您好，根据 https://forums.developer.nvidia.com/t/provided-ptx-was-compiled-with-an-unsupported-toolchain-error-using-cub/168292 这个回答，导致这个问题的原因是您环境中的driver版本和nvcc版本不匹配导致的，根据建议您需要升级一下您机器的driver版本。

可以参考这个文档获取CUDA版本和driver的匹配信息 https://docs.nvidia.com/deploy/cuda-compatibility/index.html

我的cuda版本和driver是匹配的。

@XYZ-916 根据这个错误提示

This indicates that the provided PTX was compiled with an unsupported toolchain. The most common reason for this, is the PTXwas generated by a compiler newer than what is supported by the CUDA driver and PTX JIT compiler.] (at

应该是您编译环境里面的nvcc的版本高于您的driver能支持的版本了，可能是您环境中 nvcc 的版本和cuda11.4不一致导致的，你能检查一下您编译环境里面的nvcc的版本吗？

# 检查下 nvcc 的地址
which nvcc
# 类似如下输出
/usr/local/cuda/bin/nvcc

# 检查下 nvcc 的版本
nvcc --version
# 类似如下输出
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Mon_Oct_11_21:27:02_PDT_2021
Cuda compilation tools, release 11.4, V11.4.152
Build cuda_11.4.r11.4/compiler.30521435_0

如果nvcc版本也没有问题，建议可以试试看使用nvidia的官方镜像 nvidia/cuda:11.4.3-cudnn8-devel-ubuntu18.04 编译或者运行来避免环境问题。

请问这个问题解决了吗，我是用PaddleOCR做推理时抱这个错，这是我的环境信息

gym@stresstest:~/PaddleOCR$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Sun_Feb_14_21:12:58_PST_2021
Cuda compilation tools, release 11.2, V11.2.152
Build cuda_11.2.r11.2/compiler.29618528_0

gym@stresstest:~$ nvidia-smi
Mon Aug 21 15:05:10 2023       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 510.47.03    Driver Version: 510.47.03    CUDA Version: 11.6     |
|-------------------------------+----------------------+----------------------+
| 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. |

请问这个问题解决了吗，我是用PaddleOCR做推理时抱这个错，这是我的环境信息

gym@stresstest:~/PaddleOCR$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Sun_Feb_14_21:12:58_PST_2021
Cuda compilation tools, release 11.2, V11.2.152
Build cuda_11.2.r11.2/compiler.29618528_0

gym@stresstest:~$ nvidia-smi
Mon Aug 21 15:05:10 2023       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 510.47.03    Driver Version: 510.47.03    CUDA Version: 11.6     |
|-------------------------------+----------------------+----------------------+
| 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. |

i solved my problem!!!
nvidia-smi show as flows ,
Fri Jul 12 19:24:40 2024
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.57.02 Driver Version: 470.57.02 CUDA Version: 11.4 |
|-------------------------------+----------------------+----------------------+

this means you can install paddlepaddle post112. i install post117 and came across this problem.

https://www.paddlepaddle.org.cn/whl/linux/gpu/develop.html

wget https://paddle-wheel.bj.bcebos.com/develop/linux/linux-gpu-cuda11.2-cudnn8-mkl-gcc8.2-avx/paddlepaddle_gpu-0.0.0.post112-cp39-cp39-linux_x86_64.whl

python -m pip install paddlepaddle-gpu==0.0.0.post112 -f paddlepaddle_gpu-0.0.0.post112-cp39-cp39-linux_x86_64.whl

and now run check.py
import paddle
paddle.utils.run_check()

output as follows, fixed the problem!

Running verify PaddlePaddle program ...
I0712 19:22:10.413522 124943 program_interpreter.cc:243] New Executor is Running.
W0712 19:22:10.414631 124943 gpu_resources.cc:119] Please NOTE: device: 0, GPU Compute Capability: 8.0, Driver API Version: 11.4, Runtime API Version: 11.7
W0712 19:22:10.415433 124943 gpu_resources.cc:164] device: 0, cuDNN Version: 8.9.
I0712 19:22:23.808157 124943 interpreter_util.cc:646] Standalone Executor is Used.
PaddlePaddle works well on 1 GPU.