Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines
Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.
REPEAT: Do NOT report bugs for outdated packages!
https://wiki.archlinux.org/title/Bug_reporting_guidelines
Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.
REPEAT: Do NOT report bugs for outdated packages!
FS#60511 - [python-pytorch-cuda] missing support for compute capability 3.0, 3.2, 3.5
Attached to Project:
Community Packages
Opened by Benoit Brummer (trougnouf) - Saturday, 20 October 2018, 22:06 GMT
Last edited by Doug Newgard (Scimmia) - Saturday, 20 October 2018, 23:37 GMT
Opened by Benoit Brummer (trougnouf) - Saturday, 20 October 2018, 22:06 GMT
Last edited by Doug Newgard (Scimmia) - Saturday, 20 October 2018, 23:37 GMT
|
Detailspython-pytorch-cuda is missing TORCH_CUDA_ARCH_LIST 3.0, 3.2, and 3.5 since 1.0rc1-2. I believe these architectures are still supported by NVidia and I was able to restore functionality with my GTX770 by setting TORCH_CUDA_ARCH_LIST=3.0 and remaking the package.
|
This task depends upon