No CUDA or OpenCL Support Detected for GPU

If you think it might be a driver problem, see viewforum.php?f=79

Moderators: Site Moderators, FAHC Science Team

gunnarre
Posts: 567
Joined: Sun May 24, 2020 7:23 pm
Location: Norway

Re: No CUDA or OpenCL Support Detected for GPU

Post by gunnarre »

aetch wrote:I think Ubuntu had been installing security updates in the background, something got broken and it took rebooting the machine to highlight it.

I'm not asking for help, the machine is running just fine.
If it was a security update for the kernel, then the Nvidia kernel module might not have been ready to be upgraded at the same time as the kernel. As long as the Nvidia driver is compatible with the kernel, it should put itself into the kernel every time you upgrade the kernel, but something could have gone wrong in that upgrade. Good to see that it was fixed.
Image
Online: GTX 1660 Super, GTX 1080, GTX 1050 Ti 4G OC, RX580 + occasional CPU folding in the cold.
Offline: Radeon HD 7770, GTX 960, GTX 950
psaam0001
Posts: 383
Joined: Mon May 18, 2020 2:02 am
Location: Ruckersville, Virginia, USA

Re: No CUDA or OpenCL Support Detected for GPU

Post by psaam0001 »

tomc001 wrote:Unfortunately, Microsoft thinks it is their computer, not yours.
And M/S can take a hike with their 16-20 character activation codes that one almost needs "Hubble Space Telescope" strength lenses to read. :P

Free open-source O/S > proprietary closed source O/S's in my book.

Paul
Neil-B
Posts: 2027
Joined: Sun Mar 22, 2020 5:52 pm
Hardware configuration: 1: 2x Xeon E5-2697v3@2.60GHz, 512GB DDR4 LRDIMM, SSD Raid, Win10 Ent 20H2, Quadro K420 1GB, FAH 7.6.21
2: Xeon E3-1505Mv5@2.80GHz, 32GB DDR4, NVME, Win10 Pro 20H2, Quadro M1000M 2GB, FAH 7.6.21 (actually have two of these)
3: i7-960@3.20GHz, 12GB DDR3, SSD, Win10 Pro 20H2, GTX 750Ti 2GB, GTX 1080Ti 11GB, FAH 7.6.21
Location: UK

Re: No CUDA or OpenCL Support Detected for GPU

Post by Neil-B »

For me it depends on circumstances ... Free Open Source works/is best in certain environments and for certain use cases but in others not so (anecdote follows):

A large government organisation (in a country I will not identify) dropped the use of a certain enterprise software vendor (supplying proprietary closed source suite of software integral to functioning of organisation) due to the percieved high cost of annual licensing (low 8 figure annual bill) along with concerns about the tie in issues and openly mandated the use of Free Open Source alternatives (banning any new use and any further development of capability with the proprietary closed source suite of software). Five years later after trying to develop an alternative infrastructure capability and trashing all development schedules and existing capabilities whilst doing so the decision was reversed - Post mortem indicated that the impact of this attempted move to Free Open Source to save money had actually costs far more (mid 9 figure over 5 years - not factoring in damage to business) than would ever be spent on pre-existing arrangements and that the organisation had simply not been able to replicate in house using this approach anything close to the complexity or maturity of the Proprietary Closed Source capability ... Given the vendor concerned has 10,000s of developers and has been around for decades it is I suppose not surprising that a few 1000 developers in house could not replicate the functionality and robustness, but hey someone sold the Organisation Leadership on the idea Free Open Source is best.

