Search papers, labs, and topics across Lattice.
The paper introduces MIP, a novel approach to automated patch porting that addresses implicit inconsistencies arising from non-local differences between codebases. MIP leverages an LLM, compiler diagnostics, and code analysis to identify and resolve these inconsistencies, using either compiler feedback when source identifiers exist in the target or retrieving matched code segments as mapping knowledge. Experiments on cross-fork and cross-branch patch porting demonstrate that MIP resolves more than twice as many patches compared to existing methods, validated by a user study with an industry partner.
Automated patch porting can now handle twice as many inconsistencies by leveraging LLMs to reason about global codebase differences, not just local context.
Promptly porting patches from a source codebase to its variants (e.g., forks and branches) is essential for mitigating propagated defects and vulnerabilities. Recent studies have explored automated patch porting to reduce manual effort and delay, but existing approaches mainly handle inconsistencies visible in a patch's local context and struggle with those requiring global mapping knowledge between codebases. We refer to such non-local inconsistencies as implicit inconsistencies. Implicit inconsistencies pose greater challenges for developers to resolve due to their non-local nature. To address them, we propose MIP, which enables collaboration among an LLM, a compiler, and code analysis utilities. MIP adopts different strategies for different cases: when source identifiers exist in the target codebase, it leverages compiler diagnostics; otherwise, it retrieves matched code segment pairs from the two codebases as mapping knowledge for mitigation. Experiments on two representative scenarios, cross-fork and cross-branch patch porting, show that MIP successfully resolves more than twice as many patches as the best-performing baseline in both settings. A user study with our industry partner further demonstrates its practical effectiveness.
