Please take a look at my OpenCL 2.0 tutorial on the use of enqueue_kernel and work-group scan functions. It also has a very cool algorithm, GPU-Quicksort, implemented in both OpenCL 1.2 and 2.0.
https://software.intel.com/en-us/articles/gpu-quicksort-in-opencl-20-usi...
Let me know what you think!
Hello,
I'm quite new to OpenCL and parallel computing in general.
I have just installed the SDK on a Linux SLES11 computer with a Core i7-4770 Processor.
Files installed were intel_sdk_for_ocl_applications_2014_4.6.0.92_x64.tgz and opencl_runtime_14.2_x64_4.5.0.8.tgz.
I've tried running two tests, first the sample code for general matrix multiply, and second a simple program to perform an addition and print out the processors capabilities, like MAX_COMPUTEEE_UNITS, MAX_WORK_GROUP_SIZE, etc. They both fail upon trying to compile the kernel, with the following message:
symbol lookup error: /opt/intel/opencl-1.2-4.5.0.8/lib64/libclang_compiler.so: undefined symbol: _ZN5Intel6OpenCL5Utils6Logger11GetInstanceEv, version Base
I guess it's looking for a library it can't find? Which one, and how to fix this problem.
Thanks, Bob
comile command:
/opt/intel/bin/icpc -g -o test -I/opt/intel/opencl-sdk/include -L/opt/intel/opencl/lib64 -lOpenCL *.cpp
run command: ./test test.cl -debug
output:
CL_PLATFORM_NAME: Intel(R) OpenCL
CL_PLATFORM_VERSION: OpenCL 1.2 LINUX
run command: ./test test.ir
output:
CL_PLATFORM_NAME: Intel(R) OpenCL
CL_PLATFORM_VERSION: OpenCL 1.2 LINUX
Error CL_INVALID_PROGRAM in oclLogBuildInfo Call !!!
// oclTest.cpp
//
#include<stdio.h>
#include <stdlib.h>
#include <memory.h>
#include <string.h>
// All OpenCL headers
#if defined (__APPLE__) || defined(MACOSX)
#include <OpenCL/opencl.h>
#else
#include <CL/opencl.h>
#endif
#define VENDOR_INTEL "Intel(R)"
cl_int oclGetPlatformID(cl_platform_id* clSelectedPlatformID)
{
char chBuffer[1024];
cl_uint num_platforms;
cl_platform_id* clPlatformIDs;
cl_int ciErrNum;
*clSelectedPlatformID = NULL;
// Get OpenCL platform count
ciErrNum = clGetPlatformIDs (0, NULL, &num_platforms);
if (ciErrNum != CL_SUCCESS)
{
printf(" Error %i in clGetPlatformIDs Call !!!\n\n", ciErrNum);
return -1000;
}
else
{
if(num_platforms == 0)
{
printf("No OpenCL platform found!\n\n");
return -2000;
}
else
{
// if there's a platform or more, make space for ID's
if ((clPlatformIDs = (cl_platform_id*)malloc(num_platforms * sizeof(cl_platform_id))) == NULL)
{
printf("Failed to allocate memory for cl_platform ID's!\n\n");
return -3000;
}
// get platform info for each platform and trap the NVIDIA platform if found
ciErrNum = clGetPlatformIDs (num_platforms, clPlatformIDs, NULL);
for(cl_uint i = 0; i < num_platforms; ++i)
{
ciErrNum = clGetPlatformInfo (clPlatformIDs[i], CL_PLATFORM_NAME, 1024, &chBuffer, NULL);
if(ciErrNum == CL_SUCCESS)
{
if(strstr(chBuffer, VENDOR_INTEL) != NULL)
{
*clSelectedPlatformID = clPlatformIDs[i];
break;
}
}
}
// default to zeroeth platform if NVIDIA not found
if(*clSelectedPlatformID == NULL)
{
printf("WARNING: NVIDIA OpenCL platform not found - defaulting to first platform!\n\n");
*clSelectedPlatformID = clPlatformIDs[0];
}
free(clPlatformIDs);
}
}
return CL_SUCCESS;
}
char* oclLoadProgSource(const char* cFilename, size_t* szFinalLength)
{
// locals
FILE* pFileStream = NULL;
size_t szSourceLength;
// open the OpenCL source code file
#ifdef _WIN32 // Windows version
if(fopen_s(&pFileStream, cFilename, "rb") != 0)
{
return NULL;
}
#else // Linux version
pFileStream = fopen(cFilename, "rb");
if(pFileStream == 0)
{
return NULL;
}
#endif
// get the length of the source code
fseek(pFileStream, 0, SEEK_END);
szSourceLength = ftell(pFileStream);
fseek(pFileStream, 0, SEEK_SET);
// allocate a buffer for the source code string and read it in
char* cSourceString = (char *)malloc(szSourceLength + 1);
if (fread((cSourceString), szSourceLength, 1, pFileStream) != 1)
{
fclose(pFileStream);
free(cSourceString);
return 0;
}
// close the file and return the total length of the combined (preamble + source) string
fclose(pFileStream);
if(szFinalLength != 0)
{
*szFinalLength = szSourceLength ;
}
cSourceString[szSourceLength] = '\0';
return cSourceString;
}
const char* oclErrorString(cl_int error)
{
static const char* errorString[] = {
"CL_SUCCESS","CL_DEVICE_NOT_FOUND","CL_DEVICE_NOT_AVAILABLE","CL_COMPILER_NOT_AVAILABLE","CL_MEM_OBJECT_ALLOCATION_FAILURE","CL_OUT_OF_RESOURCES","CL_OUT_OF_HOST_MEMORY","CL_PROFILING_INFO_NOT_AVAILABLE","CL_MEM_COPY_OVERLAP","CL_IMAGE_FORMAT_MISMATCH","CL_IMAGE_FORMAT_NOT_SUPPORTED","CL_BUILD_PROGRAM_FAILURE","CL_MAP_FAILURE","","","","","","","","","","","","","","","","","","CL_INVALID_VALUE","CL_INVALID_DEVICE_TYPE","CL_INVALID_PLATFORM","CL_INVALID_DEVICE","CL_INVALID_CONTEXT","CL_INVALID_QUEUE_PROPERTIES","CL_INVALID_COMMAND_QUEUE","CL_INVALID_HOST_PTR","CL_INVALID_MEM_OBJECT","CL_INVALID_IMAGE_FORMAT_DESCRIPTOR","CL_INVALID_IMAGE_SIZE","CL_INVALID_SAMPLER","CL_INVALID_BINARY","CL_INVALID_BUILD_OPTIONS","CL_INVALID_PROGRAM","CL_INVALID_PROGRAM_EXECUTABLE","CL_INVALID_KERNEL_NAME","CL_INVALID_KERNEL_DEFINITION","CL_INVALID_KERNEL","CL_INVALID_ARG_INDEX","CL_INVALID_ARG_VALUE","CL_INVALID_ARG_SIZE","CL_INVALID_KERNEL_ARGS","CL_INVALID_WORK_DIMENSION","CL_INVALID_WORK_GROUP_SIZE","CL_INVALID_WORK_ITEM_SIZE","CL_INVALID_GLOBAL_OFFSET","CL_INVALID_EVENT_WAIT_LIST","CL_INVALID_EVENT","CL_INVALID_OPERATION","CL_INVALID_GL_OBJECT","CL_INVALID_BUFFER_SIZE","CL_INVALID_MIP_LEVEL","CL_INVALID_GLOBAL_WORK_SIZE",
