System crashing (unresponsive screen) when running OpenCL code - opencl

Question

I am writing this simple opencl code and everytime I execute the code it causes system to crash, only my mouse cursor is responsive and there is a one screen blank. Similar thing happened to me when I was using clinfo with up-to-date AMDGPU-PRO driver. I am currently using slightly older version which is 20.40, and I had not had any problem with this version for running other OpenCL codes. I did try downgrading the driver to even older one, 20.30, and I am having the same problem.
One edit that I made for my code is changing the arg_size paramters in kernel arguments, clSetKernelArg(), from sizeof(cl_mem) to sizeof(cl_int) for the first four variables and sizeof(cl_mem) to sizeof(cl_float) or sizeof(float) for the last four variables. I made the changes because I was getting CL_INVALID_KERNEL_ARGS error. After the changes were made, I've noticed that I was not getting any CL errors anymore, so I thought I finally fixed the code, but now it is crashing my system.... Looking at journalctl log, it seems that it has to do with some shared libraries. Other than that I have no idea what the log is saying.
Below is the main code, kernel code, and the journalctl log.
Main code
#include <stdio.h>
#include <stdlib.h>
#include <netcdf.h>
#define CL_TARGET_OPENCL_VERSION 120
#include <CL/cl.h>
#include "cl_err.h"
// netCDF constants
#define err(e) {printf("Error: %s\n", nc_strerror(e)); return(2);}
#define clerrchk(arg, e) {printf(" %-40s : %s\n",arg, geterrstr(e));}
#define fname "leap3d.nc"
// Variable sizes and dimensions (constants)
#define ndims 4
void data_init(int in_x_siz, int in_y_siz, int in_z_siz, float *in_arr);
void pbndry(int in_x_siz, int in_y_siz, int in_z_siz, float *in_arr);
int main()
{
int i,j,k;
int nx = 128,
ny = 128,
nz = 16,
nt = 1000;
int *p_nx = &nx,
*p_ny = &ny,
*p_nz = &nz,
*p_nt = &nt;
float u = 0.0,
v = 5.0,
w = 0.0,
c = 0.01;
float *p_u = &u,
*p_v = &v,
*p_w = &w,
*p_c = &c;
// p_tf : p at future
// p_tn : p at now
// p_tp : p at past
float q_tf[nz+2][ny+2][nx+2];
float q_tn[nz+2][ny+2][nx+2];
float q_tp[nz+2][ny+2][nx+2];
float (*p_tf)[ny+2][nx+2] = q_tf;
float (*p_tn)[ny+2][nx+2] = q_tn;
float (*p_tp)[ny+2][nx+2] = q_tp;
size_t p_siz = sizeof(float) * (nx+2) * (ny+2) * (nz+2);
size_t n_siz = sizeof(int) ,
c_siz = sizeof(float) ;
int ncid, retval, varid, x_dimid, y_dimid, z_dimid, t_dimid;
int dimids[ndims];
size_t start[ndims], count[ndims];
// netCDF file operation
// Creating netCDF file
if ((retval = nc_create(fname, NC_CLOBBER, &ncid)))
err(retval);
// Define dimensions
if ((retval = nc_def_dim(ncid, "z", nz+2, &z_dimid)))
err(retval);
if ((retval = nc_def_dim(ncid, "y", ny+2, &y_dimid)))
err(retval);
if ((retval = nc_def_dim(ncid, "x", nx+2, &x_dimid)))
err(retval);
if ((retval = nc_def_dim(ncid, "t", NC_UNLIMITED, &t_dimid)))
err(retval);
// Dimension ids
dimids[0] = t_dimid;
dimids[1] = z_dimid;
dimids[2] = y_dimid;
dimids[3] = x_dimid;
// Variable for writing netCDF data one timestep at a time
count[0] = 1; // For time dimension : 1 timestep
count[1] = nz+2; // For z : write everything
count[2] = ny+2; // For y : write everything
count[3] = nx+2; // For x : write everything
start[1] = 0; // For z : don't do anything
start[2] = 0; // For y : don't do anything
start[3] = 0; // For x : don't do anything
