<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Can't install OpenCL driver for Intel HD Graphics 620 from a Gen7 (Kaby Lake) processor in GPU Compute Software</title>
    <link>https://community.intel.com/t5/GPU-Compute-Software/Can-t-install-OpenCL-driver-for-Intel-HD-Graphics-620-from-a/m-p/1303473#M200</link>
    <description>&lt;P&gt;Hello.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;I'm having some trouble running OpenCL on my system. My clinfo and other OpenCL applications I have say that an OpenCL context could not be created (error code -2).&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;I've followed this excellent article (&lt;A href="https://github.com/bashbaug/OpenCLPapers/blob/master/OpenCLOnLinux.asciidoc" target="_blank"&gt;https://github.com/bashbaug/OpenCLPapers/blob/master/OpenCLOnLinux.asciidoc&lt;/A&gt;)&amp;nbsp;and was able to see that my system has a working libOpenCL.so and that it has a readable .icd file.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;I had a previous working installation of OpenCL, and my .icd file is pointing to a file that doesn't exists (/usr/lib/x86_64-linux-gnu/intel-opencl/libigdrcl.so). I've assumed that the pointed file was a driver file the was missing, so I downloaded the file given in&amp;nbsp;&lt;A href="https://software.intel.com/content/www/us/en/develop/articles/legacy-opencl-drivers.html#latest_linux_driver" target="_blank"&gt;https://software.intel.com/content/www/us/en/develop/articles/legacy-opencl-drivers.html#latest_linux_driver.&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;By installing the file mentioned above I got the directory /opt/intel/opencl which has a libigdrcl.so file. I wrote another .icd pointing to this new file but clinfo still tells me it was impossible to create an OpenCL context.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Does anyone know whats is going on? The problem is with the driver? Where can I get correct Linux drivers for my specific hardware?&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;P.S.: here is the output of clinfo:&lt;/P&gt;
&lt;P&gt;Number of platforms 1&lt;BR /&gt;Platform Name Intel(R) OpenCL&lt;BR /&gt;Platform Vendor Intel(R) Corporation&lt;BR /&gt;Platform Version OpenCL 2.1 LINUX&lt;BR /&gt;Platform Profile FULL_PROFILE&lt;BR /&gt;Platform 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_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_khr_il_program cl_intel_unified_shared_memory_preview cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer&lt;BR /&gt;Platform Host timer resolution 1ns&lt;BR /&gt;Platform Extensions function suffix INTEL&lt;/P&gt;
&lt;P&gt;Platform Name Intel(R) OpenCL&lt;BR /&gt;Number of devices 1&lt;BR /&gt;Device Name Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz&lt;BR /&gt;Device Vendor Intel(R) Corporation&lt;BR /&gt;Device Vendor ID 0x8086&lt;BR /&gt;Device Version OpenCL 2.1 (Build 0)&lt;BR /&gt;Driver Version 2020.10.3.0.04&lt;BR /&gt;Device OpenCL C Version OpenCL C 2.0&lt;BR /&gt;Device Type CPU&lt;BR /&gt;Device Profile FULL_PROFILE&lt;BR /&gt;Device Available Yes&lt;BR /&gt;Compiler Available Yes&lt;BR /&gt;Linker Available Yes&lt;BR /&gt;Max compute units 4&lt;BR /&gt;Max clock frequency 2700MHz&lt;BR /&gt;Device Partition (core)&lt;BR /&gt;Max number of sub-devices 4&lt;BR /&gt;Supported partition types by counts, equally, by names (Intel)&lt;BR /&gt;Supported affinity domains (n/a)&lt;BR /&gt;Max work item dimensions 3&lt;BR /&gt;Max work item sizes 8192x8192x8192&lt;BR /&gt;Max work group size 8192&lt;BR /&gt;Preferred work group size multiple &amp;lt;getWGsizes:1171: create context : error -2&amp;gt;&lt;BR /&gt;Max sub-groups per work group 1&lt;BR /&gt;Preferred / native vector sizes&lt;BR /&gt;char 1 / 32&lt;BR /&gt;short 1 / 16&lt;BR /&gt;int 1 / 8&lt;BR /&gt;long 1 / 4&lt;BR /&gt;half 0 / 0 (n/a)&lt;BR /&gt;float 1 / 8&lt;BR /&gt;double 1 / 4 (cl_khr_fp64)&lt;BR /&gt;Half-precision Floating-point support (n/a)&lt;BR /&gt;Single-precision Floating-point support (core)&lt;BR /&gt;Denormals Yes&lt;BR /&gt;Infinity and NANs Yes&lt;BR /&gt;Round to nearest Yes&lt;BR /&gt;Round to zero No&lt;BR /&gt;Round to infinity No&lt;BR /&gt;IEEE754-2008 fused multiply-add No&lt;BR /&gt;Support is emulated in software No&lt;BR /&gt;Correctly-rounded divide and sqrt operations No&lt;BR /&gt;Double-precision Floating-point support (cl_khr_fp64)&lt;BR /&gt;Denormals Yes&lt;BR /&gt;Infinity and NANs Yes&lt;BR /&gt;Round to nearest Yes&lt;BR /&gt;Round to zero Yes&lt;BR /&gt;Round to infinity Yes&lt;BR /&gt;IEEE754-2008 fused multiply-add Yes&lt;BR /&gt;Support is emulated in software No&lt;BR /&gt;Address bits 64, Little-Endian&lt;BR /&gt;Global memory size 16684314624 (15.54GiB)&lt;BR /&gt;Error Correction support No&lt;BR /&gt;Max memory allocation 4171078656 (3.885GiB)&lt;BR /&gt;Unified memory for Host and Device Yes&lt;BR /&gt;Shared Virtual Memory (SVM) capabilities (core)&lt;BR /&gt;Coarse-grained buffer sharing Yes&lt;BR /&gt;Fine-grained buffer sharing Yes&lt;BR /&gt;Fine-grained system sharing Yes&lt;BR /&gt;Atomics Yes&lt;BR /&gt;Minimum alignment for any data type 128 bytes&lt;BR /&gt;Alignment of base address 1024 bits (128 bytes)&lt;BR /&gt;Preferred alignment for atomics&lt;BR /&gt;SVM 64 bytes&lt;BR /&gt;Global 64 bytes&lt;BR /&gt;Local 0 bytes&lt;BR /&gt;Max size for global variable 65536 (64KiB)&lt;BR /&gt;Preferred total size of global vars 65536 (64KiB)&lt;BR /&gt;Global Memory cache type Read/Write&lt;BR /&gt;Global Memory cache size 262144 (256KiB)&lt;BR /&gt;Global Memory cache line size 64 bytes&lt;BR /&gt;Image support Yes&lt;BR /&gt;Max number of samplers per kernel 480&lt;BR /&gt;Max size for 1D images from buffer 260692416 pixels&lt;BR /&gt;Max 1D or 2D image array size 2048 images&lt;BR /&gt;Base address alignment for 2D image buffers 64 bytes&lt;BR /&gt;Pitch alignment for 2D image buffers 64 pixels&lt;BR /&gt;Max 2D image size 16384x16384 pixels&lt;BR /&gt;Max 3D image size 2048x2048x2048 pixels&lt;BR /&gt;Max number of read image args 480&lt;BR /&gt;Max number of write image args 480&lt;BR /&gt;Max number of read/write image args 480&lt;BR /&gt;Max number of pipe args 16&lt;BR /&gt;Max active pipe reservations 65535&lt;BR /&gt;Max pipe packet size 1024&lt;BR /&gt;Local memory type Global&lt;BR /&gt;Local memory size 32768 (32KiB)&lt;BR /&gt;Max number of constant args 480&lt;BR /&gt;Max constant buffer size 131072 (128KiB)&lt;BR /&gt;Max size of kernel argument 3840 (3.75KiB)&lt;BR /&gt;Queue properties (on host)&lt;BR /&gt;Out-of-order execution Yes&lt;BR /&gt;Profiling Yes&lt;BR /&gt;Local thread execution (Intel) Yes&lt;BR /&gt;Queue properties (on device)&lt;BR /&gt;Out-of-order execution Yes&lt;BR /&gt;Profiling Yes&lt;BR /&gt;Preferred size 4294967295 (4GiB)&lt;BR /&gt;Max size 4294967295 (4GiB)&lt;BR /&gt;Max queues on device 4294967295&lt;BR /&gt;Max events on device 4294967295&lt;BR /&gt;Prefer user sync for interop No&lt;BR /&gt;Profiling timer resolution 1ns&lt;BR /&gt;Execution capabilities&lt;BR /&gt;Run OpenCL kernels Yes&lt;BR /&gt;Run native kernels Yes&lt;BR /&gt;Sub-group independent forward progress No&lt;BR /&gt;IL version SPIR-V_1.0&lt;BR /&gt;SPIR versions 1.2&lt;BR /&gt;printf() buffer size 1048576 (1024KiB)&lt;BR /&gt;Built-in kernels (n/a)&lt;BR /&gt;Device 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_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_khr_il_program cl_intel_unified_shared_memory_preview cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer&lt;/P&gt;
&lt;P&gt;NULL platform behavior&lt;BR /&gt;clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Intel(R) OpenCL&lt;BR /&gt;clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [INTEL]&lt;BR /&gt;clCreateContext(NULL, ...) [default] &amp;lt;checkNullCtx:2758: create context with device from default platform : error -2&amp;gt;&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) No devices available in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices available in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) No devices available in platform&lt;/P&gt;
&lt;P&gt;ICD loader properties&lt;BR /&gt;ICD loader Name OpenCL ICD Loader&lt;BR /&gt;ICD loader Vendor OCL Icd free software&lt;BR /&gt;ICD loader Version 2.2.11&lt;BR /&gt;ICD loader Profile OpenCL 2.1&lt;/P&gt;
&lt;P&gt;Number of platforms 1&lt;BR /&gt;Platform Name Intel(R) OpenCL&lt;BR /&gt;Platform Vendor Intel(R) Corporation&lt;BR /&gt;Platform Version OpenCL 2.1 LINUX&lt;BR /&gt;Platform Profile FULL_PROFILE&lt;BR /&gt;Platform 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_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_khr_il_program cl_intel_unified_shared_memory_preview cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer&lt;BR /&gt;Platform Host timer resolution 1ns&lt;BR /&gt;Platform Extensions function suffix INTEL&lt;/P&gt;
&lt;P&gt;Platform Name Intel(R) OpenCL&lt;BR /&gt;Number of devices 1&lt;BR /&gt;Device Name Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz&lt;BR /&gt;Device Vendor Intel(R) Corporation&lt;BR /&gt;Device Vendor ID 0x8086&lt;BR /&gt;Device Version OpenCL 2.1 (Build 0)&lt;BR /&gt;Driver Version 2020.10.3.0.04&lt;BR /&gt;Device OpenCL C Version OpenCL C 2.0&lt;BR /&gt;Device Type CPU&lt;BR /&gt;Device Profile FULL_PROFILE&lt;BR /&gt;Device Available Yes&lt;BR /&gt;Compiler Available Yes&lt;BR /&gt;Linker Available Yes&lt;BR /&gt;Max compute units 4&lt;BR /&gt;Max clock frequency 2700MHz&lt;BR /&gt;Device Partition (core)&lt;BR /&gt;Max number of sub-devices 4&lt;BR /&gt;Supported partition types by counts, equally, by names (Intel)&lt;BR /&gt;Supported affinity domains (n/a)&lt;BR /&gt;Max work item dimensions 3&lt;BR /&gt;Max work item sizes 8192x8192x8192&lt;BR /&gt;Max work group size 8192&lt;BR /&gt;Preferred work group size multiple &amp;lt;getWGsizes:1171: create context : error -2&amp;gt;&lt;BR /&gt;Max sub-groups per work group 1&lt;BR /&gt;Preferred / native vector sizes&lt;BR /&gt;char 1 / 32&lt;BR /&gt;short 1 / 16&lt;BR /&gt;int 1 / 8&lt;BR /&gt;long 1 / 4&lt;BR /&gt;half 0 / 0 (n/a)&lt;BR /&gt;float 1 / 8&lt;BR /&gt;double 1 / 4 (cl_khr_fp64)&lt;BR /&gt;Half-precision Floating-point support (n/a)&lt;BR /&gt;Single-precision Floating-point support (core)&lt;BR /&gt;Denormals Yes&lt;BR /&gt;Infinity and NANs Yes&lt;BR /&gt;Round to nearest Yes&lt;BR /&gt;Round to zero No&lt;BR /&gt;Round to infinity No&lt;BR /&gt;IEEE754-2008 fused multiply-add No&lt;BR /&gt;Support is emulated in software No&lt;BR /&gt;Correctly-rounded divide and sqrt operations No&lt;BR /&gt;Double-precision Floating-point support (cl_khr_fp64)&lt;BR /&gt;Denormals Yes&lt;BR /&gt;Infinity and NANs Yes&lt;BR /&gt;Round to nearest Yes&lt;BR /&gt;Round to zero Yes&lt;BR /&gt;Round to infinity Yes&lt;BR /&gt;IEEE754-2008 fused multiply-add Yes&lt;BR /&gt;Support is emulated in software No&lt;BR /&gt;Address bits 64, Little-Endian&lt;BR /&gt;Global memory size 16684314624 (15.54GiB)&lt;BR /&gt;Error Correction support No&lt;BR /&gt;Max memory allocation 4171078656 (3.885GiB)&lt;BR /&gt;Unified memory for Host and Device Yes&lt;BR /&gt;Shared Virtual Memory (SVM) capabilities (core)&lt;BR /&gt;Coarse-grained buffer sharing Yes&lt;BR /&gt;Fine-grained buffer sharing Yes&lt;BR /&gt;Fine-grained system sharing Yes&lt;BR /&gt;Atomics Yes&lt;BR /&gt;Minimum alignment for any data type 128 bytes&lt;BR /&gt;Alignment of base address 1024 bits (128 bytes)&lt;BR /&gt;Preferred alignment for atomics&lt;BR /&gt;SVM 64 bytes&lt;BR /&gt;Global 64 bytes&lt;BR /&gt;Local 0 bytes&lt;BR /&gt;Max size for global variable 65536 (64KiB)&lt;BR /&gt;Preferred total size of global vars 65536 (64KiB)&lt;BR /&gt;Global Memory cache type Read/Write&lt;BR /&gt;Global Memory cache size 262144 (256KiB)&lt;BR /&gt;Global Memory cache line size 64 bytes&lt;BR /&gt;Image support Yes&lt;BR /&gt;Max number of samplers per kernel 480&lt;BR /&gt;Max size for 1D images from buffer 260692416 pixels&lt;BR /&gt;Max 1D or 2D image array size 2048 images&lt;BR /&gt;Base address alignment for 2D image buffers 64 bytes&lt;BR /&gt;Pitch alignment for 2D image buffers 64 pixels&lt;BR /&gt;Max 2D image size 16384x16384 pixels&lt;BR /&gt;Max 3D image size 2048x2048x2048 pixels&lt;BR /&gt;Max number of read image args 480&lt;BR /&gt;Max number of write image args 480&lt;BR /&gt;Max number of read/write image args 480&lt;BR /&gt;Max number of pipe args 16&lt;BR /&gt;Max active pipe reservations 65535&lt;BR /&gt;Max pipe packet size 1024&lt;BR /&gt;Local memory type Global&lt;BR /&gt;Local memory size 32768 (32KiB)&lt;BR /&gt;Max number of constant args 480&lt;BR /&gt;Max constant buffer size 131072 (128KiB)&lt;BR /&gt;Max size of kernel argument 3840 (3.75KiB)&lt;BR /&gt;Queue properties (on host)&lt;BR /&gt;Out-of-order execution Yes&lt;BR /&gt;Profiling Yes&lt;BR /&gt;Local thread execution (Intel) Yes&lt;BR /&gt;Queue properties (on device)&lt;BR /&gt;Out-of-order execution Yes&lt;BR /&gt;Profiling Yes&lt;BR /&gt;Preferred size 4294967295 (4GiB)&lt;BR /&gt;Max size 4294967295 (4GiB)&lt;BR /&gt;Max queues on device 4294967295&lt;BR /&gt;Max events on device 4294967295&lt;BR /&gt;Prefer user sync for interop No&lt;BR /&gt;Profiling timer resolution 1ns&lt;BR /&gt;Execution capabilities&lt;BR /&gt;Run OpenCL kernels Yes&lt;BR /&gt;Run native kernels Yes&lt;BR /&gt;Sub-group independent forward progress No&lt;BR /&gt;IL version SPIR-V_1.0&lt;BR /&gt;SPIR versions 1.2&lt;BR /&gt;printf() buffer size 1048576 (1024KiB)&lt;BR /&gt;Built-in kernels (n/a)&lt;BR /&gt;Device 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_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_khr_il_program cl_intel_unified_shared_memory_preview cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer&lt;/P&gt;
&lt;P&gt;NULL platform behavior&lt;BR /&gt;clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Intel(R) OpenCL&lt;BR /&gt;clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [INTEL]&lt;BR /&gt;clCreateContext(NULL, ...) [default] &amp;lt;checkNullCtx:2758: create context with device from default platform : error -2&amp;gt;&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) No devices available in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices available in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) No devices available in platform&lt;/P&gt;
&lt;P&gt;ICD loader properties&lt;BR /&gt;ICD loader Name OpenCL ICD Loader&lt;BR /&gt;ICD loader Vendor OCL Icd free software&lt;BR /&gt;ICD loader Version 2.2.11&lt;BR /&gt;ICD loader Profile OpenCL 2.1&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Wed, 04 Aug 2021 00:51:47 GMT</pubDate>
    <dc:creator>ducbueno__</dc:creator>
    <dc:date>2021-08-04T00:51:47Z</dc:date>
    <item>
      <title>Can't install OpenCL driver for Intel HD Graphics 620 from a Gen7 (Kaby Lake) processor</title>
      <link>https://community.intel.com/t5/GPU-Compute-Software/Can-t-install-OpenCL-driver-for-Intel-HD-Graphics-620-from-a/m-p/1303473#M200</link>
      <description>&lt;P&gt;Hello.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;I'm having some trouble running OpenCL on my system. My clinfo and other OpenCL applications I have say that an OpenCL context could not be created (error code -2).&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;I've followed this excellent article (&lt;A href="https://github.com/bashbaug/OpenCLPapers/blob/master/OpenCLOnLinux.asciidoc" target="_blank"&gt;https://github.com/bashbaug/OpenCLPapers/blob/master/OpenCLOnLinux.asciidoc&lt;/A&gt;)&amp;nbsp;and was able to see that my system has a working libOpenCL.so and that it has a readable .icd file.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;I had a previous working installation of OpenCL, and my .icd file is pointing to a file that doesn't exists (/usr/lib/x86_64-linux-gnu/intel-opencl/libigdrcl.so). I've assumed that the pointed file was a driver file the was missing, so I downloaded the file given in&amp;nbsp;&lt;A href="https://software.intel.com/content/www/us/en/develop/articles/legacy-opencl-drivers.html#latest_linux_driver" target="_blank"&gt;https://software.intel.com/content/www/us/en/develop/articles/legacy-opencl-drivers.html#latest_linux_driver.&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;By installing the file mentioned above I got the directory /opt/intel/opencl which has a libigdrcl.so file. I wrote another .icd pointing to this new file but clinfo still tells me it was impossible to create an OpenCL context.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Does anyone know whats is going on? The problem is with the driver? Where can I get correct Linux drivers for my specific hardware?&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;P.S.: here is the output of clinfo:&lt;/P&gt;
&lt;P&gt;Number of platforms 1&lt;BR /&gt;Platform Name Intel(R) OpenCL&lt;BR /&gt;Platform Vendor Intel(R) Corporation&lt;BR /&gt;Platform Version OpenCL 2.1 LINUX&lt;BR /&gt;Platform Profile FULL_PROFILE&lt;BR /&gt;Platform 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_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_khr_il_program cl_intel_unified_shared_memory_preview cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer&lt;BR /&gt;Platform Host timer resolution 1ns&lt;BR /&gt;Platform Extensions function suffix INTEL&lt;/P&gt;
&lt;P&gt;Platform Name Intel(R) OpenCL&lt;BR /&gt;Number of devices 1&lt;BR /&gt;Device Name Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz&lt;BR /&gt;Device Vendor Intel(R) Corporation&lt;BR /&gt;Device Vendor ID 0x8086&lt;BR /&gt;Device Version OpenCL 2.1 (Build 0)&lt;BR /&gt;Driver Version 2020.10.3.0.04&lt;BR /&gt;Device OpenCL C Version OpenCL C 2.0&lt;BR /&gt;Device Type CPU&lt;BR /&gt;Device Profile FULL_PROFILE&lt;BR /&gt;Device Available Yes&lt;BR /&gt;Compiler Available Yes&lt;BR /&gt;Linker Available Yes&lt;BR /&gt;Max compute units 4&lt;BR /&gt;Max clock frequency 2700MHz&lt;BR /&gt;Device Partition (core)&lt;BR /&gt;Max number of sub-devices 4&lt;BR /&gt;Supported partition types by counts, equally, by names (Intel)&lt;BR /&gt;Supported affinity domains (n/a)&lt;BR /&gt;Max work item dimensions 3&lt;BR /&gt;Max work item sizes 8192x8192x8192&lt;BR /&gt;Max work group size 8192&lt;BR /&gt;Preferred work group size multiple &amp;lt;getWGsizes:1171: create context : error -2&amp;gt;&lt;BR /&gt;Max sub-groups per work group 1&lt;BR /&gt;Preferred / native vector sizes&lt;BR /&gt;char 1 / 32&lt;BR /&gt;short 1 / 16&lt;BR /&gt;int 1 / 8&lt;BR /&gt;long 1 / 4&lt;BR /&gt;half 0 / 0 (n/a)&lt;BR /&gt;float 1 / 8&lt;BR /&gt;double 1 / 4 (cl_khr_fp64)&lt;BR /&gt;Half-precision Floating-point support (n/a)&lt;BR /&gt;Single-precision Floating-point support (core)&lt;BR /&gt;Denormals Yes&lt;BR /&gt;Infinity and NANs Yes&lt;BR /&gt;Round to nearest Yes&lt;BR /&gt;Round to zero No&lt;BR /&gt;Round to infinity No&lt;BR /&gt;IEEE754-2008 fused multiply-add No&lt;BR /&gt;Support is emulated in software No&lt;BR /&gt;Correctly-rounded divide and sqrt operations No&lt;BR /&gt;Double-precision Floating-point support (cl_khr_fp64)&lt;BR /&gt;Denormals Yes&lt;BR /&gt;Infinity and NANs Yes&lt;BR /&gt;Round to nearest Yes&lt;BR /&gt;Round to zero Yes&lt;BR /&gt;Round to infinity Yes&lt;BR /&gt;IEEE754-2008 fused multiply-add Yes&lt;BR /&gt;Support is emulated in software No&lt;BR /&gt;Address bits 64, Little-Endian&lt;BR /&gt;Global memory size 16684314624 (15.54GiB)&lt;BR /&gt;Error Correction support No&lt;BR /&gt;Max memory allocation 4171078656 (3.885GiB)&lt;BR /&gt;Unified memory for Host and Device Yes&lt;BR /&gt;Shared Virtual Memory (SVM) capabilities (core)&lt;BR /&gt;Coarse-grained buffer sharing Yes&lt;BR /&gt;Fine-grained buffer sharing Yes&lt;BR /&gt;Fine-grained system sharing Yes&lt;BR /&gt;Atomics Yes&lt;BR /&gt;Minimum alignment for any data type 128 bytes&lt;BR /&gt;Alignment of base address 1024 bits (128 bytes)&lt;BR /&gt;Preferred alignment for atomics&lt;BR /&gt;SVM 64 bytes&lt;BR /&gt;Global 64 bytes&lt;BR /&gt;Local 0 bytes&lt;BR /&gt;Max size for global variable 65536 (64KiB)&lt;BR /&gt;Preferred total size of global vars 65536 (64KiB)&lt;BR /&gt;Global Memory cache type Read/Write&lt;BR /&gt;Global Memory cache size 262144 (256KiB)&lt;BR /&gt;Global Memory cache line size 64 bytes&lt;BR /&gt;Image support Yes&lt;BR /&gt;Max number of samplers per kernel 480&lt;BR /&gt;Max size for 1D images from buffer 260692416 pixels&lt;BR /&gt;Max 1D or 2D image array size 2048 images&lt;BR /&gt;Base address alignment for 2D image buffers 64 bytes&lt;BR /&gt;Pitch alignment for 2D image buffers 64 pixels&lt;BR /&gt;Max 2D image size 16384x16384 pixels&lt;BR /&gt;Max 3D image size 2048x2048x2048 pixels&lt;BR /&gt;Max number of read image args 480&lt;BR /&gt;Max number of write image args 480&lt;BR /&gt;Max number of read/write image args 480&lt;BR /&gt;Max number of pipe args 16&lt;BR /&gt;Max active pipe reservations 65535&lt;BR /&gt;Max pipe packet size 1024&lt;BR /&gt;Local memory type Global&lt;BR /&gt;Local memory size 32768 (32KiB)&lt;BR /&gt;Max number of constant args 480&lt;BR /&gt;Max constant buffer size 131072 (128KiB)&lt;BR /&gt;Max size of kernel argument 3840 (3.75KiB)&lt;BR /&gt;Queue properties (on host)&lt;BR /&gt;Out-of-order execution Yes&lt;BR /&gt;Profiling Yes&lt;BR /&gt;Local thread execution (Intel) Yes&lt;BR /&gt;Queue properties (on device)&lt;BR /&gt;Out-of-order execution Yes&lt;BR /&gt;Profiling Yes&lt;BR /&gt;Preferred size 4294967295 (4GiB)&lt;BR /&gt;Max size 4294967295 (4GiB)&lt;BR /&gt;Max queues on device 4294967295&lt;BR /&gt;Max events on device 4294967295&lt;BR /&gt;Prefer user sync for interop No&lt;BR /&gt;Profiling timer resolution 1ns&lt;BR /&gt;Execution capabilities&lt;BR /&gt;Run OpenCL kernels Yes&lt;BR /&gt;Run native kernels Yes&lt;BR /&gt;Sub-group independent forward progress No&lt;BR /&gt;IL version SPIR-V_1.0&lt;BR /&gt;SPIR versions 1.2&lt;BR /&gt;printf() buffer size 1048576 (1024KiB)&lt;BR /&gt;Built-in kernels (n/a)&lt;BR /&gt;Device 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_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_khr_il_program cl_intel_unified_shared_memory_preview cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer&lt;/P&gt;
&lt;P&gt;NULL platform behavior&lt;BR /&gt;clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Intel(R) OpenCL&lt;BR /&gt;clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [INTEL]&lt;BR /&gt;clCreateContext(NULL, ...) [default] &amp;lt;checkNullCtx:2758: create context with device from default platform : error -2&amp;gt;&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) No devices available in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices available in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) No devices available in platform&lt;/P&gt;
&lt;P&gt;ICD loader properties&lt;BR /&gt;ICD loader Name OpenCL ICD Loader&lt;BR /&gt;ICD loader Vendor OCL Icd free software&lt;BR /&gt;ICD loader Version 2.2.11&lt;BR /&gt;ICD loader Profile OpenCL 2.1&lt;/P&gt;
&lt;P&gt;Number of platforms 1&lt;BR /&gt;Platform Name Intel(R) OpenCL&lt;BR /&gt;Platform Vendor Intel(R) Corporation&lt;BR /&gt;Platform Version OpenCL 2.1 LINUX&lt;BR /&gt;Platform Profile FULL_PROFILE&lt;BR /&gt;Platform 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_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_khr_il_program cl_intel_unified_shared_memory_preview cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer&lt;BR /&gt;Platform Host timer resolution 1ns&lt;BR /&gt;Platform Extensions function suffix INTEL&lt;/P&gt;
&lt;P&gt;Platform Name Intel(R) OpenCL&lt;BR /&gt;Number of devices 1&lt;BR /&gt;Device Name Intel(R) Core(TM) i7-7500U CPU @ 2.70GHz&lt;BR /&gt;Device Vendor Intel(R) Corporation&lt;BR /&gt;Device Vendor ID 0x8086&lt;BR /&gt;Device Version OpenCL 2.1 (Build 0)&lt;BR /&gt;Driver Version 2020.10.3.0.04&lt;BR /&gt;Device OpenCL C Version OpenCL C 2.0&lt;BR /&gt;Device Type CPU&lt;BR /&gt;Device Profile FULL_PROFILE&lt;BR /&gt;Device Available Yes&lt;BR /&gt;Compiler Available Yes&lt;BR /&gt;Linker Available Yes&lt;BR /&gt;Max compute units 4&lt;BR /&gt;Max clock frequency 2700MHz&lt;BR /&gt;Device Partition (core)&lt;BR /&gt;Max number of sub-devices 4&lt;BR /&gt;Supported partition types by counts, equally, by names (Intel)&lt;BR /&gt;Supported affinity domains (n/a)&lt;BR /&gt;Max work item dimensions 3&lt;BR /&gt;Max work item sizes 8192x8192x8192&lt;BR /&gt;Max work group size 8192&lt;BR /&gt;Preferred work group size multiple &amp;lt;getWGsizes:1171: create context : error -2&amp;gt;&lt;BR /&gt;Max sub-groups per work group 1&lt;BR /&gt;Preferred / native vector sizes&lt;BR /&gt;char 1 / 32&lt;BR /&gt;short 1 / 16&lt;BR /&gt;int 1 / 8&lt;BR /&gt;long 1 / 4&lt;BR /&gt;half 0 / 0 (n/a)&lt;BR /&gt;float 1 / 8&lt;BR /&gt;double 1 / 4 (cl_khr_fp64)&lt;BR /&gt;Half-precision Floating-point support (n/a)&lt;BR /&gt;Single-precision Floating-point support (core)&lt;BR /&gt;Denormals Yes&lt;BR /&gt;Infinity and NANs Yes&lt;BR /&gt;Round to nearest Yes&lt;BR /&gt;Round to zero No&lt;BR /&gt;Round to infinity No&lt;BR /&gt;IEEE754-2008 fused multiply-add No&lt;BR /&gt;Support is emulated in software No&lt;BR /&gt;Correctly-rounded divide and sqrt operations No&lt;BR /&gt;Double-precision Floating-point support (cl_khr_fp64)&lt;BR /&gt;Denormals Yes&lt;BR /&gt;Infinity and NANs Yes&lt;BR /&gt;Round to nearest Yes&lt;BR /&gt;Round to zero Yes&lt;BR /&gt;Round to infinity Yes&lt;BR /&gt;IEEE754-2008 fused multiply-add Yes&lt;BR /&gt;Support is emulated in software No&lt;BR /&gt;Address bits 64, Little-Endian&lt;BR /&gt;Global memory size 16684314624 (15.54GiB)&lt;BR /&gt;Error Correction support No&lt;BR /&gt;Max memory allocation 4171078656 (3.885GiB)&lt;BR /&gt;Unified memory for Host and Device Yes&lt;BR /&gt;Shared Virtual Memory (SVM) capabilities (core)&lt;BR /&gt;Coarse-grained buffer sharing Yes&lt;BR /&gt;Fine-grained buffer sharing Yes&lt;BR /&gt;Fine-grained system sharing Yes&lt;BR /&gt;Atomics Yes&lt;BR /&gt;Minimum alignment for any data type 128 bytes&lt;BR /&gt;Alignment of base address 1024 bits (128 bytes)&lt;BR /&gt;Preferred alignment for atomics&lt;BR /&gt;SVM 64 bytes&lt;BR /&gt;Global 64 bytes&lt;BR /&gt;Local 0 bytes&lt;BR /&gt;Max size for global variable 65536 (64KiB)&lt;BR /&gt;Preferred total size of global vars 65536 (64KiB)&lt;BR /&gt;Global Memory cache type Read/Write&lt;BR /&gt;Global Memory cache size 262144 (256KiB)&lt;BR /&gt;Global Memory cache line size 64 bytes&lt;BR /&gt;Image support Yes&lt;BR /&gt;Max number of samplers per kernel 480&lt;BR /&gt;Max size for 1D images from buffer 260692416 pixels&lt;BR /&gt;Max 1D or 2D image array size 2048 images&lt;BR /&gt;Base address alignment for 2D image buffers 64 bytes&lt;BR /&gt;Pitch alignment for 2D image buffers 64 pixels&lt;BR /&gt;Max 2D image size 16384x16384 pixels&lt;BR /&gt;Max 3D image size 2048x2048x2048 pixels&lt;BR /&gt;Max number of read image args 480&lt;BR /&gt;Max number of write image args 480&lt;BR /&gt;Max number of read/write image args 480&lt;BR /&gt;Max number of pipe args 16&lt;BR /&gt;Max active pipe reservations 65535&lt;BR /&gt;Max pipe packet size 1024&lt;BR /&gt;Local memory type Global&lt;BR /&gt;Local memory size 32768 (32KiB)&lt;BR /&gt;Max number of constant args 480&lt;BR /&gt;Max constant buffer size 131072 (128KiB)&lt;BR /&gt;Max size of kernel argument 3840 (3.75KiB)&lt;BR /&gt;Queue properties (on host)&lt;BR /&gt;Out-of-order