};
const int errorCount = sizeof(errorString) / sizeof(errorString[0]);
const int index = -error;
return (index >= 0 && index < errorCount) ? errorString[index] : "Unspecified Error";
}
#define HDASHLINE "-----------------------------------------------------------\n"
void oclLogBuildInfo(cl_program cpProgram, cl_device_id cdDevice)
{
// write out the build log and ptx, then exit
char cBuildLog[10240]={0};
cl_int ciErrNum =clGetProgramBuildInfo(cpProgram, cdDevice, CL_PROGRAM_BUILD_LOG, sizeof(cBuildLog), cBuildLog, NULL );
if(ciErrNum!=CL_SUCCESS)
printf(" Error %s in oclLogBuildInfo Call !!!\n\n", oclErrorString(ciErrNum));
else
{
printf("\n%s\nBuild Log:\n%s\n%s\n", HDASHLINE, cBuildLog, HDASHLINE);
FILE *stream = fopen( "BuildInfo.log", "w" );
if (stream==NULL)
{
return;
}
fprintf( stream, "%s", cBuildLog);
fclose( stream );
}
}
int main(int argc, char** argvs)
{
cl_platform_id clSelectedPlatformID=NULL;
char *pOclSource=NULL;
char cBuffer[1024];
cl_uint num_platforms;
cl_uint ciDeviceCount;
cl_platform_id* clPlatformIDs;
cl_int ciErrNum;
cl_device_id devices[8];
cl_context contexts[8];
cl_program programs[8];
cl_command_queue commandQueues[8];
size_t srcLen=0;
pOclSource=oclLoadProgSource(argvs[1],&srcLen);
ciErrNum = oclGetPlatformID (&clSelectedPlatformID);
ciErrNum = clGetPlatformInfo (clSelectedPlatformID, CL_PLATFORM_NAME, sizeof(cBuffer), cBuffer, NULL);
if (ciErrNum == CL_SUCCESS)
printf(" CL_PLATFORM_NAME: \t%s\n", cBuffer);
ciErrNum = clGetPlatformInfo (clSelectedPlatformID, CL_PLATFORM_VERSION, sizeof(cBuffer), cBuffer, NULL);
if (ciErrNum == CL_SUCCESS)
{
printf(" CL_PLATFORM_VERSION: \t%s\n", cBuffer);
}
ciErrNum = clGetDeviceIDs (clSelectedPlatformID, CL_DEVICE_TYPE_ACCELERATOR|CL_DEVICE_TYPE_GPU, 0, NULL, &ciDeviceCount);
ciErrNum = clGetDeviceIDs (clSelectedPlatformID, CL_DEVICE_TYPE_ACCELERATOR|CL_DEVICE_TYPE_GPU, ciDeviceCount, devices, &ciDeviceCount);
for(unsigned int i = 0; i < ciDeviceCount; ++i )
{
contexts[i] = clCreateContext(0, 1, &devices[i], NULL, NULL, &ciErrNum);
if (ciErrNum != CL_SUCCESS)
{
printf("Error %s in clCreateContext call !!!\n\n", oclErrorString(ciErrNum));
break;
}
commandQueues[i]=clCreateCommandQueue(contexts[i],devices[i],0,&ciErrNum);
if (ciErrNum != CL_SUCCESS)
{
printf("Error %s in clCreateCommandQueue call !!!\n\n", oclErrorString(ciErrNum));
break;
}
if((argc>2)&&(memcmp(argvs[2],"-debug",strlen("-debug"))==0))
{
programs[i]=clCreateProgramWithSource(contexts[i],1,(const char**)&pOclSource,&srcLen,&ciErrNum);
if (ciErrNum != CL_SUCCESS)
{
printf("Error %s in clCreateProgramWithSource call !!!\n\n", oclErrorString(ciErrNum));
break;
}
ciErrNum=clBuildProgram(programs[i],1,&(devices[i]),NULL,NULL,NULL);
if (ciErrNum != CL_SUCCESS)
{
printf("Error %s in clBuildProgram call !!!\n\n", oclErrorString(ciErrNum));
oclLogBuildInfo(programs[i],devices[i]);
break;
}
size_t binary_size=0;
ciErrNum=clGetProgramInfo(programs[i],CL_PROGRAM_BINARY_SIZES,sizeof(size_t),&binary_size,NULL);
binary_size++;
char *binary=new char[binary_size];
memset(binary,0,binary_size);
ciErrNum=clGetProgramInfo(programs[i],CL_PROGRAM_BINARIES,sizeof(size_t),&binary,NULL);
char filename[256]="test.ir";
FILE *fp=fopen(filename,"wb");
fwrite(binary,1,binary_size,fp);
fclose(fp);
delete binary;
}
else
{
programs[i]=clCreateProgramWithBinary(contexts[i],1,&devices[i],&srcLen,(const unsigned char**)&pOclSource,&ciErrNum,NULL);
if (ciErrNum != CL_SUCCESS)
{
printf("Error %s in clCreateProgramWithBinary call !!!\n\n", oclErrorString(ciErrNum));
break;
}
ciErrNum=clBuildProgram(programs[i],1,&(devices[i]),NULL,NULL,NULL);
if (ciErrNum != CL_SUCCESS)
{
oclLogBuildInfo(programs[i],devices[i]);
break;
}
}
//#endif
printf(" ---------------------------------\n");
//clGetDeviceInfo(devices[i], CL_DEVICE_NAME, sizeof(oclDevInfo.deviceName), &oclDevInfo.deviceName, NULL);
//printf(" Device: %s\n", deviceName);
printf(" ---------------------------------\n");
//oclPrintDevInfo(devices[i]);
}
return 0;
}
| Attachment | Size |
|---|---|
| DownloadImage may be NSFW. Clik here to view.  cl+ir.7z | 1.97 KB |
| DownloadImage may be NSFW. Clik here to view. oclTest_0.cpp | 8.72 KB |
This is very similar to an issue I ran into a while ago (in which I was using VS2012 rather than 2013), which turned out to be caused by (silently!) incompatible versions of Intel's OpenCL runtime, and SDK.
