- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I am learning OpenCL and configuring the environment right now.
My system is Ubuntu 18.04. The HW is Intel® Core™ i7-8700K with GeForce GTX 1080 Ti.
I download intel_sdk_for_opencl_applications_2020.0.270.tar.gz from Intel® SDK for OpenCL™ Applications, extract it, run
sudo ./install.sh
and then follow the GUI instruction to complete the installation.
Then, I install NEO from intel/compute-runtime by
mkdir neo cd neo wget https://github.com/intel/compute-runtime/releases/download/19.51.15145/intel-gmmlib_19.3.4_amd64.deb wget https://github.com/intel/compute-runtime/releases/download/19.51.15145/intel-igc-core_1.0.3041_amd64.deb wget https://github.com/intel/compute-runtime/releases/download/19.51.15145/intel-igc-opencl_1.0.3041_amd64.deb wget https://github.com/intel/compute-runtime/releases/download/19.51.15145/intel-opencl_19.51.15145_amd64.deb wget https://github.com/intel/compute-runtime/releases/download/19.51.15145/intel-ocloc_19.51.15145_amd64.deb sudo dpkg -i *.deb
I check the platform by clinfo which gives me the results:
Number of platforms 2
Platform Name Intel(R) CPU Runtime for OpenCL(TM) Applications
Platform Vendor Intel(R) Corporation
Platform Version OpenCL 2.1 LINUX
Platform Profile FULL_PROFILE
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_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_intel_exec_by_local_thread cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer cl_intel_vec_len_hint
Platform Host timer resolution 1ns
Platform Extensions function suffix INTEL
Platform Name NVIDIA CUDA
Platform Vendor NVIDIA Corporation
Platform Version OpenCL 1.2 CUDA 10.2.95
Platform Profile FULL_PROFILE
Platform Extensions 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_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer cl_khr_int64_base_atomics cl_khr_int64_extended_atomics
Platform Extensions function suffix NV
Platform Name Intel(R) CPU Runtime for OpenCL(TM) Applications
Number of devices 1
Device Name Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz
Device Vendor Intel(R) Corporation
Device Vendor ID 0x8086
Device Version OpenCL 2.1 (Build 0)
Driver Version 18.1.0.0920
Device OpenCL C Version OpenCL C 2.0
Device Type CPU
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 12
Max clock frequency 3700MHz
Device Partition (core)
Max number of sub-devices 12
Supported partition types by counts, equally, by names (Intel)
Max work item dimensions 3
Max work item sizes 8192x8192x8192
Max work group size 8192
Preferred work group size multiple 128
Max sub-groups per work group 1
Preferred / native vector sizes
char 1 / 32
short 1 / 16
int 1 / 8
long 1 / 4
half 0 / 0 (n/a)
float 1 / 8
double 1 / 4 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
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 42109607936 (39.22GiB)
Error Correction support No
Max memory allocation 10527401984 (9.804GiB)
Unified memory for Host and Device Yes
Shared Virtual Memory (SVM) capabilities (core)
Coarse-grained buffer sharing Yes
Fine-grained buffer sharing Yes
Fine-grained system sharing Yes
Atomics Yes
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 0 bytes
Max size for global variable 65536 (64KiB)
Preferred total size of global vars 65536 (64KiB)
Global Memory cache type Read/Write
Global Memory cache size 262144 (256KiB)
Global Memory cache line size 64 bytes
Image support Yes
Max number of samplers per kernel 480
Max size for 1D images from buffer 657962624 pixels
Max 1D or 2D image array size 2048 images
Base address alignment for 2D image buffers 64 bytes
Pitch alignment for 2D image buffers 64 pixels
Max 2D image size 16384x16384 pixels
Max 3D image size 2048x2048x2048 pixels
Max number of read image args 480
Max number of write image args 480
Max number of read/write image args 480
Max number of pipe args 16
Max active pipe reservations 21845
Max pipe packet size 1024
Local memory type Global
Local memory size 32768 (32KiB)
Max number of constant args 480
Max constant buffer size 131072 (128KiB)
Max size of kernel argument 3840 (3.75KiB)
Queue properties (on host)
Out-of-order execution Yes
Profiling Yes
Local thread execution (Intel) Yes
Queue properties (on device)
Out-of-order execution Yes
Profiling Yes
Preferred size 4294967295 (4GiB)
Max size 4294967295 (4GiB)
Max queues on device 4294967295
Max events on device 4294967295
Prefer user sync for interop No
