WARNING: CL_DEVICE_NOT_FOUND when calling clGetDeviceIDs for of devices.
Opened this issue · 17 comments
Get above error while trying to run the XMR-Stack , I have a AMD 590 GPU below is the "clinfo" detail
Number of platforms 1
Platform Name Clover
Platform Vendor Mesa
Platform Version OpenCL 1.1 Mesa 17.2.4
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd
Platform Extensions function suffix MESA
Platform Name Clover
Number of devices 1
Device Name Radeon RX 580 Series (AMD POLARIS10 / DRM 3.20.0 / 4.13.0-31-generic, LLVM 5.0.0)
Device Vendor AMD
Device Vendor ID 0x1002
Device Version OpenCL 1.1 Mesa 17.2.4
Driver Version 17.2.4
Device OpenCL C Version OpenCL C 1.1
Device Type GPU
Device Profile FULL_PROFILE
Max compute units 36
Max clock frequency 1340MHz
Max work item dimensions 3
Max work item sizes 256x256x256
Max work group size 256
Preferred work group size multiple 64
Preferred / native vector sizes
char 16 / 16
short 8 / 8
int 4 / 4
long 2 / 2
half 0 / 0 (n/a)
float 4 / 4
double 2 / 2 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals No
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
Correctly-rounded divide and sqrt operations No
Address bits 64, Little-Endian
Global memory size 8583028736 (7.994GiB)
Error Correction support No
Max memory allocation 6008120115 (5.595GiB)
Unified memory for Host and Device No
Minimum alignment for any data type 128 bytes
Alignment of base address 1024 bits (128 bytes)
Global Memory cache type None
Image support No
Local memory type Local
Local memory size 32768 (32KiB)
Max constant buffer size 2147483647 (2GiB)
Max number of constant args 16
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Profiling timer resolution 0ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Device Available Yes
Compiler Available Yes
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_byte_addressable_store cl_khr_fp64
NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Clover
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [MESA]
clCreateContext(NULL, ...) [default] Success [MESA]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1)
Platform Name Clover
Device Name Radeon RX 580 Series (AMD POLARIS10 / DRM 3.20.0 / 4.13.0-31-generic, LLVM 5.0.0)
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 Clover
Device Name Radeon RX 580 Series (AMD POLARIS10 / DRM 3.20.0 / 4.13.0-31-generic, LLVM 5.0.0)
ICD loader properties
ICD loader Name OpenCL ICD Loader
ICD loader Vendor OCL Icd free software
ICD loader Version 2.2.8
ICD loader Profile OpenCL 1.2
NOTE: your OpenCL library declares to support OpenCL 1.2,
but it seems to support up to OpenCL 2.1 too.
I am kind of stuck any help will be much appriceated
I installed the AMD drivers only from AMD site
I am using Ubuntu 16.04... I did the amdgpu-pro-install without --compute option . DO you think that could have caused the issue
@psychocrypt correct me if I'm wrong but isn't the lowest version of CL supported by the miner 1.2? It looks like @umeshgarg1 has OpenCL 1.1 which is below the minimum requirement
Did a reinstall of prodrives also did created the lync still getting the same error
[2018-02-06 22:21:48] : Start mining: MONERO
[2018-02-06 22:21:49] : Found AMD platform index id = 0, name = Advanced Micro Devices, Inc.
[2018-02-06 22:21:49] : WARNING: CL_DEVICE_NOT_FOUND when calling clGetDeviceIDs for of devices.
[2018-02-06 22:21:49] : WARNING: No AMD device found.
[2018-02-06 22:21:49] : WARNING: backend AMD disabled.
[2018-02-06 22:21:49] : CPU configuration stored in file 'cpu.txt'
[2018-02-06 22:21:49] : Starting 1x thread, affinity: 0.
[2018-02-06 22:21:49] : hwloc: memory pinned
I have used command "cmake .. -DCUDA_ENABLE=OFF -DOpenCL_INCLUDE_DIR=/opt/AMDAPPSDK-3.0/include -DOpenCL_LIBRARY=/opt/AMDAPPSDK-3.0/lib/x86_64/sdk/libOpenCL.so " to make the file
I am having the same issue too.
Am pretty new at this yet i managed to get xmr-stak using my puny AMD GPU (an R2) from the first time. But it only lasted until reboot - got an unusable system and took me quite some time to restore it.
System is Lubuntu 16.04.3 kernel 4.13.0-32, so HWE. Latest AMDAPP 3.0 and amd-gpu-pro 17.40. So far so good except amd-gpu-pro fails to build DKMS for the kernel due to some api mismatches, wrong number of args and the likes. So i got an unbootable system, quite hard to restore.
Since amd-gpu-pro was no longer an option for this system as i couldn't get DKMS to build i switched back to the native amdgpu hwe driver, which needs "radeon.cik.support=0 radeon.si_support=0 amdgpu.cik_support=1 amdgpu.si_support=1" passed to kernel at boot. The AMD-APP does link to /usr/lib/libOpenCL.so.1 so as @psychocrypt commented i linked the amd-gpu-pro one. To do that, instead of running its install script (--compute tried installing DKMS too), i extracted the amd-gpu-pro .deb files with dpkg-deb as i didn't want it to start building DKMS and fail partially. Next step was link the extracted tree under /opt/amdgpu-pro (probably a default hardwired if no install is done). Last step was LD_LIBRARY_PATH to there for xmr-stak. So now it's mining a wee bit faster :). This is a low end laptop, 2 CPU threads achieve like 14 H/s, one GPU thread like 21 H/S.
After pulling the xmr-stak master branch and recompiling I'm having same issue here.
[2018-04-07 09:59:24] : Mining coin: monero7
[2018-04-07 09:59:24] : Found AMD platform index id = 0, name = Advanced Micro Devices, Inc.
[2018-04-07 09:59:24] : WARNING: CL_DEVICE_NOT_FOUND when calling clGetDeviceIDs for of devices.
[2018-04-07 09:59:24] : WARNING: No AMD device found.
[2018-04-07 09:59:24] : WARNING: backend AMD disabled.
Note that can't be a driver issue because with the same setup I was mining monero fine one month ago.
Thank you for your fast answer! :) The link was already fixed in my compilation a month ago. So, the issue was not related with that. It was related with cmake configuration.
This doesn't work:
cmake -DCUDA_ENABLE=OFF -D OpenCL_LIBRARY=/opt/AMDAPPSDK-3.0/lib/x86_64/sdk/libOpenCL.so -D OpenCL_INCLUDE_DIR=/opt/AMDAPPSDK-3.0/include/ ..
But this works fine!
cmake -DCUDA_ENABLE=OFF -D OpenCL_LIBRARY=/opt/amdgpu-pro/lib/x86_64-linux-gnu/libOpenCL.so -D OpenCL_INCLUDE_DIR=/opt/AMDAPPSDK-3.0/include/ ..
Now it works perfectly. Hope this helps somebody else. Thank you!
I see this is a very common issue with all xmr-stak users, and its a shame because its a great mining software! Maybe you can provide an editable wiki where users with success in compiling all this stuff can provide working steps. I work with Ubuntu 16.04, with AMD 17.40 drivers and with latest (master branch) xmr-stak sources. I could repeat all steps from scratch and provide you a working recipe.