I have attached screenshots showing the problem, using Intel's own 'MedianFilter' example project (trivially modified with a printf).
The breakpoints show the solid red circles which indicate that they are properly bound, yet they do not work.
I am using:
Visual Studio 2013 Ultimate.
Windows 7, 64-bit.
OpenCL runtime for Intel Core and Xeon Processors [...] version 4.5.0.8 [the latest release]
Intel SDK for OpenCL Applications 2014 version 4.6.0.92 [the latest release]
| Attachment | Size |
|---|---|
| DownloadImage may be NSFW. Clik here to view. moreIntelDebugFail2_0.png | 117.21 KB |
| DownloadImage may be NSFW. Clik here to view. moreIntelDebugFail_0.png | 202.7 KB |
Hi,
I have an existing C++ application that uses OpenCL C++ Wrapper API, which compiled on an earlier Intel OpenCL SDK (I forgot the version), but after I upgrade my laptop to Windows 8.1, Visual Studio 2013 Ultmate, with the below spec:
Intel® Driver Update Utility
Product Detected Intel® HD Graphics 4400
Current Driver Installed 10.18.10.3907
Your driver is current.
Intel® SDK for OpenCL Applications 2014 version 4.6.0.92
My code would no longer compile. The errors is:
Warning 1 warning C4996: 'clCreateSampler': was declared deprecated C:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL\cl.hpp 4312 1 OpenCLProject1
Warning 2 warning C4996: 'clCreateCommandQueue': was declared deprecated C:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL\cl.hpp 5118 1 OpenCLProject1
Warning 3 warning C4996: 'clCreateCommandQueue': was declared deprecated C:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL\cl.hpp 5148 1 OpenCLProject1
Warning 4 warning C4996: 'clCreateCommandQueue': was declared deprecated C:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL\cl.hpp 5166 1 OpenCLProject1
Warning 5 warning C4996: 'clEnqueueTask': was declared deprecated C:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL\cl.hpp 5947 1 OpenCLProject1
6 IntelliSense: member "cl::detail::Wrapper<cl_device_id>::object_" (declared at line 1787 of "C:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL/cl.hpp") is inaccessible c:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL\cl.hpp 1819 23 OpenCLProject1
7 IntelliSense: member "cl::detail::Wrapper<cl_device_id>::object_" (declared at line 1787 of "C:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL/cl.hpp") is inaccessible c:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL\cl.hpp 1827 23 OpenCLProject1
8 IntelliSense: member "cl::detail::Wrapper<cl_device_id>::referenceCountable_" (declared at line 1788 of "C:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL/cl.hpp") is inaccessible c:\Program Files (x86)\Intel\OpenCL SDK\4.6\include\CL\cl.hpp 1828 35 OpenCLProject1
Is this because of version incompatiblity? If so, is there a way I can find out the correct version and specifically build my application against that version?
As part of a research project on finding bugs in multi- and many-core compilers, my team have found what we believe are some bugs in the Intel OpenCL 2.0 SDK compiler.
Would Intel be interested in receiving these bug reports? If so, how do you like to receive them?
Best wishes
Alastair Donaldson
My project work well if i compile it from opencl source file(test.cl) with clCreateProgramWithSource function. and save the cl program binary into an ir file(test.ir). When i recompile the cl program with clCreateProgramWithBinary , it gets an error while clBuildProgram is called.
The same code works well both on NVIDIA and AMD GPU platforms, except the intel mic card(5110p).
Is there something wrong with my code or intel opencl sdk?
This problem has trapped me for a long time, can anyone help me? It is very important for me, thank you!
comile command: /opt/intel/bin/icpc -g -o test -I/opt/intel/opencl-sdk/include -L/opt/intel/opencl/lib64 -lOpenCL *.cpp run command: ./test test.cl -debug output: CL_PLATFORM_NAME: Intel(R) OpenCL CL_PLATFORM_VERSION: OpenCL 1.2 LINUX run command: ./test test.ir output: CL_PLATFORM_NAME: Intel(R) OpenCL CL_PLATFORM_VERSION: OpenCL 1.2 LINUX Error CL_INVALID_PROGRAM in oclLogBuildInfo Call !!!
| Attachment | Size |
|---|---|
| DownloadImage may be NSFW. Clik here to view. oclTest_0.cpp | 8.69 KB |
| DownloadImage may be NSFW. Clik here to view. cl+ir.7z | 1.97 KB |
If a kernel takes an argument which is a pointer to a __global enum type, and that enum type has at least one member defined as having the value of other another (previous) member, then attempting to use breakpoints in the kernel will cause a catastrophic crash of Visual Studio 2013.
The program being debugged will continue to run.
The problem is transitive through inclusion of the enum type in a struct.
OpenCL kernel code illustrating the issue:
typedef enum MyEnum_e {
FIRST, SECOND = FIRST // further entries make no difference
} MyEnum;
__kernel void helloworld(
__global MyEnum* in // remove this arg to avoid breakpoint-time crash
)
{
int i;
i = 41;
i = 42; // set breakpoint here
i = 43;
}
I have attached a screenshot showing VS2013 having crashed.
As a slight aside: I presume it is dangerous to use such types to pass data from the host to the device, or vice-versa; as I understand it the size of an enum type is compiler-dependent (see these StackOverflow threads).
I am using:
Visual Studio 2013 Ultimate.
Windows 7, 64-bit.
OpenCL runtime for Intel Core and Xeon Processors [...] version 4.5.0.8
Intel SDK for OpenCL Applications 2014 version 4.6.0.92
| Attachment | Size |
|---|---|
| DownloadImage may be NSFW. Clik here to view. VS2013crash.png | 130.84 KB |
Is there any documentation on this extension in the new 3960 driver?
