|Enrolment status||New students|
|Student type||Domestic students, International students|
|Level of study||Higher Degree by Research|
|Study area||Engineering and Computing|
|HDR funding type||Living stipend scholarship|
|Scholarship value||$40,000 per annum tax-free for three years, with the possibility of two 6-month extensions at the standard RTP rate ($28,092 per annum, indexed annually) in approved circumstances|
|Opening date||10 March 2020|
|Closing date||30 June 2020|
Compilers are an essential ingredient of the computing base. Software developers need to be able to trust their compilers because an error in a compiler can manifest as erroneous generated code for any of the myriad of programs it compiles. The traditional approach to compiler verification is testing, but this cannot cover all cases and hence can only show the presence of errors, not their absence. The gold standard for producing trusted software is mechanized formal verification by theorem provers. This research project focuses on verifying optimization passes of the Graal compiler, using the Isabelle interactive theorem prover.
Verification of a compiler is a complex, time-intensive task that requires expertise in formal semantics, verification methods, mechanized theorem-proving technology, and of course compilers. Fortunately, compilers are organized into multiple passes, thus enabling the verification process to be structured to match the passes, and allowing one to build up a verified compiler pass-by-pass. Our project is focusing on modelling the Graal Intermediate Representation (IR), which is a sophisticated graph structure, and then verifying several key optimisation passes of the open-source Graal compiler. If you are interested in working with cutting-edge compiler technology and doing a PhD as part of this project, we invite you to join our team.
To be eligible, you must meet the entry requirements for a higher degree by research.