if ((retval = nc_def_var(ncid, "data", NC_FLOAT, ndims, dimids, &varid)))
err(retval);
if ((retval = nc_enddef(ncid)))
err(retval);
data_init(nx,ny,nz,(float*)p_tf);
data_init(nx,ny,nz,(float*)p_tn);
data_init(nx,ny,nz,(float*)p_tp);
// Euler scheme for the first time step
for(k=1;k<nz+1;k++)
for(j=1;j<ny+1;j++)
for(i=1;i<nx+1;i++)
{
p_tf[k][j][i] = p_tn[k][j][i]
- u * c * (p_tn[k][j][i] - p_tn[k][j][i-1])
- v * c * (p_tn[k][j][i] - p_tn[k][j-1][i])
- w * c * (p_tn[k][j][i] - p_tn[k-1][j][i]);
}
pbndry(nx,ny,nz,(float*)p_tf);
p_tp = p_tn;
p_tn = p_tf;
start[0] = 0;
if (retval = nc_put_vara_float(ncid, varid, start, count, &p_tf[0][0][0]))
err(retval);
// OpenCL part //
// Use this to check the output of each API call
cl_int status;
// Retrieve the number of Platforms
cl_uint numPlatforms = 0;
status = clGetPlatformIDs(0, NULL, &numPlatforms);
// Allocate enough space for each Platform
cl_platform_id *platforms = (cl_platform_id*)malloc(numPlatforms*sizeof(cl_platform_id));
// Fill in the Platforms
status = clGetPlatformIDs(numPlatforms, platforms, NULL);
// Retrieve the number of Devices
cl_uint numDevices = 0;
status = clGetDeviceIDs(platforms[0],CL_DEVICE_TYPE_ALL, 0, NULL, &numDevices);
clerrchk("get number of devices", status);
// Allocate enough spaces for each Devices
char name_data[100];
int *comp_units;
cl_device_fp_config cfg;
cl_device_id *devices = (cl_device_id*)malloc(numDevices*sizeof(cl_device_id));
// Fill in the Devices
status = clGetDeviceIDs(platforms[0], CL_DEVICE_TYPE_ALL, numDevices, devices, NULL);
clerrchk("get device ids", status);
// for(i=0;i<numDevices;i++)
// {
// status = clGetDeviceInfo(devices[i], CL_DEVICE_NAME, sizeof(name_data), name_data, NULL);
//
// printf("Device Name #%d: %s\n", i, name_data);
// status = clGetDeviceInfo(devices[i], CL_DEVICE_MAX_COMPUTE_UNITS, sizeof(comp_units), &comp_units, NULL);
//
// printf("Max Work-Group %d\n", comp_units);
// status = clGetDeviceInfo(devices[i], CL_DEVICE_DOUBLE_FP_CONFIG, sizeof(cfg), &cfg, NULL);
//
// printf("Double FP config = %llu, Support? = %d\n", cfg, status);
// }
// Create a context and associate it with the devices
cl_context context = clCreateContext(NULL, numDevices, devices, NULL, NULL, &status);
clerrchk("create context", status);
// Create a command queue and associate it with the devices
cl_command_queue cmdQueue = clCreateCommandQueue(context, devices[0], 0, &status);
clerrchk("create cmd queue", status);
cl_mem buf_p_tf = clCreateBuffer(context, CL_MEM_READ_WRITE, p_siz, NULL, &status);
clerrchk("create buffer buf_p_tf", status);
cl_mem buf_p_tn = clCreateBuffer(context, CL_MEM_READ_ONLY , p_siz, NULL, &status);
clerrchk("create buffer buf_p_tn", status);
cl_mem buf_p_tp = clCreateBuffer(context, CL_MEM_READ_ONLY , p_siz, NULL, &status);
clerrchk("create buffer buf_p_tp", status);
cl_mem buf_nx = clCreateBuffer(context, CL_MEM_READ_ONLY , n_siz, NULL, &status);
clerrchk("create buffer buf_nx", status);
cl_mem buf_ny = clCreateBuffer(context, CL_MEM_READ_ONLY , n_siz, NULL, &status);
clerrchk("create buffer buf_ny", status);
cl_mem buf_nz = clCreateBuffer(context, CL_MEM_READ_ONLY , n_siz, NULL, &status);
clerrchk("create buffer buf_nz", status);
cl_mem buf_nt = clCreateBuffer(context, CL_MEM_READ_ONLY , n_siz, NULL, &status);
clerrchk("create buffer buf_nt", status);
cl_mem buf_u = clCreateBuffer(context, CL_MEM_READ_ONLY , c_siz, NULL, &status);
clerrchk("create buffer buf_u", status);