- Added support for OpenCL extension cl_intel_simultaneous_sharing to allow sharing of memory buffers between OpenCL, OpenGL, and DirectX. This feature is used by Adobe* applications such Photoshop CC*
Check out the specification for cl_intel_simultaneous_sharing OpenCL extension that was released with a new driver:
Enjoy!
Robert
Can host threads execute kernels concurrently with intel sdk for opencl? I heard that kernels(commands) from different command-queues will be executed concurrently on the device. Is that true? And, is "Device Fission" supported on GPU with Intel opencl driver now? That may be another way to implement it.
I use: Intel Core i7, Intel HD Graphics 4600, Intel sdk for opencl.
THX,
Lingzhi
I am experiencing a segmentation fault inside the call to clBuildProgram when using the Intel platform and Intel Xenon CPU. This is for a relatively complex kernel / OpenCL program. Simple kernels build fine. I am not having any problem building the same source code on the Apple or NVIDIA platforms.
I am attaching a bug demonstration program with as simple kernel. Please private message or email me for the offending kernel source code file.
g++ -o build_bug_demo opencl_program_build.cpp bugDemoSupport.cpp -I$OPENCL_INC_DIR -L$OPENCL_LIB_DIR -lOpenCL # Here is what I see for the offending program source: ./build_bug_demo -p 2 Selected CL_PLATFORM_NAME: Intel(R) OpenCL CL_DEVICE_NAME: Intel(R) Xeon(R) CPU X5650 @ 2.67GHz CL_DRIVER_VERSION: 1.2.0.92 Loading Source... clCreateProgramWithSource... clBuildProgram... Stack dump: 0. Running pass 'Intel OpenCL Vectorizer' on module 'Program'. 1. Running pass 'Intel OpenCL VectorizerCore' on function '@__Vectorized_.sum_reduce_over_velocity_space_stage1' 2. Running pass 'Predicator' on function '@__Vectorized_.sum_reduce_over_velocity_space_stage1' Segmentation fault (core dumped)
| Attachment | Size |
|---|---|
| DownloadImage may be NSFW. Clik here to view. bugDemoSupport.cpp | 7.36 KB |
| DownloadImage may be NSFW. Clik here to view. opencl_program_build.cpp | 2.82 KB |
| DownloadImage may be NSFW. Clik here to view. opencl_source_cl.txt | 394 bytes |
Hi,
I am a newbie to opencl and would like to know whether I can use Intel OpenCL SDK in a Windows Kernel driver. For eg: In windows display driver?
Regards
I've installed the Intel SDK on my Ubuntu VM, and the install seems to have worked fine. I can run the program here:
https://gist.github.com/rmcgibbo/6314452/download#
Which results in this info. As near as I can tell, it's all right. (Dunno if preferred vector widths are supposed to be 1, though.)
$ ./clDeviceQuery
clDeviceQuery Starting...
1 OpenCL Platforms found
CL_PLATFORM_NAME: Intel(R) OpenCL
CL_PLATFORM_VERSION: OpenCL 1.2 LINUX
OpenCL Device Info:
1 devices found supporting OpenCL on: Intel(R) OpenCL
----------------------------------
Device Intel(R) Core(TM) i7-4850HQ CPU @ 2.30GHz
---------------------------------
CL_DEVICE_NAME: Intel(R) Core(TM) i7-4850HQ CPU @ 2.30GHz
CL_DEVICE_VENDOR: Intel(R) Corporation
CL_DRIVER_VERSION: 1.2.0.8
CL_DEVICE_TYPE: CL_DEVICE_TYPE_CPU
CL_DEVICE_MAX_COMPUTE_UNITS: 2
CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS: 3
CL_DEVICE_MAX_WORK_ITEM_SIZES: 8192 / 8192 / 8192
CL_DEVICE_MAX_WORK_GROUP_SIZE: 8192
CL_DEVICE_MAX_CLOCK_FREQUENCY: 2300 MHz
CL_DEVICE_ADDRESS_BITS: 64
CL_DEVICE_MAX_MEM_ALLOC_SIZE: 626 MByte
CL_DEVICE_GLOBAL_MEM_SIZE: 2507 MByte
CL_DEVICE_ERROR_CORRECTION_SUPPORT: no
CL_DEVICE_LOCAL_MEM_TYPE: global
CL_DEVICE_LOCAL_MEM_SIZE: 32 KByte
CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE: 128 KByte
CL_DEVICE_QUEUE_PROPERTIES: CL_QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE
CL_DEVICE_QUEUE_PROPERTIES: CL_QUEUE_PROFILING_ENABLE
CL_DEVICE_IMAGE_SUPPORT: 1
CL_DEVICE_MAX_READ_IMAGE_ARGS: 480
CL_DEVICE_MAX_WRITE_IMAGE_ARGS: 480
CL_DEVICE_IMAGE <dim> 2D_MAX_WIDTH 16384
2D_MAX_HEIGHT 16384
3D_MAX_WIDTH 2048
3D_MAX_HEIGHT 2048
3D_MAX_DEPTH 2048
CL_DEVICE_PREFERRED_VECTOR_WIDTH_<t> CHAR 1, SHORT 1, INT 1, FLOAT 1, DOUBLE 1
clDeviceQuery, Platform Name = Intel(R) OpenCL, Platform Version = OpenCL 1.2 LINUX, NumDevs = 1, Device = Intel(R) Core(TM) i7-4850HQ CPU @ 2.30GHz
System Info:
Local Time/Date = 19:13:19, 10/24/2014
CPU Name: Intel(R) Core(TM) i7-4850HQ CPU @ 2.30GHz
# of CPU processors: 2
Linux version 3.2.0-23-generic (buildd@crested) (gcc version 4.6.3 (Ubuntu/Linaro 4.6.3-1ubuntu4) ) #36-Ubuntu SMP Tue Apr 10 20:39:51 UTC 2012
TEST PASSED
However, I can't get further than CL_DEVICE_NOT_AVAILABLE when I call clCreateContext(). If I run the test program on this page:
http://wiki.tiker.net/OpenCLHowTo
The output looks like this:
$ ./cl-demo 1000000 10
Choose platform:
[0] Intel(R) Corporation
Enter choice:
Choose device:
[0] Intel(R) Core(TM) i7-4850HQ CPU @ 2.30GHz
Enter choice:
*** 'clCreateContext' in 'cl-helper.c' on line 342 failed with error 'device not available'.
Aborted
Any ideas on what I'm doing wrong? As near as I can tell, the processor *should* be supported.
struct RGB16
{
ushort R, G, B;
};
__kernel void scale2D_3u16(__global ushort* restrict in, __global ushort* restrict out)
{
int xPos = get_global_id(0);
int yPos = get_global_id(1);
int p = yPos*get_global_size(0)+xPos;
__global struct RGB16* pRGBIn = (__global struct RGB16*)in;
__global struct RGB16* pRGBOut = (__global struct RGB16*)out;
pRGBOut[p] = pRGBIn[p];
}
Input data (memory view) is:
0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3, 4, 4, 4 ...
