All (especially Greg and Beth),
My group has been working on some C analyses and refactorings for OpenACC. We’d like to move our code into PLDT.
Currently, we’re working off of a separate GitHub repo. We have three projects: - CLI - command line interface, so you can run the refactorings outside Eclipse (not planning to contribute unless you really want it)
- Core - various data flow analyses, dependence analysis and Fourier-Motzkin solver, refactorings, and parser for OpenACC directives
- UI - obvious
My question is, where should we put this code in PTP/PLDT?
- Option 1: Contribute 2 or 3 new projects (maybe org.eclipse.ptp.pldt.openacc.refactoring.{core,ui})
- Option 2: Split the existing org.eclipse.ptp.pldt.openacc into separate core and UI projects, and then contribute to those
- Option 3: Dump everything into the existing org.eclipse.ptp.pldt.openacc (completely losing the core/UI separation)
What is your preference?
Our static analyses might be useful to CDT (constant propagation, reaching definitions, etc.), but
Jeff
--Jeffrey L. Overbey, Ph.D.Assistant ProfessorDepartment of Computer Science & Software EngineeringAuburn University3101H Shelby Center / (334) 844-8834 / http://jeff.over.bz
|