-
See subj.
Wondering why AMD always releases their frameworks on Linux / Unix platforms only.
Guys, do you have in mind that GPGPU developers can't build their products for Windows users (98% of them…
-
Now I am trying to translate CuPy(known as Numpy-like array API which uses Cuda library)
to hip compatible code.
So I first used hipify-perl for converting the original code, then I got to replaci…
-
Hi all,
I've tried to install newest build on Ubuntu 16.04 ROCM 1.8 machine.
I used the command
sudo WITH_ROCM=1 NO_CUDA=1 python setup.py install
but I always get the same error **cuda_runtim…
-
I installed ROCm 1.6:
```
apt-get install rocm rocm-opencl rocm-opencl-dev
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following addition…
-
Total number of operators - 210
CUDA operators - 77 (some of underlying dependency on CUB/Thrust)
cuDNN operators - 11
Tasks includes but not limited to :
- [ ] hipify ops with cuda implemen…
-
when I try build rocFFT form source, I meet
~/ROCm/rocFFT/release# CXX=/opt/rocm/bin/hcc cmake ..
-- HCC compiler set; ROCm backend selected [ CXX=/opt/rocm/bin/hcc cmake ... ]
-- Building with R…
-
-
Could you revisit the OpenCL implementation in the current landscape (2016)?
Maybe things have improved.
Worst case, ignore *ntel and focus on AMD/Nvidia.
ghost updated
6 years ago
-
When I run the following command
**bazel test -c opt --config=sycl --test_output=all //tensorflow/python/kernel_tests:basic_gpu_test**
the following log is displayed,
```
WARNING: /home/gautha…
-
Is it just completely missing? Especially now that the AMDGPU-Pro drivers work for Linux... That's dope...
maxcr updated
6 years ago