HetGPU: The pursuit of making binary compatibility towards GPUs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Yiwei, Zheng, Yusheng, Yu, Tong, Quinn, Andi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912441560989696
author Yang, Yiwei
Zheng, Yusheng
Yu, Tong
Quinn, Andi
author_facet Yang, Yiwei
Zheng, Yusheng
Yu, Tong
Quinn, Andi
contents Heterogeneous GPU infrastructures present a binary compatibility challenge: code compiled for one vendor's GPU will not run on another due to divergent instruction sets, execution models, and driver stacks . We propose hetGPU, a new system comprising a compiler, runtime, and abstraction layer that together enable a single GPU binary to execute on NVIDIA, AMD, Intel, and Tenstorrent hardware. The hetGPU compiler emits an architecture-agnostic GPU intermediate representation (IR) and inserts metadata for managing execution state. The hetGPU runtime then dynamically translates this IR to the target GPU's native code and provides a uniform abstraction of threads, memory, and synchronization. Our design tackles key challenges: differing SIMT vs. MIMD execution (warps on NVIDIA/AMD vs. many-core RISC-V on Tenstorrent), varied instruction sets, scheduling and memory model discrepancies, and the need for state serialization for live migration. We detail the hetGPU architecture, including the IR transformation pipeline, a state capture/reload mechanism for live GPU migration, and an abstraction layer that bridges warp-centric and core-centric designs. Preliminary evaluation demonstrates that unmodified GPU binaries compiled with hetGPU can be migrated across disparate GPUs with minimal overhead, opening the door to vendor-agnostic GPU computing.
format Preprint
id arxiv_https___arxiv_org_abs_2506_15993
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HetGPU: The pursuit of making binary compatibility towards GPUs
Yang, Yiwei
Zheng, Yusheng
Yu, Tong
Quinn, Andi
Hardware Architecture
Distributed, Parallel, and Cluster Computing
Heterogeneous GPU infrastructures present a binary compatibility challenge: code compiled for one vendor's GPU will not run on another due to divergent instruction sets, execution models, and driver stacks . We propose hetGPU, a new system comprising a compiler, runtime, and abstraction layer that together enable a single GPU binary to execute on NVIDIA, AMD, Intel, and Tenstorrent hardware. The hetGPU compiler emits an architecture-agnostic GPU intermediate representation (IR) and inserts metadata for managing execution state. The hetGPU runtime then dynamically translates this IR to the target GPU's native code and provides a uniform abstraction of threads, memory, and synchronization. Our design tackles key challenges: differing SIMT vs. MIMD execution (warps on NVIDIA/AMD vs. many-core RISC-V on Tenstorrent), varied instruction sets, scheduling and memory model discrepancies, and the need for state serialization for live migration. We detail the hetGPU architecture, including the IR transformation pipeline, a state capture/reload mechanism for live GPU migration, and an abstraction layer that bridges warp-centric and core-centric designs. Preliminary evaluation demonstrates that unmodified GPU binaries compiled with hetGPU can be migrated across disparate GPUs with minimal overhead, opening the door to vendor-agnostic GPU computing.
title HetGPU: The pursuit of making binary compatibility towards GPUs
topic Hardware Architecture
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2506.15993