一句话：提出开源全栈 RISC-V GPU 架构 Skybox，以软硬件混合方案加速 Vulkan 图形渲染并支持端到端 GPU 研究。
图形渲染是 GPU 上计算与内存密集的负载，但为提升面积与能效，GPU 微架构需兼顾图形与其他通用计算，而现有开源全栈 GPU 研究平台稀缺。本文提出 Skybox，一个集成软件、编译器、硬件与仿真环境的开源全栈 GPU 架构，基于可编程 RISC-V ISA。它探索软件渲染与硬件渲染的设计空间，提出混合微架构加速 Vulkan 图形 API，并引入编译器和系统优化。作者在 Altera 与 Xilinx FPGA 上评估，在 Altera Stratix 10 上生成并执行 32 核（512 线程）图形处理器，230 MHz 下峰值填充率达 3.7 GPixels；Skybox 是首个支持 Vulkan API 的开源全栈 GPU 软硬件实现。
Graphics rendering remains one of the most compute intensive and memory bound applications of GPUs and has been driving their push for performance and energy efficiency since its inception. Early GPU architectures focused only on accelerating graphics rendering and implemented dedicated fixed- function rasterizer hardware to speed-up their rendering pipeline. As GPUs have become more programmable and ubiquitous in other application domains such as scientific computing, machine learning, graph analytics, and crypto-currency, generalizing GPU microarchitectures for area and power efficiency becomes necessary, especially for mobile and IoT devices. In this work, we present Skybox, a full-stack open-source GPU architecture with integrated software, compiler, hardware, and simulation environment, that enables end-to-end GPU research. Using Skybox, we explore the design space of software versus hardware graphics rendering and propose and hybrid micro-architecture that accelerates the state-of-the art Vulkan graphics API. Skybox also introduces novel compiler and system optimizations to support its unique RISC-V ISA baseline. We evaluated Skybox on high- end Altera and also Xilinx FPGAs. We were able to generate and execute a 32 cores (512 threads) Skybox graphics processor on Altera Stratix 10 FPGA, delivering a peak fill rate of 3.7 GPixels at 230 MHz. Skybox is the first open-source full-stack GPU software and hardware implementation that supports the Vulkan API
@inproceedings{paperbot10083,
title = {Skybox: Open-Source Graphic Rendering on Programmable RISC-V GPUs},
author = {Blaise Tine and Varun Saxena and Santosh Srivatsan and Joshua R. Simpson and Fadi Alzammar and Liam Cooper and Hyesoon Kim},
booktitle = {ASPLOS},
year = {2023},
doi = {10.1145/3582016.3582024}
}