#OpenCL

🧵 Turns out, after a journey of many hours with #Lightworks… (and hoped-for alternatives)

… they have introduced new video scopes based on #OpenCL which do not always work.

Since I just recently got to work AMD OpenCL on my #Linux box, these new scopes changed #LKWS behavior to constant crashing, yet it enabled to run #DavinciResolve at all.

The Lightworks #NLE people were smart enough to put a switch in the project workspace's menu to deactivate these “advanced scopes”… yet the switch forgets its setting and needs to be flicked on every start.

This is exactly underlining my criticism about Lightworks these days: They have good ideas, they try to tackle them, but they make things worse by not fully managing.

E.g., there's VST3 audio plugin support now which does work – but on export of the video, it introduces an ugly noise in the beginning of your exported clip. So it's perfect but still useless.

Now trying to remember what I was about to do half a day ago.

{ Pengouin(BSD|Pdt) }hucste@framapiaf.org
2026-02-25

RE: framapiaf.org/@journalduhacker

Petit article mémo relatif à l'ajout d'OpenCL pour les GPU AMD, sous Debian, de mon cru.

Pour rappel, le support d'OpenCL n'est pas inclus dans le pilote Open Source `amdgpu`, il est nécessaire d'ajouter le paquet lié à ROCm… qui lui est aussi fait par AMD et Open Source

#Debian #amdgpu #ROCm #OpenCL #OpenMP #HIP

2026-02-22

HPC++: An LLVM-Based Automatic Parallelization Framework with Heterogeneous CPU–GPU Execution

#OpenCL #HPC #LLVM

hgpu.org/?p=30590

Pekka Jääskeläinenpekka@fosstodon.org
2026-02-16

A journal article about chipStar is finally published! doi.org/10.1177/10943420261423

chipStar is a compilation tool/runtime for porting CUDA/HIP applications on OpenCL/SPIR-V-capable platforms. Its origin is in the HIPCL "prototype" developed within my research group mainly by Michal Babej. Then refined to chipStar in close collab with Argonne, Intel and Paulius Velesko (PGLC) among others. #opencl #spirv #cuda #ijhpca

The code is here: github.com/CHIP-SPV/chipStar

2026-02-12

Interestingly, the segfault seems to be from clang::DiagonsticOptions from llvm-21, so this is likely one of the neverending series of issues that arise when mixing #LLVM versions that has plagued #FLOSS #OpenCL ICDs since forever. I thought we had finally gotten rid of these issues, but apparently this is not the case: dynamic loading independent libraries that depend on different versions of LLVM *still* causes issues. Is it again some unversioned global? Another weird gimmick? Who knows!

2026-02-11

OpenCL, SYCL и матрицы

В данной статье я описываю свой опыт разработки приложений с OpenCL/SYCL. Вычисления на видеокартах ассоциируются преимущественно с графикой, научными вычислениями и с недавних пор с нейросетями. Но чаще всего с графикой. Тем не менее, графические процессоры обладают свойствами, за счёт которых их очень удобно использовать в задачах, напрямую не связанных с перечисленным выше. И главное из этих свойств – массовый параллелизм. Самый краткий ввод в гетерогенное программирование: у нас есть две роли — хост и девайс , задача хоста – формировать задачи и отдавать их на девайс, в то время как задача девайса обработать их и вернуть результат. При этом хост и девайс могут быть одним вычислительным устройством (это не обязательно CPU + GPU, так как CPU может отдавать задачи сам себе).

habr.com/ru/articles/994986/

#gpgpu #opencl #sycl #gpu_computing

2026-02-08

Inside VOLT: Designing an Open-Source GPU Compiler (Tool)

#OpenCL #CUDA #FPGA #Compilers #Package

hgpu.org/?p=30544

2026-02-01

ProfInfer: An eBPF-based Fine-Grained LLM Inference Profiler

#OpenCL #LLM

hgpu.org/?p=30518

Nils D.nilsd
2026-01-25

Dropped the launcher, now using plain wine prefixes and finally also got working in 2. I think it became useable under now 😍

Valentin Churavy :julia:vchuravy@fosstodon.org
2026-01-25

#JuliaCon 2026 will have a minisymposium on "Julia, GPUs, and Accelerators" hosted by Julian Samaroo, Tim Besard, and @vchuravy find out more on pretalx.com/juliacon-2026/talk and submit your talks through the Call for Proposals juliacon.org/2026/cfp/ until Feb 28th!

