Radeon ROCm 5.0 Released With Some RDNA2 GPU Support

[bridgman]

I agree with you. The lack of ROCm support for two-year-old graphics cards, along with the very limited range of supported distributions is frustrating. If Intel Arc cards support NEO I might switch over from my RDNA 1 card as soon as I can justify it.

The ROCm stack up to OpenCL on RDNA 1/2 has been included in the AMDGPU-PRO driver packages for over a year - it's the math libraries and ML framework support (which have optimized assembly code for each GPU) that are still under development.

I believe our range of supported distros is actually a bit wider than Intel's although I expect both will continue to grow as a consequence of current distro packaging/integration efforts.

[aufkrawall]

Though OpenCL driver quality of AMDGPU-PRO seemed to be hit and miss (mostly the latter) when I had an RDNA2 GPU. At some point it simply stopped working with upstream Arch kernel at all. HPC/academic customers seem to be fine enough with the OCL support AMD offers them (otherwise they'd probably go elsewhere), but for consumer purposes, AMD unfortuantely did their fair share putting the OCL patient into a coma...

[Grinness]

Reading the DOCs, nothing has changed with regards to Polaris and older dGPUS:

he following list of GPUs are enabled in the ROCm software, though full support is not guaranteed:

• GFX8 GPUs
  • "Polaris 11" chips, such as on the AMD Radeon RX 570 and Radeon Pro WX 4100
  • "Polaris 12" chips, such as on the AMD Radeon RX 550 and Radeon RX 540
• GFX7 GPUs
  • "Hawaii" chips, such as the AMD Radeon R9 390X and FirePro W9100

As described in the next section, GFX8 GPUs require PCI Express 3.0 (PCIe 3.0) with support for PCIe atomics. This requires both CPU and motherboard support. GFX9 GPUs require PCIe 3.0 with support for PCIe atomics by default, but they can operate in most cases without this capability.

https://github.com/RadeonOpenCompute/ROCm

I am running rcom 4.3 fine on my rx480 -- I'll test 5.0 if I have a chance building from Arc AUR when available

[hajj_3]

So OpenCL is basically a dead technology now. Nvidia is still putting all new features into CUDA and leaving OpenCL at 3.0 (aka 1.2). AMD is focusing on ROCm runtime and multiple tools that make it easier to port CUDA code sets to the AMD stack. I expect that OpenCL runtime support will continue for a long time, but the language itself will not advance in any meaningful way going forward.

Isn't there Intel's OneAPI? That is open source, i think i read that a university was working on adding support for AMD gpus.

[Steffo]

Can the AMD mesa driver run in parallel to AMDGPU-PRO when I want to use ROCm for i. e. PyTorch?

[Maxzor]

Can the AMD mesa driver run in parallel to AMDGPU-PRO when I want to use ROCm for i. e. PyTorch?

I would enjoy finding a good explanation about how the two stacks get along. AFAIK they don't interact.

[SvenK]

It would be nice if we could manage it to port rendering from opencl/CUDA/HIP to vulkan/opengl compute shaders, than at least some parts are where they should be.

[piorunz]

Yeah, should work!? Just do

sudo apt install rocm-opencl

only, and "upstream kernel" of course.

Uhm, what?

"You have searched for packages that names contain rocm-opencl in all suites, all sections, and all architectures.
Sorry, your search gave no results"

[Maxzor]

Uhm, what?

"You have searched for packages that names contain rocm-opencl in all suites, all sections, and all architectures.
Sorry, your search gave no results"

Must have been referring to the homemade AMD archive: https://repo.radeon.com/rocm/apt/5.0...r/rocm-opencl/

[Grinness]

Can the AMD mesa driver run in parallel to AMDGPU-PRO when I want to use ROCm for i. e. PyTorch?

Not sure what you want to do.
I run latest kernel and mesa (arch) and I have ROCM + pytorch, torchvision, torchtext, torchaudio (with ROCM support) running fine.
Mesa is for 3d applications, ROCM general purpose compute on GPU
I guess your question should have in scope only AMDGPU-PRO and MESA in parallel, not ROCM