Section: New Results

Reasoning about C11 Program Transformations

Participants : Francesco Zappa Nardelli, Thibaut Balabonski, Robin Morisset.

We have shown that the weak memory model introduced by the 2011 C and C++ standards does not permit many of common source-to-source program transformations (such as expression linearisation and "roach motel" reordering) that modern compilers perform and that are deemed to be correct. As such it cannot be used to define the semantics of intermediate languages of compilers, as, for instance, LLVM aimed to. We consider a number of possible local fixes, some strengthening and some weakening the model. We have evaluated the proposed fixes by determining which program transformations are valid with respect to each of the patched models. We have provided formal Coq proofs of their correctness or counterexamples as appropriate.

A paper on this work has been accepted in [13] . In collaboration with Viktor Vafeiadis (MPI-SWS, Germany).