The aggressive optimization of heavily used kernels is an important problem in high-performance computing. However, both general purpose compilers and highly specialized tools such as superoptimizers often do not have sufficient static knowledge of restrictions on program inputs that could be exploited to produce the very best code. For many applications, the best possible code is conditionally correct: the optimized kernel is equal to the code that it replaces only under certain preconditions on the kernel’s inputs. The main technical challenge in producing conditionally correct optimizations is in obtaining non-trivial and useful conditions and proving conditional equivalence formally in the presence of loops. We combine abstract interpretation, decision procedures, and testing to yield a verification strategy that can address both of these problems. This approach yields a superoptimizer for x86 that in our experiments produces binaries that are often multiple times faster than those produced by production compilers.
Wed 28 OctDisplayed time zone: Eastern Time (US & Canada) change
13:30 - 15:00 | |||
13:30 22mTalk | Conditionally Correct Superoptimization OOPSLA Rahul Sharma Stanford University, Eric Schkufza Stanford University, Berkeley Churchill Stanford University, Alex Aiken Stanford University DOI | ||
13:52 22mTalk | Selective Control-Flow Abstraction via Jumping OOPSLA Sam Blackshear University of Colorado at Boulder, USA, Bor-Yuh Evan Chang University of Colorado at Boulder, USA, Manu Sridharan Samsung Research America Link to publication | ||
14:15 22mTalk | Automating Grammar Comparison OOPSLA Ravichandhran Madhavan EPFL, Switzerland, Mikaël Mayer EPFL, Switzerland, Sumit Gulwani Microsoft Research, USA, Viktor Kunčak EPFL, Switzerland Link to publication | ||
14:37 22mTalk | Reasoning about the POSIX File System: Local Update and Global Pathnames OOPSLA DOI |