# Pytorch finds cuda despite nvcc not found? **URL:** \ **Category:** Uncategorized\ **Created:** [November 23, 2022, 11:17pm UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754 "2022-11-23T23:17:57Z")\ **Posts on this page:** 10\ **Page:** 1
**Author:** ![noctildon](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/noctildon/32/55114_2.png) [@noctildon](https://discuss.pytorch.org/u/noctildon)\ **Post date:** [November 23, 2022, 11:17pm UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/1 "2022-11-23T23:17:57Z")
Pytorch sees cuda and runs well on GPU, but nvcc appears to be not found ```python import torch torch.cuda.is_available() # True torch.cuda.device_count() #1 torch.cuda.current_device() # 0 torch.cuda.get_device_name(0) # NVIDIA GeForce RTX 3090 ``` ```bash $ nvidia-smi +-----------------------------------------------------------------------------+ | NVIDIA-SMI 515.65.01 Driver Version: 515.65.01 CUDA Version: 11.7 | |-------------------------------+----------------------+----------------------+ | 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 ... Off | 00000000:01:00.0 Off | N/A | | 68% 72C P2 341W / 420W | 9876MiB / 24576MiB | 100% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | 0 N/A N/A 234092 C python3 9871MiB | +-----------------------------------------------------------------------------+ ``` ```bash $nvcc --version Command 'nvcc' not found, but can be installed with: sudo apt install nvidia-cuda-toolkit ``` Once I run `$sudo apt install nvidia-cuda-toolkit`, nvidia-smi gets removed and pytorch can’t recongize GPU. To install nvidia-smi, I run `$sudo apt install nvidia-utils-515-server`, but meanwhile nvcc get uninstalled. This looks like a chicken egg problem. ```bash $sudo updatedb; locate nvcc /etc/nvcc.profile ~/.local/lib/python3.10/site-packages/torch/share/cmake/Caffe2/Modules_CUDA_fix/upstream/FindCUDA/run_nvcc.cmak ... ``` OS: Ubuntu server 22.04 ---
**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\ **Post date:** [November 24, 2022, 12:13am UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/2 "2022-11-24T00:13:44Z")
The PyTorch binaries ship with their own CUDA runtime (as well as cuDNN, NCCL etc.) and don’t need a locally installed CUDA toolkit to execute code but only a properly installed NVIDIA driver. Your local CUDA toolkit (with the compiler) will be used if you build PyTorch from source or a custom CUDA extension. Based on your described issue, I guess your CUDA toolkit installation uninstalled the NVIDIA driver as well and/or broke it, so try a new full install and make sure CUDA applications work again. ---
**Author:** ![SandeepMenonIntSum](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/sandeepmenonintsum/32/59010_2.png) [@SandeepMenonIntSum](https://discuss.pytorch.org/u/SandeepMenonIntSum)\ **Post date:** [May 24, 2023, 6:28pm UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/3 "2023-05-24T18:28:13Z")
I started off by installed the new pytorch and I can see in the `torch.version.cuda` that it is using cuda 11.7 version. But nvcc command is not found. Also when I try to install other packages which requires CUDA\_HOME to be set I get error saying (OSError: CUDA\_HOME environment variable is not set. Please set it to your CUDA install root). How should I get around this? Should I install the nvidia-cuda-toolkit? When I downloaded the cuda toolkit 11.7 and tried to install I get the message ```auto Existing package manager installation of the driver found. It is strongly │ recommended that you remove this before continuing. │ Abort │ Continue ``` ---
**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\ **Post date:** [May 24, 2023, 7:13pm UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/4 "2023-05-24T19:13:08Z")
The PyTorch binaries ship with their required CUDA runtime dependencies, not a full CUDA toolkit with a compiler. If you want to build PyTorch from source or a custom CUDA extension you would need to install the full CUDA toolkit locally. ---
**Author:** ![SandeepMenonIntSum](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/sandeepmenonintsum/32/59010_2.png) [@SandeepMenonIntSum](https://discuss.pytorch.org/u/SandeepMenonIntSum)\ **Post date:** [May 24, 2023, 8:18pm UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/5 "2023-05-24T20:18:07Z")
I am not building Pytorch from source. Just some other repositories that require CUDA. When I try to install the full CUDA toolkit I get this message from the installer ```auto Existing package manager installation of the driver found. It is strongly │ recommended that you remove this before continuing. │ Abort │ Continue ``` Should I uninstall the cuda runtime by pytorch along with pytorch and then install the full toolkit and then pytorch again? Or is there a way to keep things intact and just install the remaining ones of the cudatoolkit. ---
**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\ **Post date:** [May 24, 2023, 11:12pm UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/6 "2023-05-24T23:12:35Z")
No, you don’t need to uninstall any PyTorch binaries or their dependencies and the warning is raised because of your already locally installed CUDA toolkit and driver. ---
**Author:** ![SandeepMenonIntSum](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/sandeepmenonintsum/32/59010_2.png) [@SandeepMenonIntSum](https://discuss.pytorch.org/u/SandeepMenonIntSum)\ **Post date:** [May 25, 2023, 1:08am UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/7 "2023-05-25T01:08:32Z")
When I try to install using ```bash sudo sh cuda_11.7.0_515.43.04_linux.run ``` I downloaded the correct versions from nvidia. I get ```auto [INFO]: Driver installation detected by command: apt list --installed | grep -e nvidia-driver-[0-9][0-9][0-9] -e nvidia-[0-9][0-9][0-9] [INFO]: Cleaning up window [INFO]: Complete [INFO]: Checking compiler version... [INFO]: gcc location: /usr/bin/gcc [INFO]: gcc version: gcc version 9.4.0 (Ubuntu 9.4.0-1ubuntu1~20.04.1) [INFO]: Initializing menu [INFO]: Setup complete [INFO]: Components to install: [INFO]: Driver [INFO]: 515.43.04 [INFO]: Executing NVIDIA-Linux-x86_64-515.43.04.run --ui=none --no-questions --accept-license --disable-nouveau --no-cc-version-check --install-libglvnd 2>&1 [INFO]: Finished with code: 256 [ERROR]: Install of driver component failed. [ERROR]: Install of 515.43.04 failed, quitting ``` ---
**Author:** ![skyler14](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/skyler14/32/64453_2.png) [@skyler14](https://discuss.pytorch.org/u/skyler14)\ **Post date:** [October 7, 2023, 9:59pm UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/8 "2023-10-07T21:59:13Z")
is there a way to get an nvcc installed and running for the cuda bound to the pytorch you guys ship (in conda here specifically). Some libraries I’m using I think rely on nvcc to check for cuda support so that would help ---
**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\ **Post date:** [October 7, 2023, 11:25pm UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/9 "2023-10-07T23:25:18Z")
You could install the matching CUDA toolkit from the NVIDIA website or could try to use the `conda` package. Also, checking for `nvcc` for CUDA support is wrong as these packages check for a build toolchain. Which packages have these checks without using `nvcc`? ---
**Author:** ![skyler14](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/skyler14/32/64453_2.png) [@skyler14](https://discuss.pytorch.org/u/skyler14)\ **Post date:** [October 8, 2023, 1:35am UTC](https://discuss.pytorch.org/t/pytorch-finds-cuda-despite-nvcc-not-found/166754/10 "2023-10-08T01:35:45Z")
There’s a pretty novice developed packaged for performant poisson blending called fpie. Upon further inspection I suspect this might be doing the nvcc check more at the C++ layer since it seems like the python is just an abstraction on top so there might be bigger problems tbh.