 16, sizeof(cl_mem), (void *) &dtdz)); \
92  CL_CHECK(clSetKernelArg(kernel, 17, sizeof(cl_mem), (void *) &jacinv)); \
93  \
94  CL_CHECK(clEnqueueNDRangeKernel((cl_command_queue) glb_cmd_queue, \
95  kernel, 1, NULL, &global_item_size, \
96  &local_item_size, 0, NULL, NULL)); \
97  } \
98  break
99 
100  switch(*lx) {
101  CASE(2);
102  CASE(3);
103  CASE(4);
104  CASE(5);
105  CASE(6);
106  CASE(7);
107  CASE(8);
108  CASE(9);
109  CASE(10);
110  CASE(11);
111  CASE(12);
112  }
113 }
drdy
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ drdy
Definition: ax_helm_full_kernel.h:54
drdz
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ drdz
Definition: ax_helm_full_kernel.h:55
dsdz
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dsdz
Definition: ax_helm_full_kernel.h:58
dsdy
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dsdy
Definition: ax_helm_full_kernel.h:57
dtdy
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dtdy
Definition: ax_helm_full_kernel.h:60
u
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ u
Definition: ax_helm_full_kernel.h:46
dx
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dx
Definition: ax_helm_full_kernel.h:49
drdx
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ drdx
Definition: ax_helm_full_kernel.h:53
dtdz
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dtdz
Definition: ax_helm_full_kernel.h:61
dsdx
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dsdx
Definition: ax_helm_full_kernel.h:56
dtdx
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dtdx
Definition: ax_helm_full_kernel.h:59
dz
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dz
Definition: ax_helm_full_kernel.h:51
dy
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ dy
Definition: ax_helm_full_kernel.h:50
jacinv
__global__ void T *__restrict__ T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ jacinv
Definition: ax_helm_full_kernel.h:62
vz
__global__ void const T *__restrict__ const T *__restrict__ const T *__restrict__ const T *__restrict__ vz
Definition: conv1_kernel.h:135
vx
__global__ void const T *__restrict__ const T *__restrict__ vx
Definition: conv1_kernel.h:133
vy
__global__ void const T *__restrict__ const T *__restrict__ const T *__restrict__ vy
Definition: conv1_kernel.h:134
device_config.h
opencl_kernel_jit
void opencl_kernel_jit(const char *kernel, cl_program *program)
Definition: jit.c:50
CASE
#define CASE(LX)
opencl_conv1
void opencl_conv1(void *du, void *u, void *vx, void *vy, void *vz, void *dx, void *dy, void *dz, void *drdx, void *dsdx, void *dtdx, void *drdy, void *dsdy, void *dtdy, void *drdz, void *dsdz, void *dtdz, void *jacinv, int *nel, int *gdim, int *lx)
Definition: opr_conv1.c:52
prgm_lib.h
conv1_program
void * conv1_program