CUDA ERROR UNSUPPORTED PTX VERSION (222)
Posted: Mon Nov 22, 2021 2:17 pm
Hi everyone, I hope you are all as well as can be.
I have had some issues recently, not sure what changed. My daily PPD had increased, all was well until yesterday I noticed 1 machine had dropped a GPU PPD, it had dropped nearly 200000 credits per unit.
So I checked the logs and found this error. Error loading CUDA module: CUDA_ERROR_UNSUPPORTED_PTX_VERSION (222)
I cannot figure what has changed that has dropped that GPU from running and throwing the error out.
Both machines are semi remote connected with me visiting the site twice a week.
Anyone any clues to what may have caused this?
Thanks in advance.
Extract from log.
13:48:14:WU03:FS01:0x22:There are 4 platforms available.
13:48:14:WU03:FS01:0x22:Platform 0: Reference
13:48:14:WU03:FS01:0x22:Platform 1: CPU
13:48:14:WU03:FS01:0x22:Platform 2: OpenCL
13:48:14:WU03:FS01:0x22: opencl-device 1 specified
13:48:14:WU03:FS01:0x22:Platform 3: CUDA
13:48:14:WU03:FS01:0x22: cuda-device 1 specified
13:48:33:WU03:FS01:0x22:Attempting to create CUDA context:
13:48:33:WU03:FS01:0x22: Configuring platform CUDA
13:48:34:WU03:FS01:0x22:Failed to create CUDA context:
13:48:34:WU03:FS01:0x22:Error loading CUDA module: CUDA_ERROR_UNSUPPORTED_PTX_VERSION (222)
13:48:34:WU03:FS01:0x22:Attempting to create OpenCL context:
13:48:34:WU03:FS01:0x22: Configuring platform OpenCL
13:48:39:WU03:FS01:0x22: Using OpenCL on platformId 0 and gpu 1
Win 10 Enterprise 64
i5 6600K @3.5Ghz
2X1060 3GB NVIDIA
16GB Ram
Loads of virtual memory
Shes an old machine but gets stuff done, all I could afford at the time, gradually changed HDs and installed a few SSDs and hybrid SSDs.
I have had some issues recently, not sure what changed. My daily PPD had increased, all was well until yesterday I noticed 1 machine had dropped a GPU PPD, it had dropped nearly 200000 credits per unit.
So I checked the logs and found this error. Error loading CUDA module: CUDA_ERROR_UNSUPPORTED_PTX_VERSION (222)
I cannot figure what has changed that has dropped that GPU from running and throwing the error out.
Both machines are semi remote connected with me visiting the site twice a week.
Anyone any clues to what may have caused this?
Thanks in advance.
Extract from log.
13:48:14:WU03:FS01:0x22:There are 4 platforms available.
13:48:14:WU03:FS01:0x22:Platform 0: Reference
13:48:14:WU03:FS01:0x22:Platform 1: CPU
13:48:14:WU03:FS01:0x22:Platform 2: OpenCL
13:48:14:WU03:FS01:0x22: opencl-device 1 specified
13:48:14:WU03:FS01:0x22:Platform 3: CUDA
13:48:14:WU03:FS01:0x22: cuda-device 1 specified
13:48:33:WU03:FS01:0x22:Attempting to create CUDA context:
13:48:33:WU03:FS01:0x22: Configuring platform CUDA
13:48:34:WU03:FS01:0x22:Failed to create CUDA context:
13:48:34:WU03:FS01:0x22:Error loading CUDA module: CUDA_ERROR_UNSUPPORTED_PTX_VERSION (222)
13:48:34:WU03:FS01:0x22:Attempting to create OpenCL context:
13:48:34:WU03:FS01:0x22: Configuring platform OpenCL
13:48:39:WU03:FS01:0x22: Using OpenCL on platformId 0 and gpu 1
Win 10 Enterprise 64
i5 6600K @3.5Ghz
2X1060 3GB NVIDIA
16GB Ram
Loads of virtual memory
Shes an old machine but gets stuff done, all I could afford at the time, gradually changed HDs and installed a few SSDs and hybrid SSDs.