New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
failed to build from source with CUDA architecture of 86 #1024
Comments
What's your CUDA version? Looks like you're trying to use an arch version which is newer than your CUDA version's nvcc will support. |
Thanks for your quick reply! The CUDA version is 11.5 |
I have the same architecture 86. With CUDA 11.8 I haven't such issue. |
@jasonppy — this command works for me without issue:
a few odd things about yours -- |
Thanks for your reply. The command in your reply leads to the same error |
@jasonppy ah, you're on 0.3.2 — looks like new versions of CMake that set the CUDA arch flag in the way they do don't work with the copied CUDA shims. Can you checking |
are you referring to #1033 ? This is not me... |
Bug Description
I tried to build from source by following the instructions in readme, at step
cmake .. -DCMAKE_BUILD_TYPE=Release -DFL_BUILD_ARRAYFIRE=ON -DCMAKE_CUDA_ARCHITECTURES=86
, it gives an errorNote that A40 has architecture 86.
Platform and Hardware
OS:
NAME="Ubuntu"
VERSION="20.04.3 LTS (Focal Fossa)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 20.04.3 LTS"
VERSION_ID="20.04"
CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
gcc (Ubuntu 9.3.0-17 ubuntu1~20.04) 9.3.0
cmake version 3.23.0-rc1
Thanks!
The text was updated successfully, but these errors were encountered: