Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Jiawen_L_
Beginner
68 Views

A Problem of Running OpenCL Samples Codes for Intel(R) HD Graphics P4600 by the code-samples Downloaded from Intel

I used the OpenCL 1.2 sample codes downloaded from Intel official website(https://software.intel.com/en-us/intel-opencl-support/code-samples) to test the Intel Corporation Xeon E3-1200 v3 Processor Integrated Graphics Controller(Intel® HD Graphics P4600). I have set up the environment for the GPU successfully and the GPU information are as following codes. But when I run the Intel samples codes programs into the GPU, the following information shows up. Is the problem of the code?(It was not designed for the GPU because of the out of resource) If so, could you please offer me any tested sample codes for Intel® HD Graphics P4600 GPU? If the codes are no problem for Intel GPU, what should I do to run the Intel sample codes into the GPU successfully?

root@storm:/home/storm/Test_Intel/intel_ocl_samples_linux/BitonicSort# ./bitonic_sort -t gpu

Platforms (1):

    [0] Intel(R) OpenCL [Selected]

Devices (1; filtered by type gpu):

    [0] Intel(R) HD Graphics [Selected]

Sort order is ascending

Input size is 33554432 items

OpenCL data alignment is 4096 bytes.

Executing OpenCL kernel...

[ ERROR ] Sample application specific error: OpenCL error CL_OUT_OF_RESOURCES happened in file BitonicSort.cpp at line 172.

 

root@storm:/home/storm/Test_Intel/intel_ocl_samples_linux/GEMM# ./gemm -t gpu

Platforms (1):

    [0] Intel(R) OpenCL [Selected]

Devices (1; filtered by type gpu):

    [0] Intel(R) HD Graphics [Selected]

Build program options: "-DT=float -DTILE_SIZE_M=1 -DTILE_GROUP_M=16 -DTILE_SIZE_N=128 -DTILE_GROUP_N=1 -DTILE_SIZE_K=8"

Running gemm_nn kernel with matrix size: 3968x3968

Memory row stride to ensure necessary alignment: 15872 bytes

Size of memory region for one matrix: 62980096 bytes

Using alpha = 0.57599 and beta = 0.872412

[ ERROR ] Sample application specific error: OpenCL error CL_OUT_OF_RESOURCES happened in file gemm.cpp at line 295.

 

 

root@storm:/home/storm/Test_Intel/intel_ocl_samples_linux/CapsBasic# ./capsbasic 

Number of available platforms: 1

Platform names:

    [0] Intel(R) OpenCL [Selected]

Number of devices available for each type:

    CL_DEVICE_TYPE_CPU: 1

    CL_DEVICE_TYPE_GPU: 1

    CL_DEVICE_TYPE_ACCELERATOR: 0

 

*** Detailed information for each device ***

CL_DEVICE_TYPE_GPU[0]

    CL_DEVICE_NAME: Intel(R) HD Graphics

    CL_DEVICE_AVAILABLE: 1

    CL_DEVICE_VENDOR: Intel(R) Corporation

    CL_DEVICE_PROFILE: FULL_PROFILE

    CL_DEVICE_VERSION: OpenCL 1.2 

    CL_DRIVER_VERSION: 16.4.2.1.39163

    CL_DEVICE_OPENCL_C_VERSION: OpenCL C 1.2 

    CL_DEVICE_MAX_COMPUTE_UNITS: 20

    CL_DEVICE_MAX_CLOCK_FREQUENCY: 1300

    CL_DEVICE_MAX_WORK_GROUP_SIZE: 512

    CL_DEVICE_ADDRESS_BITS: 64

    CL_DEVICE_MEM_BASE_ADDR_ALIGN: 1024

    CL_DEVICE_MAX_MEM_ALLOC_SIZE: 427399577

    CL_DEVICE_GLOBAL_MEM_SIZE: 1709598311

    CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE: 65536

    CL_DEVICE_GLOBAL_MEM_CACHE_SIZE: 262144

    CL_DEVICE_GLOBAL_MEM_CACHELINE_SIZE: 64

    CL_DEVICE_LOCAL_MEM_SIZE: 65536

    CL_DEVICE_PROFILING_TIMER_RESOLUTION: 80

    CL_DEVICE_IMAGE_SUPPORT: 1

    CL_DEVICE_ERROR_CORRECTION_SUPPORT: 0

    CL_DEVICE_HOST_UNIFIED_MEMORY: 1

    CL_DEVICE_EXTENSIONS: cl_intel_accelerator cl_intel_advanced_motion_estimation cl_intel_motion_estimation cl_intel_subgroups cl_intel_va_api_media_sharing cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_icd cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_spir 

    CL_DEVICE_PREFERRED_VECTOR_WIDTH_INT: 1

    CL_DEVICE_PREFERRED_VECTOR_WIDTH_LONG: 1

    CL_DEVICE_PREFERRED_VECTOR_WIDTH_FLOAT: 1

    CL_DEVICE_PREFERRED_VECTOR_WIDTH_DOUBLE: 0

    CL_DEVICE_NATIVE_VECTOR_WIDTH_INT: 1

    CL_DEVICE_NATIVE_VECTOR_WIDTH_LONG: 1

    CL_DEVICE_NATIVE_VECTOR_WIDTH_FLOAT: 1

    CL_DEVICE_NATIVE_VECTOR_WIDTH_DOUBLE: 0

 

 

 

 

 

0 Kudos
1 Reply
Robert_I_Intel
Employee
68 Views

Hi Jiawen,

Could you please try to run these samples with smaller input sizes and see if they work for you?

Could you also please try the sample here https://software.intel.com/en-us/articles/sgemm-for-intel-processor-graphics (at the very bottom of the page is the download link)? It should be much more recent than the samples above.