The same organisation does however use both Windows and Linux across the estate dependant upon the specific uses/needs .. so it isn't all bad for Free Open Source :)
2x Xeon E5-2697v3, 512GB DDR4 LRDIMM, SSD Raid, W10-Ent, Quadro K420
Xeon E3-1505Mv5, 32GB DDR4, NVME, W10-Pro, Quadro M1000M
i7-960, 12GB DDR3, SSD, W10-Pro, GTX1080Ti
i9-10850K, 64GB DDR4, NVME, W11-Pro, RTX3070

(Green/Bold = Active)
psaam0001
Posts: 383
Joined: Mon May 18, 2020 2:02 am
Location: Ruckersville, Virginia, USA

Re: No CUDA or OpenCL Support Detected for GPU

Post by psaam0001 »

I was expressing a personal opinion, based on what I am now using the better part of my computing horsepower for (F@H).

However, it will be up to those who know the limitations of whether the software they need to have developed for a specific "mission critical" purpose, can function on a specific O/S, and be secure/reasonably fail-safe at the same time.

Paul
GalTriX
Posts: 1
Joined: Fri Dec 22, 2023 7:50 am

Re: No CUDA or OpenCL Support Detected for GPU

Post by GalTriX »

Hello there!!

I decided to build a F@H machine w\ 2 NVIDIA 1050 that were siting around, and choose Ubuntu as OS because seemed to have better PPD.

All this is an adventure, and following the same spirit I choose also Ubuntu 23.10. Had some beginner difficulties but after Ubuntu+F@H+Nvidia Drivers consolidation good numbers start to pop up, until I rebooted.

17:10:38:WARNING:FS01:No CUDA or OpenCL 1.2+ support detected for GPU slot 01: gpu:1:0 GP107 [GeForce GTX 1050 Ti] 2138. Disabling.
17:10:38:WARNING:FS02:No CUDA or OpenCL 1.2+ support detected for GPU slot 02: gpu:3:0 GP107 [GeForce GTX 1050 LP] 1862. Disabling.
17:10:38:ERROR:No valid folding configuration


I got Disabled Status @ folding slots. Later found out that reinstalling drivers with different version got GPUs enabled, until I had to reboot, until I ran out of versions to switch.

So, now I'm stuck. How can I enable the GPUs?

FAH Control Log:

Code: Select all

*********************** Log Started 2023-12-21T17:10:37Z ***********************
17:10:37:******************************* libFAH ********************************
17:10:37:       Date: Oct 20 2020
17:10:37:       Time: 20:36:39
17:10:37:   Revision: 5ca109d295a6245e2a2f590b3d0085ad5e567aeb
17:10:37:     Branch: master
17:10:37:   Compiler: GNU 8.3.0
17:10:37:    Options: -faligned-new -std=c++11 -fsigned-char -ffunction-sections
17:10:37:             -fdata-sections -O3 -funroll-loops -fno-pie
17:10:37:   Platform: linux2 5.8.0-1-amd64
17:10:37:       Bits: 64
17:10:37:       Mode: Release
17:10:37:****************************** FAHClient ******************************
17:10:37:    Version: 7.6.21
17:10:37:     Author: Joseph Coffland <joseph@cauldrondevelopment.com>
17:10:37:  Copyright: 2020 foldingathome.org
17:10:37:   Homepage: https://foldingathome.org/
17:10:37:       Date: Oct 20 2020
17:10:37:       Time: 20:39:00
17:10:37:   Revision: 6efbf0e138e22d3963e6a291f78dcb9c6422a278
17:10:37:     Branch: master
17:10:37:   Compiler: GNU 8.3.0
17:10:37:    Options: -faligned-new -std=c++11 -fsigned-char -ffunction-sections
17:10:37:             -fdata-sections -O3 -funroll-loops -fno-pie
17:10:37:   Platform: linux2 5.8.0-1-amd64
17:10:37:       Bits: 64
17:10:37:       Mode: Release
17:10:37:       Args: -v --chdir /var/snap/folding-at-home-fcole90/common
17:10:37:     Config: /var/snap/folding-at-home-fcole90/common/config.xml
17:10:37:******************************** CBang ********************************
17:10:37:       Date: Oct 20 2020
17:10:37:       Time: 18:37:59
17:10:37:   Revision: 7e4ce85225d7eaeb775e87c31740181ca603de60
17:10:37:     Branch: master
17:10:37:   Compiler: GNU 8.3.0
17:10:37:    Options: -faligned-new -std=c++11 -fsigned-char -ffunction-sections
17:10:37:             -fdata-sections -O3 -funroll-loops -fno-pie -fPIC
17:10:37:   Platform: linux2 5.8.0-1-amd64
17:10:37:       Bits: 64
17:10:37:       Mode: Release
17:10:37:******************************* System ********************************
17:10:37:        CPU: Intel(R) Core(TM) i5-3470 CPU @ 3.20GHz
17:10:37:     CPU ID: GenuineIntel Family 6 Model 58 Stepping 9
17:10:37:       CPUs: 4
17:10:37:     Memory: 15.57GiB
17:10:37:Free Memory: 14.77GiB
17:10:37:    Threads: POSIX_THREADS
17:10:37: OS Version: 6.5
17:10:37:Has Battery: false
17:10:37: On Battery: false
17:10:37: UTC Offset: 1
17:10:37:        PID: 1084
17:10:37:        CWD: /var/snap/folding-at-home-fcole90/581
17:10:37:         OS: Linux 6.5.0-14-generic x86_64
17:10:37:    OS Arch: AMD64
17:10:37:       GPUs: 2
17:10:37:      GPU 0: Bus:1 Slot:0 Func:0 NVIDIA:5 GP107 [GeForce GTX 1050 Ti] 2138
17:10:37:      GPU 1: Bus:3 Slot:0 Func:0 NVIDIA:5 GP107 [GeForce GTX 1050 LP] 1862
17:10:37:       CUDA: Not detected: cuInit() returned 100
17:10:37:     OpenCL: Not detected: clGetPlatformIDs() returned -1001
17:10:37:***********************************************************************
17:10:37:<config>
17:10:37:  <!-- Client Control -->
17:10:37:  <client-threads v='6'/>
17:10:37:  <cycle-rate v='4'/>
17:10:37:  <cycles v='-1'/>
17:10:37:  <disable-sleep-when-active v='true'/>
17:10:37:  <exit-when-done v='false'/>
17:10:37:  <fold-anon v='false'/>
17:10:37:  <idle-seconds v='300'/>
17:10:37:  <open-web-control v='false'/>
17:10:37:  <update-gpus-txt v='true'/>
17:10:37:
17:10:37:  <!-- Configuration -->
17:10:37:  <config-rotate v='true'/>
17:10:37:  <config-rotate-dir v='configs'/>
17:10:37:  <config-rotate-max v='16'/>
17:10:37:
17:10:37:  <!-- Debugging -->
17:10:37:  <assignment-servers>
17:10:37:    assign1.foldingathome.org assign2.foldingathome.org assign3.foldingathome.org assign4.foldingathome.org 
17:10:37:  </assignment-servers>
17:10:37:  <auth-as v='true'/>
17:10:37:  <capture-directory v='capture'/>
17:10:37:  <capture-on-error v='false'/>
17:10:37:  <capture-packets v='false'/>
17:10:37:  <capture-requests v='false'/>
17:10:37:  <capture-responses v='false'/>
17:10:37:  <capture-sockets v='false'/>
17:10:37:  <debug-sockets v='false'/>
17:10:37:  <exception-locations v='true'/>
17:10:37:  <stack-traces v='false'/>
17:10:37:
17:10:37:  <!-- Error Handling -->
17:10:37:  <max-slot-errors v='10'/>
17:10:37:  <max-unit-errors v='5'/>
17:10:37:
17:10:37:  <!-- Folding Core -->
17:10:37:  <checkpoint v='15'/>
17:10:37:  <core-priority v='idle'/>
17:10:37:  <cpu-usage v='100'/>
17:10:37:  <gpu-usage v='100'/>