Using NVidia and AMD I get the same result as output. But using Intel OCL on HD4600 I get:
0, 0, 0, 0, 0, 1, 2, 2, 2, 0, 0, 3, 4, 4, 4, ...
I'm totally confused!
I am unable to pass a pointer to a kernel using intel's OpenCL drivers for Intel(R) Xeon(R) CPU E3-1245 v3 @ 3.40GHz on Linux.
I was able to get the program to work using AMD's CPU OpenCL drivers instead.
Any idea what the issue could be?
Also, I've been able to execute SPIR kernels on intel that do not have pointer arguments.
I uploaded the example program and kernel I've been working with.
| Attachment | Size |
|---|---|
| DownloadImage may be NSFW. Clik here to view. intel_spir.tgz | 62.94 KB |
Hi,
I have Windows 8.1 64-bit, Intel OpenCL SDK 4.6, Visual Studio 2013 Express Edition. Also I have added all nessesary directories in the project properties (include directory, lib directory and OpenCL.lib file for linker). Program is very simple:
#include <CL\cl.hpp>
int main() {
return 0;
}But I can't compile it: error C4996: 'clCreateSampler': was declared deprecated.
Do you have any ideas, how to solve this problem?
Thank you!
Hi everyone,
My VTune 2015 doesn't work when I choose "Overview" at "Analyze Processor Graphics Hardware Events". (work fine when choose "None". ) The log shows "The result you are opening is empty. This may be caused by an error during data collection. Try to re-run the analysis.", but also "Data collection is completed successfully." And the command window shows nothing like the application doesn't work successfully.
I use: windows 8.1 pro, vtune 2015, i7 4800mq
thx
Lingzhi
yep, its me again :)
I brought a Pendo Pad 8 tablet running Windows 8.1 (for only $200 AUD) to play with.
There is a problem with the opencl cpu driver
i.e.
- CL_DEVICE_NAME: Intel(R) Atom(TM) CPU Z3735E @ 1.33GHz
- CL_DEVICE_VENDOR: Intel(R) Corporation
- CL_DRIVER_VERSION: 4.5.0.8
- CL_DEVICE_PROFILE: FULL_PROFILE
- CL_DEVICE_VERSION: OpenCL 1.2 (Build 8)
- CL_DEVICE_TYPE: CPU
...............................
- CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE: 0
- CL_DEVICE_EXTENSIONS: 15
- Extensions:
- cl_khr_icd
- cl_khr_global_int32_base_atomics
- cl_khr_global_int32_extended_atomics
- cl_khr_local_int32_base_atomics
- cl_khr_local_int32_extended_atomics
- cl_khr_byte_addressable_store
- cl_khr_spir
- cl_intel_exec_by_local_thread
- cl_khr_depth_images
- cl_khr_3d_image_writes
- cl_khr_gl_sharing
- cl_intel_dx9_media_sharing
- cl_khr_dx9_media_sharing
- cl_khr_d3d11_sharing
- cl_khr_gl_sharing
If you ignore the lack of cl_khr_fp64 and PREFERRED_VECTOR_WIDTH_DOUBLE: 0 and create a kernel that uses double it works fine.
You can also see this using the Intel SDK 32 bit kernel build you can see it generating SSE 4,2 assembly code.
Please fix this, I assume its a bug.
It's possible to install the AMD CPU driver and it works, but it only generates SSE 2 assembly code and its not as optimized.
I made an argument to them along time ago, that if you go to the Steam Hardware Survey http://store.steampowered.com/hwsurvey and click on "Other Settings" you will see that 99.99% of the machine in the survey support SSE 2.
If you only support SSE4.2 then 30% of the machines on the survey (a lot of machine) can not support an application that uses opencl (unless it has a GPU that does).
I think its really cool that my full opencl development enviroment (I use CodeLite) runs on a $200 tablet.
Hopefully on a Core M derived tablet we might get fp64 on the GPU as well :)
As I posted in subject https://software.intel.com/en-us/forums/topic/533980
Here is more informations:
[common@localhost ~]$ workspace/BenchmarkX264/Debug/BenchmarkX264 x264 [info]: using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX x264 [info]: OpenCL acceleration enabled with Intel(R) Corporation Intel(R) Core(TM) i7-3520M CPU @ 2.90GHz x264 [info]: Compiling OpenCL kernels... *** Error in `workspace/BenchmarkX264/Debug/BenchmarkX264': munmap_chunk(): invalid pointer: 0x00007fd78c820210 *** ======= Backtrace: ========= /lib64/libc.so.6(+0x7bc07)[0x7fd7a7bf5c07] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xd6aefa)[0x7fd7969a0efa] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xdd9ae4)[0x7fd796a0fae4] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xdd9c00)[0x7fd796a0fc00] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xdae7f9)[0x7fd7969e47f9] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x2ac09d)[0x7fd795ee209d] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x2acf60)[0x7fd795ee2f60] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x29d5a4)[0x7fd795ed35a4] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x2a6d75)[0x7fd795edcd75] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x2a74d8)[0x7fd795edd4d8] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xda55d8)[0x7fd7969db5d8] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xda56b9)[0x7fd7969db6b9] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xda58bb)[0x7fd7969db8bb] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x27df89)[0x7fd795eb3f89] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xda55d8)[0x7fd7969db5d8] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xda56b9)[0x7fd7969db6b9] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xda58bb)[0x7fd7969db8bb] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x27b5b9)[0x7fd795eb15b9] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xda52af)[0x7fd7969db2af] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0xda60d9)[0x7fd7969dc0d9] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x22a1a8)[0x7fd795e601a8] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x1feefc)[0x7fd795e34efc] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x2120c7)[0x7fd795e480c7] /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so(+0x1f95d3)[0x7fd795e2f5d3] /opt/intel/opencl-1.2-4.5.0.8/lib64/libcpu_device.so(+0x2df57)[0x7fd798010f57] /opt/intel/opencl-1.2-4.5.0.8/lib64/libintelocl.so(+0xd0015)[0x7fd798a33015] /opt/intel/opencl-1.2-4.5.0.8/lib64/libtask_executor.so(+0x13837)[0x7fd798508837] /opt/intel/opencl-1.2-4.5.0.8/lib64/libtask_executor.so(+0x13c0e)[0x7fd798508c0e] /opt/intel/opencl-1.2-4.5.0.8/lib64/libtask_executor.so(+0x2ae8d)[0x7fd79851fe8d] /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbb.so.2(+0x24fea)[0x7fd7983cafea] /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbb.so.2(+0x20f39)[0x7fd7983c6f39] /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbb.so.2(+0x1f416)[0x7fd7983c5416] /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbb.so.2(+0x1b410)[0x7fd7983c1410] /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbb.so.2(+0x1b3b6)[0x7fd7983c13b6] /lib64/libpthread.so.0(+0x7df3)[0x7fd7a8761df3] /lib64/libc.so.6(clone+0x6d)[0x7fd7a7c7001d] ======= Memory map: ======== 00400000-0051b000 r-xp 00000000 08:06 1635048 /home/common/workspace/BenchmarkX264/Debug/BenchmarkX264 0071a000-0071b000 r--p 0011a000 08:06 1635048 /home/common/workspace/BenchmarkX264/Debug/BenchmarkX264 0071b000-0071c000 rw-p 0011b000 08:06 1635048 /home/common/workspace/BenchmarkX264/Debug/BenchmarkX264 0071c000-00797000 rw-p 00000000 00:00 0 025bf000-05b2d000 rw-p 00000000 00:00 0 [heap] 7fd7854ee000-7fd787b76000 r-xp 00000000 08:06 134619547 /opt/intel/opencl-1.2-4.5.0.8/lib64/libclang_compiler.so 7fd787b76000-7fd787d75000 ---p 02688000 08:06 134619547 /opt/intel/opencl-1.2-4.5.0.8/lib64/libclang_compiler.so 7fd787d75000-7fd787e98000 r--p 02687000 08:06 134619547 /opt/intel/opencl-1.2-4.5.0.8/lib64/libclang_compiler.so 7fd787e98000-7fd787ee1000 rw-p 027aa000 08:06 134619547 /opt/intel/opencl-1.2-4.5.0.8/lib64/libclang_compiler.so 7fd787ee1000-7fd788000000 rw-p 00000000 00:00 0 7fd788000000-7fd788021000 rw-p 00000000 00:00 0 7fd788021000-7fd78c000000 ---p 00000000 00:00 0 7fd78c000000-7fd78cb93000 rw-p 00000000 00:00 0 7fd78cb93000-7fd790000000 ---p 00000000 00:00 0 7fd790000000-7fd790021000 rw-p 00000000 00:00 0 7fd790021000-7fd794000000 ---p 00000000 00:00 0 7fd794280000-7fd794281000 ---p 00000000 00:00 0 7fd794281000-7fd794681000 rw-p 00000000 00:00 0 [stack:5529] 7fd794681000-7fd794682000 ---p 00000000 00:00 0 7fd794682000-7fd794a82000 rw-p 00000000 00:00 0 [stack:5528] 7fd794a82000-7fd794a83000 ---p 00000000 00:00 0 7fd794a83000-7fd795004000 rw-p 00000000 00:00 0 [stack:5527] 7fd795018000-7fd795066000 r--p 00000000 08:06 139595701 /opt/intel/opencl-1.2-4.5.0.8/lib64/clbltfne9_img_cbk.o 