Profiling timer resolution 1ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels Yes
Sub-group independent forward progress No
IL version SPIR-V_1.0
SPIR versions 1.2
printf() buffer size 1048576 (1024KiB)
Built-in kernels
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_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_intel_exec_by_local_thread cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer cl_intel_vec_len_hint
Platform Name NVIDIA CUDA
Number of devices 1
Device Name GeForce GTX 1080 Ti
Device Vendor NVIDIA Corporation
Device Vendor ID 0x10de
Device Version OpenCL 1.2 CUDA
Driver Version 440.33.01
Device OpenCL C Version OpenCL C 1.2
Device Type GPU
Device Topology (NV) PCI-E, 01:00.0
Device Profile FULL_PROFILE
Device Available Yes
Compiler Available Yes
Linker Available Yes
Max compute units 28
Max clock frequency 1645MHz
Compute Capability (NV) 6.1
Device Partition (core)
Max number of sub-devices 1
Supported partition types None
Max work item dimensions 3
Max work item sizes 1024x1024x64
Max work group size 1024
Preferred work group size multiple 32
Warp size (NV) 32
Preferred / native vector sizes
char 1 / 1
short 1 / 1
int 1 / 1
long 1 / 1
half 0 / 0 (n/a)
float 1 / 1
double 1 / 1 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
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 11718230016 (10.91GiB)
Error Correction support No
Max memory allocation 2929557504 (2.728GiB)
Unified memory for Host and Device No
Integrated memory (NV) No
Minimum alignment for any data type 128 bytes
Alignment of base address 4096 bits (512 bytes)
Global Memory cache type Read/Write
Global Memory cache size 1376256 (1.312MiB)
Global Memory cache line size 128 bytes
Image support Yes
Max number of samplers per kernel 32
Max size for 1D images from buffer 268435456 pixels
Max 1D or 2D image array size 2048 images
Max 2D image size 16384x32768 pixels
Max 3D image size 16384x16384x16384 pixels
Max number of read image args 256
Max number of write image args 16
Local memory type Local
Local memory size 49152 (48KiB)
Registers per block (NV) 65536
Max number of constant args 9
Max constant buffer size 65536 (64KiB)
Max size of kernel argument 4352 (4.25KiB)
Queue properties
Out-of-order execution Yes
Profiling Yes
Prefer user sync for interop No
Profiling timer resolution 1000ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Kernel execution timeout (NV) Yes
Concurrent copy and kernel execution (NV) Yes
Number of async copy engines 2
printf() buffer size 1048576 (1024KiB)
Built-in kernels
Device Extensions 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_fp64 cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll cl_nv_copy_opts cl_nv_create_buffer cl_khr_int64_base_atomics cl_khr_int64_extended_atomics
NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) No platform
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) No platform
clCreateContext(NULL, ...) [default] No platform
clCreateContext(NULL, ...) [other] Success [INTEL]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_DEFAULT) Success (1)
Platform Name Intel(R) CPU Runtime for OpenCL(TM) Applications
Device Name Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) Success (1)
Platform Name Intel(R) CPU Runtime for OpenCL(TM) Applications
Device Name Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No devices found in platform
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) CPU Runtime for OpenCL(TM) Applications
Device Name Intel(R) Core(TM) i7-8700K CPU @ 3.70GHz
NOTE: your OpenCL library only supports OpenCL 2.0,
but some installed platforms support OpenCL 2.1.
Programs using 2.1 features may crash
or behave unexepectedly
By the way, I already install the CUDA through other channels. The result from clinfo doesn't detect Intel Graphics.
Also, I try sample code from Get Started with Intel® SDK for OpenCL™ Applications 2019 on Linux* OS with Training Sample.
The cpu version works well.
./CPUOpenCLProjectforLinux gives me
Selected device type is CL_DEVICE_TYPE_CPU Number of available platforms: 2 Selected platform: Intel(R) CPU Runtime for OpenCL(TM) Applications NDRange performance counter time 0.488000 ms.
while ./GPUOpenCLProjectforLinux gives me
Selected device type is CL_DEVICE_TYPE_GPU Number of available platforms: 2 Selected platform: Intel(R) CPU Runtime for OpenCL(TM) Applications clGetDeviceIDs() returned CL_DEVICE_NOT_FOUND for platform Intel(R) CPU Runtime for OpenCL(TM) Applications. Error: Failed to find OpenCL platform.