17:10:37:  <no-assembly v='false'/>
17:10:37:
17:10:37:  <!-- Folding Slot Configuration -->
17:10:37:  <cause v='HUNTINGTONS'/>
17:10:37:  <client-subtype v='LINUX'/>
17:10:37:  <client-type v='normal'/>
17:10:37:  <cpu-species v='X86_PENTIUM_II'/>
17:10:37:  <cpu-type v='AMD64'/>
17:10:37:  <cpus v='-1'/>
17:10:37:  <disable-viz v='false'/>
17:10:37:  <gpu v='true'/>
17:10:37:  <gpu-beta v='false'/>
17:10:37:  <max-packet-size v='normal'/>
17:10:37:  <os-species v='UNKNOWN'/>
17:10:37:  <os-type v='LINUX'/>
17:10:37:  <project-key v='0'/>
17:10:37:  <smp v='true'/>
17:10:37:
17:10:37:  <!-- GUI -->
17:10:37:  <gui-enabled v='true'/>
17:10:37:
17:10:37:  <!-- HTTP Server -->
17:10:37:  <allow v='127.0.0.1'/>
17:10:37:  <connection-timeout v='60'/>
17:10:37:  <deny v='0/0'/>
17:10:37:  <http-addresses v='0:7396'/>
17:10:37:  <https-addresses v=''/>
17:10:37:  <max-connect-time v='900'/>
17:10:37:  <max-connections v='800'/>
17:10:37:  <max-request-length v='52428800'/>
17:10:37:  <min-connect-time v='300'/>
17:10:37:
17:10:37:  <!-- Logging -->
17:10:37:  <log v='log.txt'/>
17:10:37:  <log-color v='true'/>
17:10:37:  <log-crlf v='false'/>
17:10:37:  <log-date v='false'/>
17:10:37:  <log-date-periodically v='21600'/>
17:10:37:  <log-domain v='false'/>
17:10:37:  <log-header v='true'/>
17:10:37:  <log-level v='true'/>
17:10:37:  <log-no-info-header v='true'/>
17:10:37:  <log-redirect v='false'/>
17:10:37:  <log-rotate v='true'/>
17:10:37:  <log-rotate-dir v='logs'/>
17:10:37:  <log-rotate-max v='16'/>
17:10:37:  <log-short-level v='false'/>
17:10:37:  <log-simple-domains v='true'/>
17:10:37:  <log-thread-id v='false'/>
17:10:37:  <log-thread-prefix v='true'/>
17:10:37:  <log-time v='true'/>
17:10:37:  <log-to-screen v='true'/>
17:10:37:  <log-truncate v='false'/>
17:10:37:  <verbosity v='3'/>
17:10:37:
17:10:37:  <!-- Network -->
17:10:37:  <proxy v=':8080'/>
17:10:37:  <proxy-enable v='false'/>
17:10:37:  <proxy-pass v='*****'/>
17:10:37:  <proxy-user v=''/>
17:10:37:
17:10:37:  <!-- Process Control -->
17:10:37:  <child v='false'/>
17:10:37:  <daemon v='false'/>
17:10:37:  <fork v='false'/>
17:10:37:  <pid v='false'/>
17:10:37:  <pid-file v='FAHClient.pid'/>
17:10:37:  <respawn v='false'/>
17:10:37:  <service v='false'/>
17:10:37:
17:10:37:  <!-- Remote Command Server -->
17:10:37:  <command-address v='0.0.0.0'/>
17:10:37:  <command-allow-no-pass v='127.0.0.1'/>
17:10:37:  <command-deny-no-pass v='0/0'/>
17:10:37:  <command-enable v='true'/>
17:10:37:  <command-port v='36330'/>
17:10:37:
17:10:37:  <!-- Slot Control -->
17:10:37:  <auto-conf v='true'/>
17:10:37:  <idle v='false'/>
17:10:37:  <max-shutdown-wait v='60'/>
17:10:37:  <pause-on-battery v='true'/>
17:10:37:  <pause-on-start v='false'/>
17:10:37:  <paused v='false'/>
17:10:37:  <power v='light'/>
17:10:37:
17:10:37:  <!-- User Information -->
17:10:37:  <machine-id v='0'/>
17:10:37:  <passkey v='*****'/>
17:10:37:  <team v='262045'/>
17:10:37:  <user v='GalTrix'/>
17:10:37:
17:10:37:  <!-- Web Server -->
17:10:37:  <web-allow v='127.0.0.1'/>
17:10:37:  <web-deny v='0/0'/>
17:10:37:  <web-enable v='true'/>
17:10:37:
17:10:37:  <!-- Web Server Sessions -->
17:10:37:  <session-cookie v='sid'/>
17:10:37:  <session-lifetime v='86400'/>
17:10:37:  <session-timeout v='3600'/>
17:10:37:
17:10:37:  <!-- Work Unit Control -->
17:10:37:  <dump-after-deadline v='true'/>
17:10:37:  <max-queue v='16'/>
17:10:37:  <max-units v='0'/>
17:10:37:  <next-unit-percentage v='99'/>
17:10:37:  <stall-detection-enabled v='false'/>
17:10:37:  <stall-percent v='5'/>
17:10:37:  <stall-timeout v='1800'/>
17:10:37:
17:10:37:  <!-- Folding Slots -->
17:10:37:  <slot id='1' type='GPU'>
17:10:37:    <pci-bus v='1'/>
17:10:37:    <pci-slot v='0'/>
17:10:37:  </slot>
17:10:37:  <slot id='2' type='GPU'>
17:10:37:    <pci-bus v='3'/>
17:10:37:    <pci-slot v='0'/>
17:10:37:  </slot>
17:10:37:</config>
17:10:37:Trying to access database...
17:10:38:Successfully acquired database lock
17:10:38:WARNING:FS01:No CUDA or OpenCL 1.2+ support detected for GPU slot 01: gpu:1:0 GP107 [GeForce GTX 1050 Ti]  2138.  Disabling.
17:10:38:WARNING:FS02:No CUDA or OpenCL 1.2+ support detected for GPU slot 02: gpu:3:0 GP107 [GeForce GTX 1050 LP] 1862.  Disabling.
17:10:38:ERROR:No valid folding configuration
******************************* Date: 2023-12-22 *******************************
07:17:15:ERROR:Receive error: 4: Interrupted system call
When I run nvidia-smi:

Code: Select all

Fri Dec 22 08:31:34 2023       
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.129.03             Driver Version: 535.129.03   CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+
| 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 | 00000000:01:00.0  On |                  N/A |
|  0%   50C    P0              N/A /  72W |    946MiB /  4096MiB |      4%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------+
|   1  NVIDIA GeForce GTX 1050        Off | 00000000:03:00.0 Off |                  N/A |
|  0%   21C    P8              N/A /  70W |      9MiB /  2048MiB |      0%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------+
                                                                                         
+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
|    0   N/A  N/A      2016      G   /usr/lib/xorg/Xorg                          445MiB |
|    0   N/A  N/A      2220      G   /usr/bin/gnome-shell                        156MiB |
|    0   N/A  N/A      3035      G   ...irefox/3216/usr/lib/firefox/firefox      330MiB |
|    0   N/A  N/A      4813      G   /usr/bin/nautilus                             8MiB |
|    1   N/A  N/A      2016      G   /usr/lib/xorg/Xorg                            4MiB |
+--------------------------------------------------------------------------------
-------+
bikeaddict
Posts: 187
Joined: Sun May 03, 2020 1:20 am

Re: No CUDA or OpenCL Support Detected for GPU

Post by bikeaddict »

The system needs ocl-icd-opencl-dev installed for OpenCL support.

See viewtopic.php?t=36824

Not sure why it doesn't find CUDA. On Fedora Linux, it's part of xorg-x11-drv-nvidia-cuda-libs that's installed by RPM Fusion.
Post Reply