BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250110T023313Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241206T154300
DTEND;TZID=Asia/Tokyo:20241206T155400
UID:siggraphasia_SIGGRAPH Asia 2024_sess149_papers_194@linklings.com
SUMMARY:GPU Coroutines for Flexible Splitting and Scheduling of Rendering 
 Tasks
DESCRIPTION:Technical Papers\n\nShaokun Zheng, Xin Chen, and Zhong Shi (Ts
 inghua University); Ling-Qi Yan (University of California Santa Barbara); 
 and Kun Xu (Tsinghua University)\n\nWe introduce coroutines into GPU kerne
 l programming, providing an automated solution for flexible splitting and 
 scheduling of rendering tasks. This approach addresses a prevalent challen
 ge in harnessing the power of modern GPUs for complex, imbalanced graphics
  workloads like path tracing. Usually, to accommodate the SIMT execution m
 odel and latency-hiding architecture, developers have to decompose a monol
 ithic mega-kernel into smaller sub-tasks for improved thread coherence and
  reduced register pressure. However, involving the handling of intricate n
 ested control flows and numerous interdependent program states, this proce
 ss can be exceedingly tedious and error-prone when performed manually.\n\n
 Coroutines, a building block for asynchronous programming in many high-lev
 el CPU languages, exhibit untapped potential for restructuring GPU kernels
  due to their versatility in control representation. By extending Luisa [Z
 heng et al. 2022], we implement an asymmetric, stackless coroutine model w
 ith programming language support and multiple built-in schedulers for mode
 rn GPUs. To showcase the effectiveness of our model and implementation, we
  examine them in different application scenarios, including path tracing, 
 SDF rendering, and incorporation with custom passes.\n\nRegistration Categ
 ory: Full Access, Full Access Supporter\n\nLanguage Format: English Langua
 ge\n\nSession Chair: Valentin Deschaintre (Adobe Research)
URL:https://asia.siggraph.org/2024/program/?id=papers_194&sess=sess149
END:VEVENT
END:VCALENDAR