cl_mem buf_v = clCreateBuffer(context, CL_MEM_READ_ONLY , c_siz, NULL, &status);
clerrchk("create buffer buf_v", status);
cl_mem buf_w = clCreateBuffer(context, CL_MEM_READ_ONLY , c_siz, NULL, &status);
clerrchk("create buffer buf_w", status);
cl_mem buf_c = clCreateBuffer(context, CL_MEM_READ_ONLY , c_siz, NULL, &status);
clerrchk("create buffer buf_c", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_p_tf , CL_FALSE, 0, p_siz, p_tf ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_p_tf", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_p_tn , CL_FALSE, 0, p_siz, p_tn ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_p_tn", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_p_tp , CL_FALSE, 0, p_siz, p_tp ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_p_tp", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_nx , CL_FALSE, 0, n_siz, p_nx ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_nx", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_ny , CL_FALSE, 0, n_siz, p_ny ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_ny", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_nz , CL_FALSE, 0, n_siz, p_nz ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_nz", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_nt , CL_FALSE, 0, n_siz, p_nt ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_nt", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_u , CL_FALSE, 0, c_siz, p_u ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_u", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_v , CL_FALSE, 0, c_siz, p_v ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_v", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_w , CL_FALSE, 0, c_siz, p_w ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_w", status);
status = clEnqueueWriteBuffer(cmdQueue, buf_c , CL_FALSE, 0, c_siz, p_c ,0, NULL, NULL);
clerrchk("enqueue write buffer for buf_c", status);
// Create Program with the source code
cl_program program = NULL;
size_t program_size;
char *program_source;
FILE *program_handle = fopen("leapfrog.cl","r");
fseek(program_handle, 0, SEEK_END);
program_size = ftell(program_handle);
rewind(program_handle);
program_source = (char*)malloc(program_size+1);
program_source[program_size] = '\0';
fread(program_source, sizeof(char), program_size, program_handle);
fclose(program_handle);
program = clCreateProgramWithSource(context, 1, (const char**)&program_source, &program_size, &status);
clerrchk("create program", status);
// Compile the Program for the Device
status = clBuildProgram(program, numDevices, devices, NULL, NULL, NULL);
if(status != CL_SUCCESS)
{
//printf("Code : %d\n",status);
//printf("Program 1 %s\n",getErrorString(status));
size_t log_size;
clGetProgramBuildInfo(program, devices[0], CL_PROGRAM_BUILD_LOG, 0, NULL, &log_size);
char *log = (char *) malloc(log_size);
clGetProgramBuildInfo(program, devices[0], CL_PROGRAM_BUILD_LOG, log_size, log, NULL);
printf("%s\n", log);
}
// Create a kernel
cl_kernel kernel = NULL;
kernel = clCreateKernel(program, "leapfrog3d", &status);
clerrchk("create kernel", status);
// Associate the input and output buffers with the kernel
status = clSetKernelArg(kernel, 0, sizeof(cl_int), &buf_nx );
clerrchk("set kernel buf_nx", status);
status = clSetKernelArg(kernel, 1, sizeof(cl_int), &buf_ny );
clerrchk("set kernel buf_ny", status);
