# VortexSynth：用于GPU架构研究的基于剖析指导的基准重构

> 原文链接：https://robotics.hellobike.com/papers/2026-10-08-2610-12661
> 频道：papers | 更新日期：2026-10-08 | 标签：arxiv, cs.RO
> 核心摘要：VortexSynth通过剖析选择热点内核并合成目标语言实现，在Vortex上以机器人应用评估GPU基准迁移。

# VortexSynth：用于GPU架构研究的基于剖析指导的基准重构

::: v-pre
> 原文链接：[VortexSynth: Profile-guided benchmark reconstruction for GPU architecture research](https://arxiv.org/abs/2610.12661)（arXiv 2610.12661 · 发布日期：2026-10-08）

**VortexSynth: Profile-guided benchmark reconstruction for GPU architecture research**

VortexSynth通过剖析选择热点内核并合成目标语言实现，在Vortex上以机器人应用评估GPU基准迁移。

## 核心要点

- 许多厂商定制工作负载和库内核因闭源或编程模型不受支持，难以直接运行在研究架构上。
- 该框架用剖析指导的基准重构，从参考工作负载选择热点内核并合成目标编程语言实现。
- 流程验证输出等价性，并用翻译的代理实现测量剖析差异，将签名差距交给智能体迭代改进。
- 在Vortex open-ISA GPGPU上评估时，所有活跃内核对通过输出等价检查，案例显示可降低剖析距离。

**验证范围与局限**：验证范围为Vortex这一open-ISA GPGPU上的机器人应用领域。

## 英文摘要 (Abstract)

GPU architecture research requires representative benchmarks that can execute on new designs. However, many vendor-tailored workloads and library kernels cannot run directly on research architectures because their implementations are closed-source or rely on programming models unsupported by the target architecture. Migrating existing kernels requires manual effort to adapt their implementations and validate their behavior. This creates a need to automate workload migration to target platforms. We present VortexSynth, a framework that uses profile-guided benchmark reconstruction to construct GPU benchmarks from reference workloads. Our workflow selects hot kernels through profiling, synthesizes implementations in the target programming language, verifies output equivalence, and measures profile differences using a translated proxy implementation. The resulting profile signature gap is supplied to an agent for iterative refinement. We evaluate the method on Vortex, an open-ISA GPGPU, using robotics as the application domain. All active kernel pairs pass both output-equivalence checks, and the case studies show that iterative refinement can reduce profile distance.

作者：Huanzhi Pu, Chulhyung Park, Chihyo Ahn, Ramyad Hadidi, Hyesoon Kim
:::

## 原始参考来源 (Sources)

- [arXiv 2610.12661：VortexSynth: Profile-guided benchmark reconstruction for GPU architecture research](https://arxiv.org/abs/2610.12661)