OpenCL not supported: clGetPlatformIDs() returned -1001

[dbubbis13]

I've been trying to get my GTX 1650 to fold, but it isn't working no matter what I do. I'm pretty sure I have the correct drivers installed, but am still getting errors in the folding at home client. The PPD isn't increasing whenever I enable the GPU as well. The GPU is being passed through Proxmox VE 8.3.0.
Here is the log:

Code: Select all

22:36:18:I1: Version: 8.3.18
22:36:18:I1: Author: Joseph Coffland <[email protected]>
22:36:18:I1: Org: foldingathome.org
22:36:18:I1: Copyright: 2023-2024, foldingathome.org
22:36:18:I1: Homepage: https://foldingathome.org/
22:36:18:I1: License: GPL-3.0-or-later
22:36:18:I1: URL: https://v8-3.foldingathome.org/
22:36:18:I1: Date: Jul 12 2024
22:36:18:I1: Time: 13:26:31
22:36:18:I1: Revision: 99ae953ee7b1c0b3070161cfcf9150184f76bd96
22:36:18:I1: Branch: master
22:36:18:I1: Compiler: GNU 8.3.0
22:36:18:I1: Options: -Wsuggest-override -faligned-new -std=c++17 -fsigned-char
22:36:18:I1: -ffunction-sections -fdata-sections -O3 -funroll-loops -fno-pie
22:36:18:I1: Platform: linux 4.19.0-26-cloud-amd64
22:36:18:I1: Bits: 64
22:36:18:I1: Mode: Release
22:36:18:I1: Args: --config=/etc/fah-client/config.xml
22:36:18:I1: --log=/var/log/fah-client/log.txt
22:36:18:I1: --log-rotate-dir=/var/log/fah-client/
22:36:18:I1: Config: /etc/fah-client/config.xml
22:36:18:I1:****************************** CBang ******************************
22:36:18:I1: Version: 1.7.2
22:36:18:I1: Author: Joseph Coffland <[email protected]>
22:36:18:I1: Org: Cauldron Development
22:36:18:I1: Copyright: Cauldron Development, 2003-2024
22:36:18:I1: Homepage: https://cauldrondevelopment.com/
22:36:18:I1: License: LGPL-2.1-or-later
22:36:18:I1: Date: Jun 24 2024
22:36:18:I1: Time: 13:29:44
22:36:18:I1: Revision: 1b05ea96f0ed3043c32b78a66dbf50a9b2002289
22:36:18:I1: Branch: master
22:36:18:I1: Compiler: GNU 8.3.0
22:36:18:I1: Options: -Wsuggest-override -faligned-new -std=c++17 -fsigned-char
22:36:18:I1: -ffunction-sections -fdata-sections -O3 -funroll-loops -fno-pie
22:36:18:I1: -fPIC
22:36:18:I1: Platform: linux 4.19.0-26-cloud-amd64
22:36:18:I1: Bits: 64
22:36:18:I1: Mode: Release
22:36:18:I1:***************************** System ******************************
22:36:18:I1: CPU: QEMU Virtual CPU version 2.5+
22:36:18:I1: CPU ID: GenuineIntel Family 15 Model 107 Stepping 1
22:36:18:I1: CPUs: 4
22:36:18:I1: Memory: 1.88GiB
22:36:18:I1:Free Memory: 1.62GiB
22:36:18:I1: OS Version: 6.1
22:36:18:I1:Has Battery: false
22:36:18:I1: On Battery: false
22:36:18:I1: Hostname: boincnodeone
22:36:18:I1: UTC Offset: -6
22:36:18:I1: PID: 637
22:36:18:I1: CWD: /var/lib/fah-client
22:36:18:I1: Exec: /usr/bin/fah-client
22:36:18:I1:*******************************************************************
22:36:18:I2:<config/>
22:36:18:I1:Opening Database
22:36:18:I1:F@H ID = z1ySWIs6e9TX8755BROYo113DTPSOleRWKpH_9U5xuk
22:36:18:I3:Loading default group
22:36:18:I3:Loading default resource group
22:36:18:I1:Listening for HTTP on 127.0.0.1:7396
22:36:18:I3:WU24:Loading work unit 24 with ID c7G58aJg6P0BU-EuSq0Tx06loqSp-0XacoH6KWITzgw
22:36:18:I3:Loaded 1 wus.
22:36:18:W :OpenCL not supported: clGetPlatformIDs() returned -1001
22:36:18:W :CUDA not supported: Failed to open dynamic library 'libcuda.so': libcuda.so: cannot open shared object file: No such file or directory
22:36:18:I3:gpus = {
22:36:18:I3: "gpu:01:00:00": {"vendor": 4318, "device": 8584, "type": "nvidia", "supported": false, "description": "TU116 [GeForce GTX 1650] 3091"}
22:36:18:I3:}
22:36:34:W :CON1:DNS lookup failed for api.foldingathome.org
22:36:34:E :OUT1:Failed response: CONNECT
22:36:50:I1:OUT2:> GET https://api.foldingathome.org/machine/z1ySWIs6e9TX8755BROYo113DTPSOleRWKpH_9U5xuk HTTP/1.1
22:36:50:I1:OUT2:< HTTP/1.1 200 HTTP_OK
22:36:50:I1:OUT3:> GET wss://node1.foldingathome.org/ws/client HTTP/1.1
22:36:51:I1:OUT3:< HTTP/1.1 101 HTTP_SWITCHING_PROTOCOLS
22:36:51:I1:Logging into node account

