Efficient detection of determinacy races in Cilk programs

Abstract

A parallel multithreaded program that is ostensibly deterministic may nevertheless behave nondeterministically due to bugs in the code. These bugs are called determinacy races, and they result when one thread updates a location in shared memory while another thread is concurrently accessing the location. We have implemented a provably efficient determinacy-race detector for Cilk, an algorithmic multithreaded programming language. If a Cilk program run on a given input data set has a determinacy race, our debugging tool, which we call the "Nondeterminator," guarantees to detect and localize the race.The core of the Nondeterminator is an asymptotically efficient serial algorithm (inspired by Tarjan's nearly linear-time leastcommon-ancestors algorithm) for detecting determinacy races in series-parallel directed acyclic graphs. For a Cilk program that runs in T time on one processor and uses v shared-memory locations, the Nondeterminator runs in O (T α(v; v)) time, where α is Tarjan's functional inverse of Ackermann's function, a very slowly growing function which, for all practical purposes, is bounded above by 4. The Nondeterminator uses at most a constant factor more space than does the original program. On a variety of Cilk program benchmarks, the Nondeterminator exhibits a slowdown of less than 12 compared with the serial execution time of the original optimized code, which we contend is an acceptable slowdown for debugging purposes.

The Cilk++ concurrency platform

“…The Cilkscreen race detector is based on provably good algorithms [2,8,14] developed originally for MIT Cilk. In a single serial execution on a test input for a deterministic program, Cilkscreen guarantees to report a race bug if the race bug is exposed: that is, if two different schedulings of the parallel code would produce different results.…”

The Cilk++ concurrency platform

“…The Cilkscreen race detector is based on provably good algorithms [2,6,11] developed originally for MIT Cilk. In a single serial execution on a test input for a deterministic program, Cilkscreen guarantees to report a race bug if the race bug is exposed: that is, if two different schedulings of the parallel code would produce different results.…”

Nested parallelism in transactional memory

“…As in the "series-parallel parse tree" of (Feng and Leiserson 1997), an internal node is either an S-node or a P-node. An S-node denotes series composition of the subtrees (i.e., the subtrees must execute in left-to-right order), whereas a Pnode denotes parallel composition.…”

“…Bags are merged dynamically in a way similar to the SPbags (Feng and Leiserson 1997) in the local tier of SP-hybrid (Bender et al 2004;Fineman 2005) using a disjoint-sets data structure (Cormen et al 2001, Chapter 21). Since traces execute on a single processor, we do not lock the data structure on update (UNION) operations.…”

“…A runtime system (e.g., for Cilk) dynamically schedules the program on the number of processors, p, specified at execution time. If the language also permits only "properly nested" parallelism, i.e., any program execution can be represented as a "series-parallel dag" or "series-parallel parse tree" (Feng and Leiserson 1997), then a Cilk-like work-stealing scheduler executes the program in a provably efficient manner (Blumofe et al 1996). A natural question arises: how can transactions be added to a dynamic multithreaded language such as Cilk?…”