#julialang #cuda #rocm #opencl #gpu

The JuliaGPU community has been a strong presence at JuliaCon for many years, and continues to be a strong foundation of Julia's overall computing ecosystem. In 2026, we propose to organize a minisymposium specifically focused on the usage and programming of GPUs (and other accelerators) in Julia. There is some overlap with an HPC minisymposium, however we intend to focus our track on very GPU-specific content or low-level details that make JuliaGPU tick. Additionally, material relating to non-GPU devices (such as TPUs, APUs, IPUs, etc.) are very welcome!
2026-01-23

hm, not sure what I did, but today I seem to be able to use PoCL, Clover and rusticl simultaneously in the same program on Debian Trixie (previously my attempts failed with mysterious segfaults). Using fraktaler-3-3.1 with environment variables RUSTICL_ENABLE=radeonsi RUSTICL_FEATURES=fp64 for my AMD RX 580 GPU.

Seem rusticl is significantly slower than Clover, hopefully Clover is not deleted before rusticl catches up.

#OpenCL

benchmarks of various computation backends in Fraktaler-3 on a desktop with AMD 2700X CPU and AMD RX 580 GPU.

the figures are logarithmic, an increase of 1 unit corresponds to 2x increase in speed.

the figures show rusticl is slower than Clover, and PoCL is mostly faster than the regular CPU backend.
2026-01-19

Call for Submissions for IWOCL 2026 has been extended until January 25th!

Now is the time to submit on of four types of submissions:

+ Technical Presentations & Case Studies (500 word abstract)
+ Short Research Papers (3-4 page paper)
+ Full Research Papers (8-12 page paper)
+ Posters (1-2 page abstract)

Get your submission in now and join us in for another inspiring workshop at IWOCL 2026!
iwocl.org
#OpenCL #SYCL #IWOCL2026

IWOCL Submissions extended until January 25, 2026
2026-01-07

Khronos Group Welcomes Lisuan Technology as Contributor Member

Lisuan Technology is a company dedicated to the research and development of graphics rendering GPUs. Founded by leading figures in the GPU industry, it possesses capabilities in large-scale chip R&D, top-level architecture design, software stack design, and mass production. They are interested in #OpenCL, #OpenGL, #OpenGLES, #Vulkan, #WebGL, #SPIR, and #OpenXR.
lisuantech.com/

2025-12-19

Okay #Gamer #Gamers... A friend's 16 year old is looking to build a PC for #gaming over the holidays... Now, I have a bunch of old PC components lying around... So I have to ask the question... Is an #AMD #Radeon Vega RX 64 or a WX9100 usable for a modern game machine? I remember these as monsters for #OpenCL -- but I don't think I *ever* connected a monitor to them.

Mostly interested in running #Linux for now, but might move over to #Windows eventually...

#Linux_Gaming #LinuxGaming #Steam #SteamOS #Minecraft

cathillcathill
2025-12-01

Finally enabled OpenCL support in Darktable 5.0.1 on Debian 13 with AMD integrated GPU! First, `sudo apt-get install ocl-icd-libopencl1 rocm-opencl-icd`. However, `darktable-cltest -d verbose` shows "failed to compile program `demosaic_ppg.cl'!" and "fatal error: malformed or corrupted AST file: 'could not find file '/usr/lib/llvm-17/lib/clang/17/include/opencl-c-base.h'". Solution: `sudo apt-get install libclang-common-17-dev`.

2025-11-23

Inside VOLT: Designing an Open-Source GPU Compiler

#OpenCL #CUDA #Compilers

hgpu.org/?p=30372

Client Info

Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst