🦞🌯 Lobster Roll

Thread

Directly writing AMD GPU assembler by hacking the OpenGL shader interface (onedrive.live.com)
Shaders in OpenGL are written in GLSL which gets compiled and uploaded to the GPU by the driver. There is no way of writing hardware-specific GPU assembly code directly. Well… officially, anyway. As the last slide says: GPU-based OS anyone?

Stories related to "Directly writing AMD GPU assembler by hacking the OpenGL shader interface" across the full archive.

Directly writing AMD GPU assembler by hacking the OpenGL shader interface (onedrive.live.com)
Shaders in OpenGL are written in GLSL which gets compiled and uploaded to the GPU by the driver. There is no way of writing hardware-specific GPU assembly code directly. Well… officially, anyway. As the last slide says: GPU-based OS anyone?
How AMD Is Taking Standard C/C++ Code to Run Directly on GPUs (phoronix.com)
How AMD Is Taking Standard C/C++ Code to Run Directly on GPUs (phoronix.com)
AMD Lands LLVM Flang Fortran Runtime Support for Compiling Directly on the GPU (phoronix.com)
My First Multi-GPU Kernel: Writing All-to-All for AMD MI300X (gau-nernst.github.io)
AMD Unveils Hardware-Based Virtualized GPU (marketwatch.com)
Sorry for the press release, but it's more technical than most.
Neanderthal 0.6.0: now on Nvidia, AMD, and Intel GPU (Clojure matrix library) (neanderthal.uncomplicate.org)
This is a major release of Neanderthal, a fast native & GPU matrix library: In this release, spotlight is on the new GPU engine, that: Works on all three major hardware platforms: AMD, Nvidia, and Intel Works on all three major operating systems: Linux, Windows, and OS X Is even faster, so it is...
[RFC] Using DC in amdgpu for upcoming GPU (lists.freedesktop.org)
Intel Announces Plans To Develop Discrete GPUs, Hires AMDs Raja Koduri as Chief Architect (anandtech.com)
Writing Testable GPU Code (blog.esciencecenter.nl)
GPU-Centered Font Rendering Directly from Glyph Outlines (terathon.com)
AMD Announces 7nm Rome CPUs and MI60 GPUs (tomshardware.com)
Deep Learning in Clojure from Scratch to GPU: CUDA and OpenCL, Nvidia and AMD (dragan.rocks)
Deep Learning from Scratch to GPU - 8 - The Forward Pass (CUDA, OpenCL, Nvidia, AMD, Intel) (dragan.rocks)
Deep Learning from Scratch to GPU: The Backward Pass (CUDA, OpenCL, Nvidia, AMD, Intel) (dragan.rocks)
AMD's Big Navi and Xbox Series X GPU 'Arden' source code stolen and leaked (tomshardware.com)
Setting the amdgpu HDMI Pixel Format on Linux (wezm.net)
FreeBSD 64bit Assembler (amd64) (honeyguide.eu)
AMD announces CPU and GPU events on October 8th and 28th (twitter.com)
Writing an assembler (blog.steve.fi)
An Ecosystem of libraries and tools for writing and executing extremely fast GPU code fully in Rust (github.com)
Writing an open source GPU driver – without the hardware (collabora.com)
GPU-Centered Font Rendering Directly from Glyph Outlines (2017) (jcgt.org)
Faster, more flexible inference on both AMD and NVIDIA GPUs using AITemplate (ai.facebook.com)
nvtop: GPUs process monitoring for AMD, Intel and NVIDIA (github.com)
Training LLMs with AMD MI250 GPUs and MosaicML (mosaicml.com)
Hardcore Vulkan debugging – Digging deep on Linux + AMDGPU (themaister.net)
AMD Open-Source GPU Kernel Driver Above 5 Million Lines, Entire Linux Kernel At 34.8 Million (phoronix.com)
AMD's Long and Winding Road to the Hybrid CPU-GPU Instinct MI300A (nextplatform.com)
AMD's Long and Winding Road to the Hybrid CPU-GPU Instinct MI300A (nextplatform.com)