Not generating dummy.icd in etc/OpenCL/vendors

[MAMorgado]

Hi! Not sure if i'm doing something wrong but after the installation I don't have any icd loader in the vendors folder.

Here's what I did.

Clone git ./bootstrap ./configure --enable-update-database sudo make sudo make check (result all ok) sudo make install sudo make installcheck (result all ok)

Thank you!

MaM

[Kerilk]

Hello,

First a disclaimer: this is nor regularly tested as the structure is now public, so this mechanism is not needed at the present time. Nonetheless I tested and it seemed to work fine.

The database update has it's specific target in the Makefile, so here is how I would go about it:

./bootstrap
./configure --enable-update-database
# this should prompt for your password at some point
# unless you used sudo recently
make update-database
make check
sudo make install

The script will create the /etc/OpenCL/vendors/dummycl.icd using a sudo command and you will be prompted to input your password when needed. The dummy icd is never installed but resides in .libs/libdummycl.so. Don't forget to destroy the /etc/OpenCL/vendors/dummycl.icd as it will get picked up and used by OpenCL loaders. Hope this helps, don't hesitate if you have more questions.

[MAMorgado]

Hi Kerilk. Thanks for your help. the make update-database fixed it and it generated the dummyicd in the etc...

But now I was trying a specific release of yours to be compatible with open cl 3.0 (v2.2.13) and when I try to make install I get the following errors:

make install-recursive make[1]: Entering directory '/home/husarion/ocl-icd/ocl-icd-2.2.13' Making install in . make[2]: Entering directory '/home/husarion/ocl-icd/ocl-icd-2.2.13' gcc -g -O2 -o run_dummy_icd run_dummy_icd-run_dummy_icd.o run_dummy_icd-run_dummy_icd_gen.o run_dummy_icd-run_dummy_icd_weak_gen.o -lOpenCL -ldl /usr/bin/ld: cannot find -lOpenCL collect2: error: ld returned 1 exit status make[2]: *** [Makefile:754: run_dummy_icd] Error 1 make[2]: Leaving directory '/home/husarion/ocl-icd/ocl-icd-2.2.13' make[1]: *** [Makefile:1015: install-recursive] Error 1 make[1]: Leaving directory '/home/husarion/ocl-icd/ocl-icd-2.2.13' make: *** [Makefile:1323: install] Error 2

I suspect that I may have not completely deleted this version... or are you seing something else?

Thanks for all the help :)

[Kerilk]

If you want to do this, you just need to import the ocl_interface.yaml from the latest build.

The update database option goal was to reverse engineer an existing libOpenCL.so library (usually provided by vendors) when Khronos had not released the loader sources. So in order to achieve your goal you need a working 3.0 OpenCL loader. But all it will do is update the above yaml file.

Should you want to still do it, the simplest way is to build and install the latest ocl-icd, then build v2.2.13 updating the database.

If you don't mind asking, are you experiencing any issue with the latest version?

[MAMorgado]

Hi! No I had no problem with the latest release except that I was not getting the 3.0 functionalities which I need to run another program.

Should you want to still do it, the simplest way is to build and install the latest ocl-icd, then build v2.2.13 updating the database.

I still have the latest version installed so this would be perfect. How can I do it? I'm sorry I am quite the beginner :/

[Kerilk]

This cannot find -lOpenCL you were given means that the installation of the loader was not complete. Most probably the OpenCL library installed in /usr/local/lib/ocl-icd (or some equivalent). Depending on your system there may be simpler solutions:

Note that I checked an 2.2.13 already offers OpenCL 3.0 so no update should be required if it was the version already installed on your system.

I you want the latest ocl-icd loader installed on your distribution feel free to use this ppa I setup for this exact purpose.

https://launchpad.net/~ocl-icd/+archive/ubuntu/ppa

[MAMorgado]

Hey :)

I think it is finally OK now!

