Anuga CUDA
xuhenry opened this issue · 1 comments
xuhenry commented
Hello, every one.
Is there anyone who using the ANGUA based on CUDA?
Thanks a lot!
Best regards,
Henry
schoeller commented
@xuhenry : I have tried it make it run, but have not quite succeeded.
I am working on Ubuntu 19.04. After having set up pycuda I have the following ouput from nvidia-smi
nvidia-smi
Thu Sep 19 09:27:06 2019
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 418.56 Driver Version: 418.56 CUDA Version: 10.1 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Quadro K2000 Off | 00000000:01:00.0 On | N/A |
| 32% 49C P8 N/A / N/A | 274MiB / 1999MiB | 2% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 2643 G /usr/lib/xorg/Xorg 34MiB |
| 0 2684 G /usr/bin/gnome-shell 53MiB |
| 0 2888 G /usr/lib/xorg/Xorg 76MiB |
| 0 4615 G /usr/bin/gnome-shell 105MiB |
+-----------------------------------------------------------------------------+
Executing the merimbula example I ran into a compiler issue. I then adjusted archM and codeM set to compute_30 as understood from https://docs.nvidia.com/cuda/cuda-compiler-driver-nvcc/index.html#options-for-steering-gpu-code-generation. Now I am stuck with abdulbudiaji/anuga-cuda#4 (comment)
If you have any hints to share I'd be delighted.
Best
Sebastian