Did I miss any procedure? Thanks
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi AlterEtch,
I don't see any obvious missing steps, but here are a few additional things to try:
1) I wrote an article describing how OpenCL works on Linux that includes some troubleshooting tips that may be helpful:
2) Since you also have a discrete GPU card, you may want to verify that your system keeps the integrated GPU enabled when a discrete GPU is installed. If you run "lspci" do you see the integrated GPU?
3) If everything else looks OK, can you verify that your user is in the "video" group? See:
https://github.com/intel/compute-runtime#installation-options
Thanks!
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi AlterEtch,
I don't see any obvious missing steps, but here are a few additional things to try:
1) I wrote an article describing how OpenCL works on Linux that includes some troubleshooting tips that may be helpful:
2) Since you also have a discrete GPU card, you may want to verify that your system keeps the integrated GPU enabled when a discrete GPU is installed. If you run "lspci" do you see the integrated GPU?
3) If everything else looks OK, can you verify that your user is in the "video" group? See:
https://github.com/intel/compute-runtime#installation-options
Thanks!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ben,
first I have to thank you for your article and code samples!
I am having a similar issue installing Intel-compute-runtime (NEO).
I installed it as the release page suggests:
wget https://github.com/intel/compute-runtime/releases/download/21.27.20266/intel-gmmlib_21.2.1_amd64.deb
wget https://github.com/intel/intel-graphics-compiler/releases/download/igc-1.0.7862/intel-igc-core_1.0.7862_amd64.deb
wget https://github.com/intel/intel-graphics-compiler/releases/download/igc-1.0.7862/intel-igc-opencl_1.0.7862_amd64.deb
wget https://github.com/intel/compute-runtime/releases/download/21.27.20266/intel-opencl_21.27.20266_amd64.deb
wget https://github.com/intel/compute-runtime/releases/download/21.27.20266/intel-ocloc_21.27.20266_amd64.deb
wget https://github.com/intel/compute-runtime/releases/download/21.27.20266/intel-level-zero-gpu_1.1.20266_amd64.deb
sudo dpkg -i *.deb
"/etc/OpenCL/vendors/intel.icd" does say "/usr/local/lib/intel-opencl/libigdrcl.so".
Under /usr/local/lib:
drwxr-xr-x - root root 1 Jul 17:13 igc/
drwxr-xr-x - clli clli 1 Jul 17:13 intel-opencl/
lrwxrwxrwx 13 root root 6 Jun 3:32 libiga64.so -> libiga64.so.1
lrwxrwxrwx 20 root root 6 Jun 3:32 libiga64.so.1 -> libiga64.so.1.0.7683
.rw-r--r-- 2.0M root root 6 Jun 3:32 libiga64.so.1.0.7683
lrwxrwxrwx 11 root root 6 Jun 3:32 libigc.so -> libigc.so.1
lrwxrwxrwx 18 root root 6 Jun 3:32 libigc.so.1 -> libigc.so.1.0.7683
.rw-r--r-- 63M root root 6 Jun 3:32 libigc.so.1.0.7683
lrwxrwxrwx 14 root root 6 Jun 3:32 libigdfcl.so -> libigdfcl.so.1
lrwxrwxrwx 21 root root 6 Jun 3:32 libigdfcl.so.1 -> libigdfcl.so.1.0.7683
.rw-r--r-- 1.6M root root 6 Jun 3:32 libigdfcl.so.1.0.7683
lrwxrwxrwx 22 clli clli 6 May 19:01 libigdgmm.so.11 -> libigdgmm.so.11.3.1111
.rw-r--r-- 537k clli clli 14 Jun 16:15 libigdgmm.so.11.3.1111
.rwxr-xr-x 692k clli clli 14 Jun 16:16 libocloc.so*
lrwxrwxrwx 21 root root 6 Jun 3:32 libopencl-clang.so -> libopencl-clang.so.10
.rw-r--r-- 113M root root 8 Dec 2020 libopencl-clang.so.10
.rw-r--r-- 27M root root 6 Jun 3:31 libSPIRVDLL.so
lrwxrwxrwx 34 root root 25 Feb 2020 libz3.so -> /usr/lib/x86_64-linux-gnu/libz3.so
lrwxrwxrwx 28 clli clli 11 Jun 4:42 libze_intel_gpu.so.1 -> libze_intel_gpu.so.1.1.20098
.rwxr-xr-x 5.4M clli clli 14 Jun 16:16 libze_intel_gpu.so.1.1.20098*
Looks like everything is in place.
However, clinfo only shows Nvidia OpenCL.
What lshw says:
❯ lshw -C video
WARNING: you should run this program as super-user.
*-display
description: VGA compatible controller
product: GP102 [GeForce GTX 1080 Ti]
vendor: NVIDIA Corporation
physical id: 0
bus info: pci@0000:01:00.0
version: a1
width: 64 bits
clock: 33MHz
capabilities: vga_controller bus_master cap_list rom
configuration: driver=nvidia latency=0
resources: irq:16 memory:a3000000-a3ffffff memory:90000000-9fffffff memory:a0000000-a1ffffff ioport:3000(size=128) memory:c0000-dffff
*-display
description: Display controller
product: UHD Graphics 630 (Desktop 9 Series)
vendor: Intel Corporation
physical id: 2
bus info: pci@0000:00:02.0
version: 02
width: 64 bits
clock: 33MHz
capabilities: bus_master cap_list
configuration: driver=i915 latency=0
resources: irq:130 memory:a2000000-a2ffffff memory:80000000-8fffffff ioport:4000(size=64)
WARNING: output may be incomplete or inaccurate, you should run this program as super-user.
lspci:
00:00.0 Host bridge: Intel Corporation 8th Gen Core 8-core Desktop Processor Host Bridge/DRAM Registers [Coffee Lake S] (rev 0d)
00:01.0 PCI bridge: Intel Corporation Xeon E3-1200 v5/E3-1500 v5/6th Gen Core Processor PCIe Controller (x16) (rev 0d)
00:02.0 Display controller: Intel Corporation UHD Graphics 630 (Desktop 9 Series) (rev 02)
00:14.0 USB controller: Intel Corporation Cannon Lake PCH USB 3.1 xHCI Host Controller (rev 10)
00:14.2 RAM memory: Intel Corporation Cannon Lake PCH Shared SRAM (rev 10)
00:16.0 Communication controller: Intel Corporation Cannon Lake PCH HECI Controller (rev 10)
00:17.0 RAID bus controller: Intel Corporation SATA Controller [RAID mode] (rev 10)
00:1b.0 PCI bridge: Intel Corporation Cannon Lake PCH PCI Express Root Port #17 (rev f0)
00:1c.0 PCI bridge: Intel Corporation Cannon Lake PCH PCI Express Root Port #1 (rev f0)
00:1d.0 PCI bridge: Intel Corporation Cannon Lake PCH PCI Express Root Port #9 (rev f0)
00:1f.0 ISA bridge: Intel Corporation Z390 Chipset LPC/eSPI Controller (rev 10)
00:1f.3 Audio device: Intel Corporation Cannon Lake PCH cAVS (rev 10)
00:1f.4 SMBus: Intel Corporation Cannon Lake PCH SMBus Controller (rev 10)
00:1f.5 Serial bus controller [0c80]: Intel Corporation Cannon Lake PCH SPI Controller (rev 10)
00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection (7) I219-V (rev 10)
01:00.0 VGA compatible controller: NVIDIA Corporation GP102 [GeForce GTX 1080 Ti] (rev a1)
01:00.1 Audio device: NVIDIA Corporation GP102 HDMI Audio Controller (rev a1)
Part of strace of clinfo that I deem relevant (thank you for your article again!):
Looks like the ICD loader did find "libigdrcl.so" and "libigdgmm.so.11" through "intel.icd".
