In file src\backend\cuda\kernel/reduce.hpp:429 CUDA Error (209): no kernel image is available for execution on the device 0# af::array::host in afcuda 1# af::array::host in afcuda 2# af::array::host in afcuda 3# af::array::host in afcuda 4# af::array::host in afcuda 5# af::array::host in af 6# arrayfire::algorithm::max_all<f64>
1392492818 opened this issue · 1 comments
1392492818 commented
Description
Reproducible Code and/or Steps
System Information
Checklist
- Using the latest available ArrayFire release
- GPU drivers are up to date
9prady9 commented
@1392492818 Going forward, can you please open bugs properly with title and description in respective locations of GitHub issue. Pasting everything in title of the issue is not going to help us help you.
As far as the issue goes, I think the message means you are trying to run your program on CUDA device for which kernel hasn't been compiled for.
Please open an issue with code reproducible and the system information where you are trying to do this. Also, do note that issues such as these are to be reported in upstream project https://github.com/arrayfire/arrayfire which where the fix is likely going to go in if this is a bug in code.
This project is a bindings provider for rust language.