-
## š Bug
PyTorch fails to build for gfx1010 architecture
## To Reproduce
Steps to reproduce the behavior:
1. Install ROCm 3.7
1. Clone pytorch: `git clone --recursive https://github.com/Rā¦
-
As brought up in reddit thread, a OpenCL device would be useful to support folks with AMD gpus.
Here are roughly the tasks that need to happen:
1. [ ] Add `opencl` feature flag
2. [ ] Create an `ā¦
-
What: The AMD HIP support provide the support of running Legion applications on both AMD and Nvidia GPUs with HIP backend.
### Requirements:
1. AMD ROCm >=3.x.
2. CUDA if running HIP on Nvidia GPā¦
-
## š Feature
The proposed feature is to support GPU acceleration on AMD's ROCM stack.
We (AMD) have already ported pytorch and torchvision to support our ROCM stack, and all those changes are currā¦
-
RuntimeError: Failed to find native CUDA module, make sure that you compiled the code with K2_WITH_CUDA.
-
Hi, will composable_kernel provide an interface similar to cutlass?
Some projects based cutlass are difficult to hipify, such as fastertransformer, xformer etc.
-
I have used Deepspeed successfully on local cluster before, and never ran into a problem But now I try to install deepspeed in a docker instance on a server farm, and it always fails with the message:ā¦
-
When I run the command: nvidia-docker build -t maskrcnn-benchmark --build-arg CUDA=10.0 --build-arg CUDNN=7 docker/
the following problem occursļ¼/bin/sh: 1: /miniconda/bin/conda: not found
but āDockā¦
-
### Steps to reproduce
slate@2021.05.02 +rocm build fails using:
spack/develop (https://github.com/spack/spack/commit/250d5d2c00ad9d6a747499b1160bc4a460266fae from Wed May 4 13:30:15 2022 -040ā¦
-
On CentOS, I installed LLVM (v15) and Clang (v15) and started the build using CMake (v3.26). The build runs smoothly but at the very end it reached a linker error.
Here is a copy of the build outpā¦