7fd795066000-7fd7950e6000 rwxp 00000000 00:00 0 7fd7950e6000-7fd79514a000 r--p 00000000 08:06 139595702 /opt/intel/opencl-1.2-4.5.0.8/lib64/clbltfne9_img_cbk.rtl 7fd79514a000-7fd79534b000 rw-p 00000000 00:00 0 7fd79534b000-7fd795453000 r--p 00000000 08:06 139595709 /opt/intel/opencl-1.2-4.5.0.8/lib64/clbltfnshared.rtl 7fd795453000-7fd795564000 r--p 00000000 08:06 139595700 /opt/intel/opencl-1.2-4.5.0.8/lib64/clbltfne9.rtl 7fd795564000-7fd795a35000 r-xp 00000000 08:06 139595697 /opt/intel/opencl-1.2-4.5.0.8/lib64/__ocl_svml_e9.so 7fd795a35000-7fd795c35000 ---p 004d1000 08:06 139595697 /opt/intel/opencl-1.2-4.5.0.8/lib64/__ocl_svml_e9.so 7fd795c35000-7fd795c36000 rw-p 004d1000 08:06 139595697 /opt/intel/opencl-1.2-4.5.0.8/lib64/__ocl_svml_e9.so 7fd795c36000-7fd79726d000 r-xp 00000000 08:06 139595710 /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so 7fd79726d000-7fd79746d000 ---p 01637000 08:06 139595710 /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so 7fd79746d000-7fd79752b000 r--p 01637000 08:06 139595710 /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so 7fd79752b000-7fd7975a8000 rw-p 016f5000 08:06 139595710 /opt/intel/opencl-1.2-4.5.0.8/lib64/libOclCpuBackEnd.so 7fd7975a8000-7fd7978c7000 rw-p 00000000 00:00 0 7fd7978c7000-7fd7978c9000 r-xp 00000000 08:06 74088815 /usr/lib64/libscif.so.0.0.1 7fd7978c9000-7fd797ac8000 ---p 00002000 08:06 74088815 /usr/lib64/libscif.so.0.0.1 7fd797ac8000-7fd797ac9000 rw-p 00001000 08:06 74088815 /usr/lib64/libscif.so.0.0.1 7fd797ac9000-7fd797b3b000 r-xp 00000000 08:06 74088826 /usr/lib64/libcoi_host.so.0 7fd797b3b000-7fd797d3b000 ---p 00072000 08:06 74088826 /usr/lib64/libcoi_host.so.0 7fd797d3b000-7fd797d3f000 rw-p 00072000 08:06 74088826 /usr/lib64/libcoi_host.so.0 7fd797d3f000-7fd797d4e000 rw-p 00000000 00:00 0 7fd797d7d000-7fd797ddc000 r-xp 00000000 08:06 140908586 /opt/intel/opencl-1.2-4.5.0.8/lib64/libmic_device.so 7fd797ddc000-7fd797fdb000 ---p 0005f000 08:06 140908586 /opt/intel/opencl-1.2-4.5.0.8/lib64/libmic_device.so 7fd797fdb000-7fd797fde000 r--p 0005e000 08:06 140908586 /opt/intel/opencl-1.2-4.5.0.8/lib64/libmic_device.so 7fd797fde000-7fd797fe2000 rw-p 00061000 08:06 140908586 /opt/intel/opencl-1.2-4.5.0.8/lib64/libmic_device.so 7fd797fe2000-7fd797fe3000 rw-p 00000000 00:00 0 7fd797fe3000-7fd798045000 r-xp 00000000 08:06 140908576 /opt/intel/opencl-1.2-4.5.0.8/lib64/libcpu_device.so 7fd798045000-7fd798245000 ---p 00062000 08:06 140908576 /opt/intel/opencl-1.2-4.5.0.8/lib64/libcpu_device.so 7fd798245000-7fd79824c000 r--p 00062000 08:06 140908576 /opt/intel/opencl-1.2-4.5.0.8/lib64/libcpu_device.so 7fd79824c000-7fd79824f000 rw-p 00069000 08:06 140908576 /opt/intel/opencl-1.2-4.5.0.8/lib64/libcpu_device.so 7fd79824f000-7fd798250000 rw-p 00000000 00:00 0 7fd798250000-7fd798280000 r-xp 00000000 08:06 140908582 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbbmalloc.so.2 7fd798280000-7fd798380000 ---p 00030000 08:06 140908582 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbbmalloc.so.2 7fd798380000-7fd798384000 rw-p 00030000 08:06 140908582 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbbmalloc.so.2 7fd798384000-7fd7983a6000 rw-p 00000000 00:00 0 7fd7983a6000-7fd7983ee000 r-xp 00000000 08:06 140908580 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbb.so.2 7fd7983ee000-7fd7984ee000 ---p 00048000 08:06 140908580 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbb.so.2 7fd7984ee000-7fd7984f3000 rw-p 00048000 08:06 140908580 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtbb.so.2 7fd7984f3000-7fd7984f5000 rw-p 00000000 00:00 0 7fd7984f5000-7fd79853b000 r-xp 00000000 08:06 140908578 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtask_executor.so 7fd79853b000-7fd79873b000 ---p 00046000 08:06 140908578 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtask_executor.so 7fd79873b000-7fd79873f000 r--p 00046000 08:06 140908578 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtask_executor.so 7fd79873f000-7fd798740000 rw-p 0004a000 08:06 140908578 /opt/intel/opencl-1.2-4.5.0.8/lib64/libtask_executor.so 7fd798740000-7fd798761000 r-xp 00000000 08:06 139595711 /opt/intel/opencl-1.2-4.5.0.8/lib64/libcl_logger.so 7fd798761000-7fd798961000 ---p 00021000 08:06 139595711 /opt/intel/opencl-1.2-4.5.0.8/lib64/libcl_logger.so 7fd798961000-7fd798962000 r--p 00021000 08:06 139595711 /opt/intel/opencl-1.2-4.5.0.8/lib64/libcl_logger.so 7fd798962000-7fd798963000 rw-p 00022000 08:06 139595711 /opt/intel/opencl-1.2-4.5.0.8/lib64/libcl_logger.so 7fd798963000-7fd798ad4000 r-xp 00000000 08:06 140908577 /opt/intel/opencl-1.2-4.5.0.8/lib64/libintelocl.so 7fd798ad4000-7fd798cd3000 ---p 00171000 08:06 140908577 /opt/intel/opencl-1.2-4.5.0.8/lib64/libintelocl.so 7fd798cd3000-7fd798cea000 r--p 00170000 08:06 140908577 /opt/intel/opencl-1.2-4.5.0.8/lib64/libintelocl.so 7fd798cea000-7fd798cec000 rw-p 00187000 08:06 140908577 /opt/intel/opencl-1.2-4.5.0.8/lib64/libintelocl.so 7fd798cec000-7fd798e00000 rw-p 00000000 00:00 0 7fd798e00000-7fd799600000 rw-p 00000000 00:00 0 7fd799600000-7fd799800000 rw-p 00000000 00:00 0 7fd799800000-7fd79a200000 rw-p 00000000 00:00 0 7fd79a200000-7fd79a600000 rw-p 00000000 00:00 0 7fd79a600000-7fd79ae00000 rw-p 00000000 00:00 0 7fd79ae00000-7fd79b000000 rw-p 00000000 00:00 0 7fd79b000000-7fd79ba00000 rw-p 00000000 00:00 0 7fd79ba00000-7fd79bc00000 rw-p 00000000 00:00 0 7fd79bc00000-7fd79c400000 rw-p 00000000 00:00 0 7fd79c400000-7fd79c600000 rw-p 00000000 00:00 0 7fd79c600000-7fd79d000000 rw-p 00000000 00:00 0 7fd79d000000-7fd79d200000 rw-p 00000000 00:00 0 7fd79d200000-7fd79da00000 rw-p 00000000 00:00 0 7fd79da00000-7fd79dc00000 rw-p 00000000 00:00 0 7fd79dc00000-7fd79e600000 rw-p 00000000 00:00 0 7fd79e600000-7fd79e800000 rw-p 00000000 00:00 0 7fd79e800000-7fd79f000000 rw-p 00000000 00:00 0 7fd79f000000-7fd79f200000 rw-p 00000000 00:00 0 7fd79f200000-7fd79fc00000 rw-p 00000000 00:00 0 7fd79fc00000-7fd7a0000000 rw-p 00000000 00:00 0 7fd7a0000000-7fd7a0800000 rw-p 00000000 00:00 0 7fd7a0800000-7fd7a0a00000 rw-p 00000000 00:00 0 7fd7a0a00000-7fd7a1400000 rw-p 00000000 00:00 0 7fd7a1400000-7fd7a147e000 rw-p 00000000 00:00 0 7fd7a147e000-7fd7a1485000 r-xp 00000000 08:06 67120903 /usr/lib64/librt-2.17.so 7fd7a1485000-7fd7a1684000 ---p 00007000 08:06 67120903 /usr/lib64/librt-2.17.so 7fd7a1684000-7fd7a1685000 r--p 00006000 08:06 67120903 /usr/lib64/librt-2.17.so 7fd7a1685000-7fd7a1686000 rw-p 00007000 08:06 67120903 /usr/lib64/librt-2.17.so 7fd7a1686000-7fd7a1690000 r-xp 00000000 08:06 67329598 /usr/lib64/libnuma.so.1 7fd7a1690000-7fd7a1890000 ---p 0000a000 08:06 67329598 /usr/lib64/libnuma.so.1 7fd7a1890000-7fd7a1891000 r--p 0000a000 08:06 67329598 /usr/lib64/libnuma.so.1 7fd7a1891000-7fd7a1892000 rw-p 0000b000 08:06 67329598 /usr/lib64/libnuma.so.1 7fd7a1892000-7fd7a1899000 r-xp 00000000 08:06 139595696 /opt/intel/opencl-1.2-4.5.0.8/lib64/libOpenCL.so.1.2 7fd7a1899000-7fd7a1a98000 ---p 00007000 08:06 139595696 /opt/intel/opencl-1.2-4.5.0.8/lib64/libOpenCL.so.1.2 7fd7a1a98000-7fd7a1a99000 r--p 00006000 08:06 139595696 /opt/intel/opencl-1.2-4.5.0.8/lib64/libOpenCL.so.1.2 7fd7a1a99000-7fd7a1a9a000 rw-p 00007000 08:06 139595696 /opt/intel/opencl-1.2-4.5.0.8/lib64/libOpenCL.so.1.2 7fd7a1a9a000-7fd7a1a9b000 ---p 00000000 00:00 0 7fd7a1a9b000-7fd7a229b000 rw-p 00000000 00:00 0 [stack:5526] 7fd7a229b000-7fd7a229c000 ---p 00000000 00:00 0 7fd7a229c000-7fd7a2a9c000 rw-p 00000000 00:00 0 [stack:5525] 7fd7a2a9c000-7fd7a2a9d000 ---p 00000000 00:00 0 7fd7a2a9d000-7fd7a329d000 rw-p 00000000 00:00 0 [stack:5524] 7fd7a329d000-7fd7a329e000 ---p 00000000 00:00 0 7fd7a329e000-7fd7a3a9e000 rw-p 00000000 00:00 0 [stack:5523] 7fd7a3a9e000-7fd7a3a9f000 ---p 00000000 00:00 0 7fd7a3a9f000-7fd7a429f000 rw-p 00000000 00:00 0 [stack:5522] 7fd7a429f000-7fd7a42a0000 ---p 00000000 00:00 0 7fd7a42a0000-7fd7a4aa0000 rw-p 00000000 00:00 0 [stack:5521] 7fd7a4aa0000-7fd7a4aa1000 ---p 00000000 00:00 0 7fd7a4aa1000-7fd7a52a1000 rw-p 00000000 00:00 0 [stack:5520] 7fd7a52a1000-7fd7a52a2000 ---p 00000000 00:00 0 7fd7a52a2000-7fd7a5aa2000 rw-p 00000000 00:00 0 [stack:5519] 7fd7a5aa2000-7fd7a5aa3000 ---p 00000000 00:00 0 7fd7a5aa3000-7fd7a62a3000 rw-p 00000000 00:00 0 [stack:5518] 7fd7a62a3000-7fd7a62a4000 ---p 00000000 00:00 0 7fd7a62a4000-7fd7a6c00000 rw-p 00000000 00:00 0 [stack:5517] 7fd7a6c00000-7fd7a7800000 rw-p 00000000 00:00 0 7fd7a7800000-7fd7a7b7a000 rw-p 00000000 00:00 0 7fd7a7b7a000-7fd7a7d30000 r-xp 00000000 08:06 67116969 /usr/lib64/libc-2.17.so 7fd7a7d30000-7fd7a7f30000 ---p 001b6000 08:06 67116969 /usr/lib64/libc-2.17.so 7fd7a7f30000-7fd7a7f34000 r--p 001b6000 08:06 67116969 /usr/lib64/libc-2.17.so 7fd7a7f34000-7fd7a7f36000 rw-p 001ba000 08:06 67116969 /usr/lib64/libc-2.17.so 7fd7a7f36000-7fd7a7f3b000 rw-p 00000000 00:00 0 7fd7a7f3b000-7fd7a7f50000 r-xp 00000000 08:06 68649257 /usr/lib64/libgcc_s-4.8.2-20140120.so.1 7fd7a7f50000-7fd7a814f000 ---p 00015000 08:06 68649257 /usr/lib64/libgcc_s-4.8.2-20140120.so.1 7fd7a814f000-7fd7a8150000 r--p 00014000 08:06 68649257 /usr/lib64/libgcc_s-4.8.2-20140120.so.1 7fd7a8150000-7fd7a8151000 rw-p 00015000 08:06 68649257 /usr/lib64/libgcc_s-4.8.2-20140120.so.1 7fd7a8151000-7fd7a8252000 r-xp 00000000 08:06 67116977 /usr/lib64/libm-2.17.so 7fd7a8252000-7fd7a8451000 ---p 00101000 08:06 67116977 /usr/lib64/libm-2.17.so 7fd7a8451000-7fd7a8452000 r--p 00100000 08:06 67116977 /usr/lib64/libm-2.17.so 7fd7a8452000-7fd7a8453000 rw-p 00101000 08:06 67116977 /usr/lib64/libm-2.17.so 7fd7a8453000-7fd7a853c000 r-xp 00000000 08:06 67120932 /usr/lib64/libstdc++.so.6.0.19 7fd7a853c000-7fd7a873b000 ---p 000e9000 08:06 67120932 /usr/lib64/libstdc++.so.6.0.19 7fd7a873b000-7fd7a8743000 r--p 000e8000 08:06 67120932 /usr/lib64/libstdc++.so.6.0.19Stack dump: 0. Running pass 'Intel OpenCL Vectorizer' on module 'Program'. 1. Running pass 'Intel OpenCL VectorizerCore' on function '@__Vectorized_.sum_inter_cost' 2. Running pass 'PacketizeFunction' on function '@__Vectorized_.sum_inter_cost' Aborted (core dumped)
Please help me resolve this problem!