execution Yes&lt;BR /&gt;Profiling Yes&lt;BR /&gt;Local thread execution (Intel) Yes&lt;BR /&gt;Queue properties (on device)&lt;BR /&gt;Out-of-order execution Yes&lt;BR /&gt;Profiling Yes&lt;BR /&gt;Preferred size 4294967295 (4GiB)&lt;BR /&gt;Max size 4294967295 (4GiB)&lt;BR /&gt;Max queues on device 4294967295&lt;BR /&gt;Max events on device 4294967295&lt;BR /&gt;Prefer user sync for interop No&lt;BR /&gt;Profiling timer resolution 1ns&lt;BR /&gt;Execution capabilities&lt;BR /&gt;Run OpenCL kernels Yes&lt;BR /&gt;Run native kernels Yes&lt;BR /&gt;Sub-group independent forward progress No&lt;BR /&gt;IL version SPIR-V_1.0&lt;BR /&gt;SPIR versions 1.2&lt;BR /&gt;printf() buffer size 1048576 (1024KiB)&lt;BR /&gt;Built-in kernels (n/a)&lt;BR /&gt;Device 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_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_khr_il_program cl_intel_unified_shared_memory_preview cl_intel_exec_by_local_thread cl_intel_vec_len_hint cl_intel_device_partition_by_names cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer&lt;/P&gt;
&lt;P&gt;NULL platform behavior&lt;BR /&gt;clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Intel(R) OpenCL&lt;BR /&gt;clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [INTEL]&lt;BR /&gt;clCreateContext(NULL, ...) [default] &amp;lt;checkNullCtx:2758: create context with device from default platform : error -2&amp;gt;&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) No devices available in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices available in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform&lt;BR /&gt;clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) No devices available in platform&lt;/P&gt;
&lt;P&gt;ICD loader properties&lt;BR /&gt;ICD loader Name OpenCL ICD Loader&lt;BR /&gt;ICD loader Vendor OCL Icd free software&lt;BR /&gt;ICD loader Version 2.2.11&lt;BR /&gt;ICD loader Profile OpenCL 2.1&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 04 Aug 2021 00:51:47 GMT</pubDate>
      <guid>https://community.intel.com/t5/GPU-Compute-Software/Can-t-install-OpenCL-driver-for-Intel-HD-Graphics-620-from-a/m-p/1303473#M200</guid>
      <dc:creator>ducbueno__</dc:creator>
      <dc:date>2021-08-04T00:51:47Z</dc:date>
    </item>
    <item>
      <title>Re: Can't install OpenCL driver for Intel HD Graphics 620 from a Gen7 (Kaby Lake) processor</title>
      <link>https://community.intel.com/t5/GPU-Compute-Software/Can-t-install-OpenCL-driver-for-Intel-HD-Graphics-620-from-a/m-p/1303703#M201</link>
      <description>&lt;P&gt;Hi, if you're using a Linux distribution that can use .deb install packages can you try downloading and installing the latest compute-runtime release?&lt;/P&gt;
&lt;P&gt;&lt;A href="https://github.com/intel/compute-runtime/releases/latest" target="_blank"&gt;https://github.com/intel/compute-runtime/releases/latest&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;These packages are much more up-to-date than the packages on the link above and will support your Kaby Lake GPU.&lt;/P&gt;
&lt;P&gt;With your current install, I suspect some of the required dependencies are missing (e.g. the GPU compiler) and this is why clinfo is only enumerating a CPU OpenCL device.&amp;nbsp; Please be sure to download all of the installation packages and install them as described in the "Installation procedure".&lt;/P&gt;
&lt;P&gt;If this still doesn't work, or if you cannot install from .deb packages, can you please include a bit more information about your system, such as which distro you are using?&amp;nbsp; Thanks!&lt;/P&gt;</description>
      <pubDate>Wed, 04 Aug 2021 18:19:26 GMT</pubDate>
      <guid>https://community.intel.com/t5/GPU-Compute-Software/Can-t-install-OpenCL-driver-for-Intel-HD-Graphics-620-from-a/m-p/1303703#M201</guid>
      <dc:creator>Ben_A_Intel</dc:creator>
      <dc:date>2021-08-04T18:19:26Z</dc:date>
    </item>
  </channel>
</rss>

