rocm-arch
ROCm-OpenCL-Runtime
Our great sponsors
rocm-arch | ROCm-OpenCL-Runtime | |
---|---|---|
19 | 15 | |
329 | 171 | |
0.3% | - | |
6.1 | 0.0 | |
13 days ago | 3 months ago | |
Shell | C++ | |
- | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
rocm-arch
- I need assistance interpreting a specific error and understanding the purpose of the code that is referenced by the error, please.
- Finally, ROCm packages in [community]!
-
Limit AUR build cores; paru
rocm-arch README
-
Looking for co-maintainers (AUR)
. I use this for my personal PKGBUILDS: https://github.com/acxz/pkgbuilds and for smaller projects with less than 30ish packages like https://github.com/acxz/gazebo-arch/ and https://github.com/rocm-arch/rocm-arch/. and of course for projects that have larger amount of PKGBUILDS just mirroring them int their own github organization like ros-noetic is the way to go.
-
First time in 2 years I was able to get Blender running with an AMD GPU on Linux!
I had to install rocm from the AUR. https://github.com/rocm-arch/rocm-arch
-
Build for unofficial supported GPU (6700XT - gfx1031)
Obviously i followed that instruction with the parameter gfx1031, also tried to recompile all rocm packages in rocm-arch/rocm-arch repository with gfx1031 but none works.
-
"error making: rocfft" when installing rocm-tensorflow
However, there is an unofficial Arch Linux ROCm project at https://github.com/rocm-arch/rocm-arch (not supported by AMD).
-
Tensorflow with Radeon GPU
As others said, rocm is the solution. However compiling all the rocm stuff is very painful and last I tried, I was not able to compile not all rocm packages on AUR. There is a binary repository called arch4edu, which has rocm and a tensorflow-rocm package. I haven't tried it myself, but if you don't mind the security implications of using a binary repository (of course containers are always an option), then this will probably be the easiest way for tensorflow on amd.
-
The State Of ROCm For HPC In Early 2021 With CUDA Porting Via HIP, Rewriting With OpenMP - Phoronix
The Arch community created a repository just with PKGBUILDs to install ROCM on Arch, here: https://github.com/rocm-arch/rocm-arch - but given that ROCM itself depends a lot on LLVM upstream it's quite hard to add it to the community repository (and also the whole installation procedure is VERY slow), like they talked in this issue: https://github.com/rocm-arch/rocm-arch/issues/262
ROCm-OpenCL-Runtime
-
ROCm Is AMD's #1 Priority, Executive Says
Its not that they're supporting buggy code, they just downgraded the quality of their implementation significantly. They made the compiler a lot worse when they swapped to rocm
https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime/iss... is the tracking issue for it filed a year ago, which appears to be wontfix largely because its a lot of work
OpenCL still unfortunately supports quite a few things that vulkan doesn't, which makes swapping away very difficult for some use cases
-
rocm-opencl (rocm-opencl-runtime) rx 6600 xt support
There's https://docs.amd.com/bundle/ROCm-Installation_FAQ/page/Frequently_Asked_Questions.html which leads to a page which doesn't list any gpus that I can see, there's https://rocm.docs.amd.com/en/latest/release/gpu_os_support.html which lists "RDNA2" , there's https://github.com/RadeonOpenCompute/ROCm/issues/1698 which is from last year and mentions changing an env for the RX 6600 XT (navi 23) . Not a lot is mentioned in the readme of https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime .
-
Install ROCm Fedora 38
$ dnf info rocm-opencl Installed Packages Name : rocm-opencl Version : 5.4.3 Release : 2.fc38 Architecture : x86_64 Size : 1.7 M Source : rocm-opencl-5.4.3-2.fc38.src.rpm Repository : @System From repo : updates Summary : ROCm OpenCL Runtime URL : https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime License : MIT Description : ROCm OpenCL language runtime. : Supports offline and in-process/in-memory compilation.
-
First time in 2 years I was able to get Blender running with an AMD GPU on Linux!
Eg this bug about shared cl/gl textures with mipmaps being broken has now breached its first birthday, without even acknowledgement - basic cl/gl functionality here. This bug took a year for a fix to make its way into a public driver. And this fairly performance critical bug is just "wontfix", and also a significant downgrade from their old driver stack
-
So far I'm unconvinced a 34MB binary blob is more free than OpenZFS.
Its definitely workable if you're willing to put in the effort (except for things that are straight up broken in some cases, like device side enqueue), but there are some issues that require.. fairly major workarounds
-
New NVIDIA Open-Source Linux Kernel Graphics Driver Appears
Their implementation is here: https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime
-
C++ Show and Tell - April 2022
After a lot of moderately annoyed testing, I discovered that the AMD OpenCL implementation is.. rather dumb. If any two kernels share any arguments arguments, it inserts a command barrier between the two, hard-stalling the GPU. After filing a bug, it turns out this is wontfix as well, which is doubly bad. There's no set of flags in OpenCL that you can use to fix this either
-
[TPU] AMD ROCm 4.5 Drops "Polaris" Architecture Support
What's particularly bizarre, is that with one bug report I filed, they claim to have fixed it internally in april, but.. no public driver has ever been released with the fix. For 7 months? Which is just a bizarre software development process
-
Who is to blame for the bad OpenCL Performance? Blender or AMD?
Could you report these issues here https://github.com/RadeonOpenCompute/ROCm-OpenCL-Runtime? Regarding the device side enqueue issue, could you attach a simple test case to the issue that reproduces the crash? The current pastebin link doesn't give enough info.
What are some alternatives?
ROCm - AMD ROCmâ„¢ Software - GitHub Home [Moved to: https://github.com/ROCm/ROCm]
tensorflow-upstream - TensorFlow ROCm port
rocFFT - Next generation FFT implementation for ROCm
aurpublish - PKGBUILD management framework for the Arch User Repository
seer - Seer - a gui frontend to gdb
HIP - HIP: C++ Heterogeneous-Compute Interface for Portability
GPU-Raytracer - GPU Raytracer from scratch in C++/CUDA
nvchecker - New version checker for software releases
pkgbuilds - PKGBUILDs for Arch Linux
chrgfx - Converts to and from tile based graphics from retro video game hardware