{"ID":6537447,"CreatedAt":"2026-07-14T02:54:43.516908796Z","UpdatedAt":"2026-07-15T03:28:55.185153975Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.11636","arxiv_id":"2607.11636","title":"On Thread Convergence","abstract":"We introduce a notion of convergence for the nodes and edges of a control-flow graph that captures whether a barrier placed at that location is guaranteed to synchronize all threads of a thread block in every execution. Convergence analysis lets a compiler determine when a barrier lies in a uniformly executed region and therefore avoid the code transformations otherwise required to implement thread-block barriers correctly on warp-synchronous hardware. We formalize convergent nodes, convergent edges, and well-synchronized programs; give two inference rules (a branch rule and a merge rule); and present a linear-time iterative work-list algorithm that propagates convergence information bidirectionally through the flow graph. We then describe refinements that improve precision using single-entry single-exit region information, path information, and thread-variance information.","short_abstract":"We introduce a notion of convergence for the nodes and edges of a control-flow graph that captures whether a barrier placed at that location is guaranteed to synchronize all threads of a thread block in every execution. Convergence analysis lets a compiler determine when a barrier lies in a uniformly executed region an...","url_abs":"https://arxiv.org/abs/2607.11636","url_pdf":"https://arxiv.org/pdf/2607.11636v1","authors":"[\"Vinod Grover\",\"Manjunath Kudlur\"]","published":"2026-07-13T14:53:37Z","proceeding":"cs.PL","tasks":"[\"cs.PL\"]","methods":"[]","has_code":false}