Here is the result from my clinfo

$ clinfo
Number of platforms                               2
  Platform Name                                   Intel(R) OpenCL HD Graphics
  Platform Vendor                                 Intel(R) Corporation
  Platform Version                                OpenCL 3.0 
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_byte_addressable_store cl_khr_fp16 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_intel_command_queue_families cl_intel_subgroups cl_intel_required_subgroup_size cl_intel_subgroups_short cl_khr_spir cl_intel_accelerator cl_intel_driver_diagnostics cl_khr_priority_hints cl_khr_throttle_hints cl_khr_create_command_queue cl_intel_subgroups_char cl_intel_subgroups_long cl_khr_il_program cl_intel_mem_force_host_memory cl_khr_subgroup_extended_types cl_khr_subgroup_non_uniform_vote cl_khr_subgroup_ballot cl_khr_subgroup_non_uniform_arithmetic cl_khr_subgroup_shuffle cl_khr_subgroup_shuffle_relative cl_khr_subgroup_clustered_reduce cl_intel_device_attribute_query cl_khr_fp64 cl_khr_subgroups cl_intel_spirv_device_side_avc_motion_estimation cl_intel_spirv_media_block_io cl_intel_spirv_subgroups cl_khr_spirv_no_integer_wrap_decoration cl_intel_unified_shared_memory_preview cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_intel_planar_yuv cl_intel_packed_yuv cl_intel_motion_estimation cl_intel_device_side_avc_motion_estimation cl_intel_advanced_motion_estimation cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_3d_image_writes cl_intel_media_block_io cl_intel_va_api_media_sharing cl_intel_sharing_format_query cl_khr_pci_bus_info 
  Platform Host timer resolution                  1ns
  Platform Extensions function suffix             INTEL

  Platform Name                                   DummyCL
  Platform Vendor                                 ocl-icd ICD test
  Platform Version                                OpenCL 1.2
  Platform Profile                                FULL_PROFILE
  Platform Extensions                             cl_khr_icd
  Platform Extensions function suffix             LIG

  Platform Name                                   Intel(R) OpenCL HD Graphics
Number of devices                                 1
  Device Name                                     Intel(R) UHD Graphics [0x9bca]
  Device Vendor                                   Intel(R) Corporation
  Device Vendor ID                                0x8086
  Device Version                                  OpenCL 3.0 NEO 
  Driver Version                                  21.29.20389
  Device OpenCL C Version                         OpenCL C 3.0 
  Device Type                                     GPU
  Device Profile                                  FULL_PROFILE
  Device Available                                Yes
  Compiler Available                              Yes
  Linker Available                                Yes
  Max compute units                               24
  Max clock frequency                             1150MHz
  Device Partition                                (core)
    Max number of sub-devices                     0
    Supported partition types                     None
    Supported affinity domains                    (n/a)
  Max work item dimensions                        3
  Max work item sizes                             256x256x256
  Max work group size                             256
  Preferred work group size multiple              32
  Max sub-groups per work group                   32
  Sub-group sizes (Intel)                         8, 16, 32
  Preferred / native vector sizes                 
    char                                                16 / 16      
    short                                                8 / 8       
    int                                                  4 / 4       
    long                                                 1 / 1       
    half                                                 8 / 8        (cl_khr_fp16)
    float                                                1 / 1       
    double                                               1 / 1        (cl_khr_fp64)
  Half-precision Floating-point support           (cl_khr_fp16)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
  Single-precision Floating-point support         (core)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
    Correctly-rounded divide and sqrt operations  Yes
  Double-precision Floating-point support         (cl_khr_fp64)
    Denormals                                     Yes
    Infinity and NANs                             Yes
    Round to nearest                              Yes
    Round to zero                                 Yes
    Round to infinity                             Yes
    IEEE754-2008 fused multiply-add               Yes
    Support is emulated in software               No
  Address bits                                    64, Little-Endian
  Global memory size                              13188349952 (12.28GiB)
  Error Correction support                        No
  Max memory allocation                           4294959104 (4GiB)
  Unified memory for Host and Device              Yes
  Shared Virtual Memory (SVM) capabilities        (core)
    Coarse-grained buffer sharing                 Yes
    Fine-grained buffer sharing                   No
    Fine-grained system sharing                   No
    Atomics                                       No
  Minimum alignment for any data type             128 bytes
  Alignment of base address                       1024 bits (128 bytes)
  Preferred alignment for atomics                 
    SVM                                           64 bytes
    Global                                        64 bytes
    Local                                         64 bytes
  Max size for global variable                    65536 (64KiB)
  Preferred total size of global vars             4294959104 (4GiB)
  Global Memory cache type                        Read/Write
  Global Memory cache size                        524288 (512KiB)
  Global Memory cache line size                   64 bytes
  Image support                                   Yes
    Max number of samplers per kernel             16
    Max size for 1D images from buffer            268434944 pixels
    Max 1D or 2D image array size                 2048 images
    Base address alignment for 2D image buffers   4 bytes
    Pitch alignment for 2D image buffers          4 pixels
    Max 2D image size                             16384x16384 pixels
    Max planar YUV image size                     16384x16352 pixels
    Max 3D image size                             16384x16384x2048 pixels
    Max number of read image args                 128
    Max number of write image args                128
    Max number of read/write image args           128
  Max number of pipe args                         16
  Max active pipe reservations                    1
  Max pipe packet size                            1024
  Local memory type                               Local
  Local memory size                               65536 (64KiB)
  Max number of constant args                     8
  Max constant buffer size                        4294959104 (4GiB)
  Max size of kernel argument                     2048 (2KiB)
  Queue properties (on host)                      
    Out-of-order execution                        Yes
    Profiling                                     Yes
  Queue properties (on device)                    
    Out-of-order execution                        Yes
    Profiling                                     Yes
    Preferred size                                131072 (128KiB)
    Max size                                      67108864 (64MiB)
  Max queues on device                            1
  Max events on device                            1024
  Prefer user sync for interop                    Yes
  Profiling timer resolution                      83ns
  Execution capabilities                          
    Run OpenCL kernels                            Yes
    Run native kernels                            No
    Sub-group independent forward progress        Yes
    IL version                                    SPIR-V_1.2 
    SPIR versions                                 1.2 
  printf() buffer size                            4194304 (4MiB)
  Built-in kernels                                block_motion_estimate_intel;block_advanced_motion_estimate_check_intel;block_advanced_motion_estimate_bidirectional_check_intel;
  Motion Estimation accelerator version (Intel)   2
    Device-side AVC Motion Estimation version     1
      Supports texture sampler use                Yes
      Supports preemption                         No
  Device Extensions                               cl_khr_byte_addressable_store cl_khr_fp16 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_intel_command_queue_families cl_intel_subgroups cl_intel_required_subgroup_size cl_intel_subgroups_short cl_khr_spir cl_intel_accelerator cl_intel_driver_diagnostics cl_khr_priority_hints cl_khr_throttle_hints cl_khr_create_command_queue cl_intel_subgroups_char cl_intel_subgroups_long cl_khr_il_program cl_intel_mem_force_host_memory cl_khr_subgroup_extended_types cl_khr_subgroup_non_uniform_vote cl_khr_subgroup_ballot cl_khr_subgroup_non_uniform_arithmetic cl_khr_subgroup_shuffle cl_khr_subgroup_shuffle_relative cl_khr_subgroup_clustered_reduce cl_intel_device_attribute_query cl_khr_fp64 cl_khr_subgroups cl_intel_spirv_device_side_avc_motion_estimation cl_intel_spirv_media_block_io cl_intel_spirv_subgroups cl_khr_spirv_no_integer_wrap_decoration cl_intel_unified_shared_memory_preview cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_intel_planar_yuv cl_intel_packed_yuv cl_intel_motion_estimation cl_intel_device_side_avc_motion_estimation cl_intel_advanced_motion_estimation cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_image2d_from_buffer cl_khr_depth_images cl_khr_3d_image_writes cl_intel_media_block_io cl_intel_va_api_media_sharing cl_intel_sharing_format_query cl_khr_pci_bus_info 

  Platform Name                                   DummyCL
Number of devices                                 0

NULL platform behavior
  clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...)  Intel(R) OpenCL HD Graphics
  clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...)   Success [INTEL]
  clCreateContext(NULL, ...) [default]            Success [INTEL]
  clCreateContext(NULL, ...) [other]              �A�;�U
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT)  Success (1)
    Platform Name                                 Intel(R) OpenCL HD Graphics
    Device Name                                   Intel(R) UHD Graphics [0x9bca]
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU)  Success (1)
    Platform Name                                 Intel(R) OpenCL HD Graphics
    Device Name                                   Intel(R) UHD Graphics [0x9bca]
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM)  No devices found in platform
  clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL)  Success (1)
    Platform Name                                 Intel(R) OpenCL HD Graphics
    Device Name                                   Intel(R) UHD Graphics [0x9bca]

ICD loader properties
  ICD loader Name                                 OpenCL ICD Loader
  ICD loader Vendor                               OCL Icd free software
  ICD loader Version                              2.3.0
  ICD loader Profile                              OpenCL 3.0
    NOTE:   your OpenCL library declares to support OpenCL 3.0,
        but it seems to support up to OpenCL 2.2 only.
    NOTE:   your OpenCL library only supports OpenCL 2.2,
        but some installed platforms support OpenCL 3.0.
        Programs using 3.0 features may crash
        or behave unexpectedly

Thank you!