status = clSetKernelArg(kernel, 2, sizeof(cl_int), &buf_nz );
clerrchk("set kernel buf_nz", status);
status = clSetKernelArg(kernel, 3, sizeof(cl_int), &buf_nt );
clerrchk("set kernel buf_nz", status);
status = clSetKernelArg(kernel, 4, sizeof(cl_mem), &buf_p_tf);
clerrchk("set kernel buf_p_tf", status);
status = clSetKernelArg(kernel, 5, sizeof(cl_mem), &buf_p_tn);
clerrchk("set kernel buf_p_tn", status);
status = clSetKernelArg(kernel, 6, sizeof(cl_mem), &buf_p_tp);
clerrchk("set kernel buf_p_tp", status);
status = clSetKernelArg(kernel, 7, sizeof(cl_float), &buf_u );
clerrchk("set kernel buf_u", status);
status = clSetKernelArg(kernel, 8, sizeof(cl_float), &buf_v );
clerrchk("set kernel buf_v", status);
status = clSetKernelArg(kernel, 9, sizeof(cl_float), &buf_w );
clerrchk("set kernel buf_w", status);
status = clSetKernelArg(kernel,10, sizeof(cl_float), &buf_c );
clerrchk("set kernel buf_c", status);
// Define index space (global work size) of work items for execution
// A workgroup size (local work size) is not required, but can be used
size_t glbworksiz[3] = {nx,ny,nz};
// Execute the kernel for execution
status = clEnqueueNDRangeKernel(cmdQueue, kernel, 3, NULL, glbworksiz, NULL, 0, NULL, NULL);
clerrchk("enqueue nd range kernel", status);
// Read the Device output buffer to the host output array
status = clEnqueueReadBuffer(cmdQueue, buf_p_tf, CL_TRUE, 0, p_siz, p_tf, 0, NULL, NULL);
clerrchk("enqueue read buffer", status);
start[0] = 1;
if (retval = nc_put_vara_float(ncid, varid, start, count, &p_tf[0][0][0]))
err(retval);
if ((retval = nc_close(ncid)))
err(retval);
clReleaseMemObject(buf_p_tf);
clReleaseMemObject(buf_p_tn);
clReleaseMemObject(buf_p_tp);
clReleaseMemObject(buf_nx);
clReleaseMemObject(buf_ny);
clReleaseMemObject(buf_nz);
clReleaseMemObject(buf_nt);
clReleaseMemObject(buf_u);
clReleaseMemObject(buf_v);
clReleaseMemObject(buf_w);
clReleaseMemObject(buf_c);
clReleaseContext(context);
clReleaseKernel(kernel);
clReleaseProgram(program);
clReleaseCommandQueue(cmdQueue);
printf("\nDone. . .\n");
return 0;
}
void data_init(int in_x_siz, int in_y_siz, int in_z_siz, float *in_arr)
{
int i,j,k;
int i_min = 50,
i_max = 70,
j_min = 50,
j_max = 70;
for(k=0;k<in_z_siz+2;k++)
for(j=0;j<in_y_siz+2;j++)
for(i=0;i<in_x_siz+2;i++)
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) +i] = 0.0;
for(k=1;k<in_z_siz+1;k++)
for(j=j_min;j<j_max;j++)
for(i=i_min;i<i_max;i++)
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) +i] = 3.0;
}
void pbndry(int in_x_siz, int in_y_siz, int in_z_siz, float *in_arr)
{
int i,j,k;
// Periodic boundary
// x-direction
for(k=1;k<in_z_siz+1;k++)
for(j=1;j<in_y_siz+1;j++)
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + 0] =
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + in_x_siz];
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + (in_x_siz+1)] =
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + 1];
// y-direction
for(k=1;k<in_z_siz+1;k++)
for(i=1;i<in_x_siz+1;i++)
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + 0 * (in_x_siz+2) + i] =
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + in_y_siz * (in_x_siz+2) + i];
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + (in_y_siz+1) * (in_x_siz+2) + i] =
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + 1 * (in_x_siz+2) + i];
// z-direction
for(j=1;j<in_y_siz+1;j++)
for(i=1;i<in_x_siz+1;i++)
in_arr[0 * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + i] =
