-
We use OpenCL 1D image for LUT kernel which can be easily replaced by OpenCL buffer like we do in HIP.
This causes some failures in MI300.
-
**Describe the motivation for the feature request**
Today we figured out why, while using Intels OpenCL runtime, we would see initially performance drops in our simulation. For the first hour of simu…
-
想全志A133 OPENCL性能测试，用下面的命令跑起来后发现CPU 一直有一个核被占用100%,GPU 利用率在30-40%的样子，这个正常吗，opengl 模式怎么会消耗CPU资源这么多
./benchmark_bin \
--model_file=MobileNetV3_small_x1_0_infer/inference.pdmodel \
--param_fil…
-
**Motivation**
It would be amazing to be able to run PyTorch nodes on AMD GPU via OpenCL.
**Description**
I think theoretically now it should be possible because now we have [RustICL](https://doc…
-
Hi,
for a benchmark software i'm making, could you add the support for an OpenCL backend ?
awesome work man !
-
Leaf, and now Spearow, always promised OpenCL but has yet to deliver. I think that's a shame!
There are plenty of great folks working on ML experiments for Rust which feature OpenCL or frameworks t…
ghost updated
6 years ago
-
is the code prepared/ready for adding an opencl backend to it?
if so, any high-level implementation plan would be welcome! i may give it a try.
now that NVIDIA also provides an opencl API for their …
-
On (Linux) CUDA machine, the backend returned by `af.get_backend()` remains at unified even if the backend is correctly set as OpenCL. `af.get_active_backend()` correctly returns OpenCL as the backend…
-
The performance gap is found in https://github.com/intel/intel-xpu-backend-for-triton/pull/2347
Need to investigate root cause of the performance drops of the column major B matrix case.
Roughly 1…
-
When the OpenCL backend falls back from GPU to CPU device, the following line causes an assert:
https://github.com/BinomialLLC/basis_universal/blob/6775d334fe8d7af95eb003512c4f5aa8cef41bbd/encoder/…