openat(AT_FDCWD, "/etc/OpenCL/vendors/intel.icd", O_RDONLY) = 4
fstat(4, {st_mode=S_IFREG|0644, st_size=41, ...}) = 0
fstat(4, {st_mode=S_IFREG|0644, st_size=41, ...}) = 0
lseek(4, 0, SEEK_SET) = 0
read(4, "/usr/local/lib/intel-opencl/libi"..., 41) = 41
lseek(4, 41, SEEK_SET) = 41
openat(AT_FDCWD, "/usr/local/lib/intel-opencl/libigdrcl.so", O_RDONLY|O_CLOEXEC) = 10
read(10, "\177ELF\2\1\1\3\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0p\346\2\0\0\0\0\0"..., 832) = 832
fstat(10, {st_mode=S_IFREG|0755, st_size=6204016, ...}) = 0
mmap(NULL, 8798672, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 10, 0) = 0x7f83b7b8d000
mprotect(0x7f83b8049000, 2093056, PROT_NONE) = 0
mmap(0x7f83b8248000, 1245184, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 10, 0x4bb000) = 0x7f83b8248000
mmap(0x7f83b8378000, 496080, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7f83b8378000
close(10) = 0
openat(AT_FDCWD, "/home/pychang/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "tls/haswell/x86_64/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "tls/haswell/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "tls/x86_64/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "tls/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "haswell/x86_64/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "haswell/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "x86_64/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "/home/pychang/.llvm-10.0.0-release/lib/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
openat(AT_FDCWD, "/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 10
fstat(10, {st_mode=S_IFREG|0644, st_size=77151, ...}) = 0
mmap(NULL, 77151, PROT_READ, MAP_PRIVATE, 10, 0) = 0x7f83c14b0000
close(10) = 0
openat(AT_FDCWD, "/usr/local/lib/libigdgmm.so.11", O_RDONLY|O_CLOEXEC) = 10
read(10, "\177ELF\2\1\1\0\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0\360p\0\0\0\0\0\0"..., 832) = 832
fstat(10, {st_mode=S_IFREG|0644, st_size=537920, ...}) = 0
mmap(NULL, 2633088, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 10, 0) = 0x7f83b790a000
mprotect(0x7f83b7988000, 2097152, PROT_NONE) = 0
mmap(0x7f83b7b88000, 20480, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 10, 0x7e000) = 0x7f83b7b88000
close(10)
The ICD loader is "/usr/local/cuda-10.2/targets/x86_64-linux/lib/libOpenCL.so.1".
I tried using LD_LIBRARY_PATH to force it to use Khronos ICD loader, but gained no luck.
Could it be iGPU not being VGA compatible controller?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Kevin,
first I have to thank you for your article and code samples!
You're very welcome, I'm glad they are useful!
I don't see anything obviously wrong in the information you have provided.
One easy thing to check is: Is your user in the "video" group or otherwise have access to /dev/dri/renderD*? See:
https://github.com/intel/compute-runtime#installation-options
If so, some next things that would be helpful to know are:
- What distro are you using and what is your kernel version?
- Do you have your display connected to the integrated GPU or the discrete GPU (or both)?
- Is this an "optimus" system?
The ICD loader is "/usr/local/cuda-10.2/targets/x86_64-linux/lib/libOpenCL.so.1".
I tried using LD_LIBRARY_PATH to force it to use Khronos ICD loader, but gained no luck.
This is probably fine, at least for now. There could be an issue using the latest OpenCL features if this is an older ICD loader, but it shouldn't cause an issue enumerating OpenCL platforms and devices.
Could it be iGPU not being VGA compatible controller?
I don't think so, but just in case, it'd be worth taking a look through your BIOS to see if there is anything related to keeping the iGPU enabled when a dGPU is installed.
Good luck!
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Thanks for your execellent troubleshooting tips, which I have studied in depth. I have an Intel(R) Core(TM) i5-3470 CPU @ 3.20GHz running Ubuntu 22.04LTS. I have installed OpenCL from intel-gpu-tools. The proper libraries ae present and intel.ics is in /etc/OpenCL/vendors. All of you non-aplication specific troubleshooting steps check out. However, clinfo returns "Number of platforms 0". Might you hae any further ideas?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi Ben,
It seems my intel graphics is not enabled. I run `lspci | grep VGA`, getting the result of only NVIDIA.
Then, I run `sudo prime-select intel` and reboot. But the result of `lspci | grep VGA` doesn't change at all.
I also try the same procedure on a laptop without dedicated GPU, and it works well.
Now I need to find a way to enable the integrated GPU.
==========================
UPDATE:
Somehow all my HDMI cables are plugged into the dedicated GPU which make integrated GPU undetectable.
Now I can find three platform: intel cpu, intel gpu, and cuda.
Thanks.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
AlterEtch,
Glad the platforms are detected.
I checked with the Intel Graphics Compute Runtime Driver developers for some comment about the hardware configuration...
They suggested the possibility that the system vendor BIOS or EFI control could disable Intel Graphics Technology hardware if the display cable was not connected to it. This would cause the iGFX device to not be resolved by the Linux* OS at 'PCI bus as 00:02.0 VGA compatible controller', and not allow the corresponding i915 video driver to load. The OpenCL implementation for iGFX relies on this OS+Driver visibility.
Windows* OS systems are known to disable device resolution already as described in the 2020 developer sdk release notes page 9.
-MichaelC
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page