-
when I run `pip install megablocks` I get this:
```
clang: error: unsupported option '--ptxas-options=-v'
clang: error: unsupported option '--generate-code=arch=compute_90,code=sm_90'…
-
### Steps to reproduce
In packages.yaml, having:
```
rocm-openmp-extras:
buildable: false
externals:
- spec: rocm-openmp-extras@5.7.1
prefix: /opt/rocm-5.7.1
```
is not en…
-
I have profiled an application like so:
```shell
$ omniperf profile -n --device=0 -- ./test_binary
```
Now I want, for multiple dispatch, raw some roofline:
```shell
$ # Calling this does n…
-
Changes need to be made in rocm-cmake to support installation of libraries at `$ROCM_PATH/lib/migraphx`.
First, a `PRIVATE` flag should be added to the `rocm_install_targets` that will install the …
-
The command used to run mlperf inference for resnet50 model on ubuntu with rocm is below:
cm run script --tags=run-mlperf,inference \
--model=resnet50 \
--implementation=reference \
--…
-
### Problem Description
CI failing for SLES 15 with PR #412. The staging will not be promoted with this failure.
GPU - `MI100[gfx908:sramecc+:xnack-]`
```
[989](http://math-ci.amd.com/job/mainli…
-
### Problem Description
I am trying to compile some of the HIP samples on my computer, I downloaded and installed the HIP SDK and everything seemed fine, I had to install perl, as mentioned in #271…
-
*TL;DR: Please add support for AMD GPUs on Linux through ROCm.*
Greetings,
I was wondering if it is being considered to support ROCm for inference? Since an RX 7900 XTX is currently the only option …
-
### Your question:
I'm interested in distributing packages that use HIP, AMD's framework for GPU compute. It's basically their equivalent of CUDA. There are a couple of packages on conda-forge rela…
-
Dear team,
Firstly, thanks for your amazing works, congratulation.
As pytorch support RocM (AMD GPU card) is it possible to extend the support too for flashlight ?
References:
- https://pytorc…