GFTrans: an on-the-fly static analysis framework for code performance profiling

Jie Li

doi:10.3389/fdata.2026.1779935

GFTrans: an on-the-fly static analysis framework for code performance profiling

2026-02-27

1

PRE

AI

DOI：10.3389/fdata.2026.1779935

Original

Original request for help

Save

Abstract

En 中文

Improving software efficiency is crucial for maintenance, but pinpointing runtime bottlenecks becomes increasingly difficult as systems expand. Traditional dynamic profiling tools require full build-execution cycles, creating significant latency that impedes agile development. To address this, we introduce GFTrans, a static analysis framework that predicts c program performance without execution. GFTrans utilizes a Transformer architecture with a novel anchor-based embedding technique to integrate control flow and data dependencies into a unified sequence. Additionally, a dynamic gating mechanism fuses these semantic representations with 16 handcrafted statistical features to comprehensively capture code complexity. Evaluated on a dataset of real-world GitHub c functions with high-precision runtime labels, GFTrans outperforms baseline models like Random Forest and Code2Vec, achieving 78.64% accuracy. The system identifies potential bottlenecks in milliseconds, enabling developers to perform optimization effectively during the coding phase.

Keywords:

code representation learning

control flow and data flow

graph linearization

on-the-fly profiling

performance prediction

static analysis

GFTrans: an on-the-fly static analysis framework for code performance profiling

Abstract

Journal

Organization