pytorch/serve

RuntimeError: CUDA error: no kernel image is available for execution on the device

Open

#1,754 opened on Jul 22, 2022

View on GitHub
 (7 comments) (0 reactions) (0 assignees)Java (3,844 stars) (790 forks)batch import
help wanted

Description

I am deploying a model using torchserve. It was easily deployed on Tesla K80 GPU. But now when i shifted it to newer GPU Nvidia A30. I am getting this error

torch_version: '1.12.0+cu113' cudnn_version: 8302

image

PFA Nvidia-smi Screenshot

image

Contributor guide

Tech stack
pythonpytorch
Domain
machine learningbackend
Issue type
bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
half day
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
mostly clear
Prerequisites
Familiarity with PyTorch and CUDAUnderstanding of GPU compute capabilities
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30
Research direction
This issue requires investigating the CUDA compute capability of the A30 GPU (8.0) and ensuring the installed PyTorch version (1.12.0+cu113) includes kernels for that architecture. Check the PyTorch documentation on CUDA compute capabilities and possibly reinstall a compatible PyTorch build. Also verify the torchserve configuration for GPU device selection.
RuntimeError: CUDA error: no kernel image is available for execution on the device · pytorch/serve#1754 | Good First Issue