in_arr[in_z_siz * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + i];
in_arr[(in_z_siz+1) * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + i] =
in_arr[1 * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + i];
}
Kernel
void pbndry(int in_x_siz, int in_y_siz, int in_z_siz, global float *in_arr)
{
int i,j,k;
// Periodic boundary
// x-direction
for(k=1;k<in_z_siz+1;k++)
for(j=1;j<in_y_siz+1;j++)
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + 0] =
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + in_x_siz];
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + (in_x_siz+1)] =
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + 1];
// y-direction
for(k=1;k<in_z_siz+1;k++)
for(i=1;i<in_x_siz+1;i++)
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + 0 * (in_x_siz+2) + i] =
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + in_y_siz * (in_x_siz+2) + i];
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + (in_y_siz+1) * (in_x_siz+2) + i] =
in_arr[k * (in_y_siz+2) * (in_x_siz+2) + 1 * (in_x_siz+2) + i];
// z-direction
for(j=1;j<in_y_siz+1;j++)
for(i=1;i<in_x_siz+1;i++)
in_arr[0 * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + i] =
in_arr[in_z_siz * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + i];
in_arr[(in_z_siz+1) * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + i] =
in_arr[1 * (in_y_siz+2) * (in_x_siz+2) + j * (in_x_siz+2) + i];
}
kernel void leapfrog3d(
const int x_siz,
const int y_siz,
const int z_siz,
const int t_siz,
global float *in_p_tf,
global float *in_p_tn,
global float *in_p_tp,
const float u_vel,
const float v_vel,
const float w_vel,
const float c
)
{
int nx = x_siz;
int ny = y_siz;
int nz = z_siz;
int nt = t_siz;
float u = u_vel;
float v = v_vel;
float w = w_vel;
float C = c ;
int i = get_global_id(0);
int j = get_global_id(1);
int k = get_global_id(2);
int idx0, idx_i0, idx_i1, idx_j0, idx_j1, idx_k0, idx_k1;
for(int t=1;t<t_siz;t++)
{
idx0 = i + j * (nx+2) + k * (nx+2) * (ny+2);
idx_i0 = (i+1) + j * (nx+2) + k * (nx+2) * (ny+2);
idx_j0 = i + (j+1) * (nx+2) + k * (nx+2) * (ny+2);
idx_k0 = i + j * (nx+2) + (k+1) * (nx+2) * (ny+2);
idx_i1 = (i-1) + j * (nx+2) + k * (nx+2) * (ny+2);
idx_j1 = i + (j-1) * (nx+2) + k * (nx+2) * (ny+2);
idx_k1 = i + j * (nx+2) + (k-1) * (nx+2) * (ny+2);
in_p_tf[idx0] = in_p_tp[idx0]
- u_vel * C * (in_p_tn[idx_i0] - in_p_tn[idx_i1])
- v_vel * C * (in_p_tn[idx_j0] - in_p_tn[idx_j1])
- w_vel * C * (in_p_tn[idx_k0] - in_p_tn[idx_k1]);
pbndry(nx,ny,nz,in_p_tf);
in_p_tp = in_p_tn;
in_p_tn = in_p_tf;
}
}
journalctl log
Aug 19 13:17:12 Angke systemd-coredump[121514]: Process 121510 (lpf.gpu) of user 1000 dumped core.
Found module /home/rangke/prog_works/opencl/adv/lpf.gpu with build-id: 8036638bab286ce6a1e81a4983bb68d19a7145fd
Found module linux-vdso.so.1 with build-id: 27d9b8c0c25b172c86a29351e47701c1d0676035
Found module libamdocl12cl64.so without build-id.
Found module libgcc_s.so.1 with build-id: 7f8508bb914546ada778809b64b99d234337d835
Found module libstdc++.so.6 with build-id: 8ab0e57054dd1dcba681f217016afc6a4e639783
Found module libamd_comgr.so.1 with build-id: 438b1fbf7c58fd6a5b555a7283a58ee1eb1808f0
Found module libdrm.so.2 with build-id: 3aeff5403ca8d7589eabc05752eb613937f454a1
Found module libdrm_amdgpu.so.1 with build-id: a89ceb7c9082e5276f39023716eb4d194e75f6b8
Found module libamdocl-orca64.so without build-id.