Would this have anything to due with how I passed through the GPU in Proxmox, or is it something else?

[PaulTV]

Usually the open source Nouveau drivers are installed on Linux, but that won't work with F@h. You'll need the official Nvidia drivers. Do install dkms before installing the Nvidia drivers, and let it register with dkms, so that the driver is recompiled after you patch your kernel.

[dbubbis13]

Even after reinstalling the drivers, I am still getting errors:

Code: Select all

21:03:11:I3:gpus = {
21:03:11:I3: "gpu:01:00:00": {"vendor": 4318, "device": 8584, "type": "nvidia", "supported": false, "description": "TU116 [GeForce GTX 1650] 3091"}
21:03:11:I3:}
21:03:27:W :CON1:DNS lookup failed for api.foldingathome.org
21:03:27:E :OUT1:Failed response: CONNECT
21:03:42:I1:OUT2:> GET https://api.foldingathome.org/machine/z1ySWIs6e9TX8755BROYo113DTPSOleRWKpH_9U5xuk HTTP/1.1
21:03:43:I1:OUT2:< HTTP/1.1 200 HTTP_OK
21:03:43:I1:OUT3:> GET wss://node1.foldingathome.org/ws/client HTTP/1.1
21:03:46:I1:OUT3:< HTTP/1.1 101 HTTP_SWITCHING_PROTOCOLS
21:03:46:I1:Logging into node account
21:03:47:W :Received unsupported remote command 'wus'
21:07:46:I1:Loaded cores/fahcore-a8-lin-64bit-sse2-0.0.12/FahCore_a8
21:07:46:I3:WU26:Running FahCore: /var/lib/fah-client/cores/fahcore-a8-lin-64bit-sse2-0.0.12/FahCore_a8 -dir rsGZfbpwzvOQ3vtCId3gc5vU-dbaroD4q6L7JDlVAuI -suffix 01 -version 8.3.18 -lifeline 664 -np 4
21:07:46:I3:WU26:Started FahCore on PID 22364
21:07:46:I1:WU26:*********************** Log Started 2024-12-05T21:07:46Z ***********************
21:07:46:I1:WU26:************************** Gromacs Folding@home Core ***************************
21:07:46:I1:WU26: Core: Gromacs
21:07:46:I1:WU26: Type: 0xa8
21:07:46:I1:WU26: Version: 0.0.12
21:07:46:I1:WU26: Author: Joseph Coffland <[email protected]>
21:07:46:I1:WU26: Copyright: 2020 foldingathome.org
21:07:46:I1:WU26: Homepage: https://foldingathome.org/
21:07:46:I1:WU26: Date: Jan 16 2021
21:07:46:I1:WU26: Time: 19:22:05
21:07:46:I1:WU26: Compiler: GNU 8.3.0
21:07:46:I1:WU26: Options: -faligned-new -std=c++14 -fsigned-char -ffunction-sections
21:07:46:I1:WU26: -fdata-sections -O3 -funroll-loops -fno-pie
21:07:46:I1:WU26: Platform: linux2 4.15.0-128-generic
21:07:46:I1:WU26: Bits: 64
21:07:46:I1:WU26: Mode: Release
21:07:46:I1:WU26: SIMD: sse2
21:07:46:I1:WU26: OpenMP: ON
21:07:46:I1:WU26: CUDA: OFF
21:07:46:I1:WU26: Args: -dir rsGZfbpwzvOQ3vtCId3gc5vU-dbaroD4q6L7JDlVAuI -suffix 01
21:07:46:I1:WU26: -version 8.3.18 -lifeline 664 -np 4
21:07:46:I1:WU26:************************************ libFAH ************************************
21:07:46:I1:WU26: Date: Jan 16 2021
21:07:46:I1:WU26: Time: 19:21:38
21:07:46:I1:WU26: Compiler: GNU 8.3.0
21:07:46:I1:WU26: Options: -faligned-new -std=c++14 -fsigned-char -ffunction-sections
21:07:46:I1:WU26: -fdata-sections -O3 -funroll-loops -fno-pie
21:07:46:I1:WU26: Platform: linux2 4.15.0-128-generic
21:07:46:I1:WU26: Bits: 64
21:07:46:I1:WU26: Mode: Release
21:07:46:I1:WU26:************************************ CBang *************************************
21:07:46:I1:WU26: Date: Jan 16 2021
21:07:46:I1:WU26: Time: 19:21:24
21:07:46:I1:WU26: Compiler: GNU 8.3.0
21:07:46:I1:WU26: Options: -faligned-new -std=c++14 -fsigned-char -ffunction-sections
21:07:46:I1:WU26: -fdata-sections -O3 -funroll-loops -fno-pie -fPIC
21:07:46:I1:WU26: Platform: linux2 4.15.0-128-generic
21:07:46:I1:WU26: Bits: 64
21:07:46:I1:WU26: Mode: Release
21:07:46:I1:WU26:************************************ System ************************************
21:07:46:I1:WU26: CPU: QEMU Virtual CPU version 2.5+
21:07:46:I1:WU26: CPU ID: GenuineIntel Family 15 Model 107 Stepping 1
21:07:46:I1:WU26: CPUs: 4
21:07:46:I1:WU26: Memory: 1.88GiB
21:07:46:I1:WU26:Free Memory: 634.54MiB
21:07:46:I1:WU26: Threads: POSIX_THREADS
21:07:46:I1:WU26: OS Version: 6.1
21:07:46:I1:WU26:Has Battery: false
21:07:46:I1:WU26: On Battery: false
21:07:46:I1:WU26: UTC Offset: -6
21:07:46:I1:WU26: PID: 22364
21:07:46:I1:WU26: CWD: /var/lib/fah-client/work
21:07:46:I1:WU26:********************************************************************************
21:07:46:I1:WU26:Project: 12419 (Run 95, Clone 6, Gen 106)
21:07:46:I1:WU26:Unit: 0x00000000000000000000000000000000
21:07:46:I1:WU26:Digital signatures verified
21:07:46:I1:WU26:Calling: mdrun -c frame106.gro -s frame106.tpr -x frame106.xtc -cpi state.cpt -cpt 5 -nt 4 -ntmpi 1
21:07:46:I1:WU26:Steps: first=530000000 total=535000000
21:07:48:I1:WU26:Completed 3226130 out of 5000000 steps (64%)
21:07:53:W :Received unsupported remote command 'wus'
21:10:10:I1:WU26:Caught signal SIGINT(2) on PID 22364
21:10:10:I1:WU26:Exiting, please wait. . .
21:10:11:I1:WU26:Folding@home Core Shutdown: INTERRUPTED

[PaulTV]

The message that triggers me is the one about not finding libcuda.so, that one definitely should be on your system.
Which Linux version are you running? Could you please give the output of the commands 'nvidia-smi' and 'locate libcuda.so'?
On one of my servers which matches your hardware most (a GTX 1050 Ti):

Code: Select all

[paul@fedora ~]$ nvidia-smi
Fri Dec  6 07:49:19 2024
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 550.135                Driver Version: 550.135        CUDA Version: 12.4     |
|-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA GeForce GTX 1050 Ti     Off |   00000000:07:00.0 Off |                  N/A |
| 26%   26C    P0             N/A /   75W |       0MiB /   4096MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI        PID   Type   Process name                              GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+
[paul@fedora ~]$ locate libcuda.so
/usr/lib/libcuda.so
/usr/lib/libcuda.so.1
/usr/lib/libcuda.so.550.135
/usr/lib64/libcuda.so
/usr/lib64/libcuda.so.1
/usr/lib64/libcuda.so.550.135

[toTOW]

These are probably some permission issues ... what is the guest distribution ?