Found module librt.so.1 with build-id: 75484da2d6f1515189eefa076e0a40328834cd16
Found module libamdocl64.so without build-id.
Found module libresolv.so.2 with build-id: c915c72668282861a813f7ea3c0780f37b681dc0
Found module libkeyutils.so.1 with build-id: ac405ddd17be10ce538da3211415ee50c8f8df79
Found module libkrb5support.so.0 with build-id: c4ee4ad1dc2da57487bc2419b88f1b6873184582
Found module libcom_err.so.2 with build-id: eb61ef71c8b97846db759fb89a115405cff6dd30
Found module libk5crypto.so.3 with build-id: 632a59ed7c35704d84645e6d1e9873348d1eb802
Found module libkrb5.so.3 with build-id: c61cb4da63b8a839ee7df99eaf9dbd3d0968534c
Found module libunistring.so.2 with build-id: 015ac6d6bcb60b7d8bea31a80d1941b06e8636ab
Found module libpthread.so.0 with build-id: 07c8f95b4f3251d08550217ad8a1f31066229996
Found module libzstd.so.1 with build-id: 4b10444c1560ebc574af4d5f488b7408b22d450e
Found module libgssapi_krb5.so.2 with build-id: 9be9d3348399b72b76161a64e6d9fd760b77163a
Found module libcrypto.so.1.1 with build-id: 81b77a8e0b6e1c0db19644a5f120890f02762021
Found module libssl.so.1.1 with build-id: 99394a6653d9c107f2e9b730bbbfd18ed43ae3b9
Found module libpsl.so.5 with build-id: 0229a201aaf5652186c9fdc192ebe52baf19d7f1
Found module libssh2.so.1 with build-id: 7f6d9edd2e793b266cae4f22fc1ba7b6b401c08c
Found module libidn2.so.0 with build-id: 1ce2b50ad9f9821c2c629b521cf5a3c99593d332
Found module libnghttp2.so.14 with build-id: 5ca39b42e7cb2770878644d57e88677df6336815
Found module libz.so.1 with build-id: 81bf6e728a6d6f5b105b0f8b25f6c614ce10452a
Found module libsz.so.2 with build-id: c114ff6d6bb52989ad7099aacac51780e5ef418e
Found module libcurl.so.4 with build-id: 49c0cb842d0e0dad11c435b7fb88b3d88b8a43ac
Found module libm.so.6 with build-id: 2b8fd1f869ecab4e0b55e92f2f151897f6818acf
Found module libhdf5.so.200 with build-id: 553f354452b0af3e7232580b8dff9e0c6584830b
Found module libhdf5_hl.so.200 with build-id: 2b0926fbab5318a556eb524497edb6e78099ff60
Found module ld-linux-x86-64.so.2 with build-id: 040cc3dd10461562f177df39e3be2f3704258c3c
Found module libdl.so.2 with build-id: 5abc547e7b0949f89f3c0e21ab0c8331a7440a8a
Found module libc.so.6 with build-id: 4b406737057708c0e4c642345a703c47a61c73dc
Found module libnetcdf.so.18 with build-id: f68d6c8120acfeaaa265d3b0750d24b669671124
Found module libOpenCL.so.1 with build-id: 4f566e048bc3b8112ba357ef29a3affd5858ccdf
Stack trace of thread 121510:
#0 0x00007efc82a9ad71 clEnqueueWriteBuffer (libamdocl64.so + 0xc8ad71)
#1 0x00007efc88a2d28d clEnqueueWriteBuffer (libOpenCL.so.1 + 0x1228d)
#2 0x000055d84a146dcd n/a (/home/rangke/prog_works/opencl/adv/lpf.gpu + 0x2dcd)
#3 0x00007efc88726b25 __libc_start_main (libc.so.6 + 0x27b25)
#4 0x000055d84a14528e n/a (/home/rangke/prog_works/opencl/adv/lpf.gpu + 0x128e)
Mind you that lpf.gpu is the executable that I ran.