Subregion Analysis and Bounds Check Elimination for High Level Arrays

Abstract

For decades, the design and implementation of arrays in programming languages has reflected a natural tension between productivity and performance. Recently introduced HPCS languages (Chapel, Fortress and X10) advocate the use of high-level arrays for improved productivity. For example, high-level arrays in the X10 language support rank-independent specification of multidimensional loop and array computations using regions and points. Three aspects of X10 high-level arrays are important for productivity but pose significant performance challenges: high-level accesses are performed through point objects rather than integer indices, variables containing references to arrays are rank-independent, and all subscripts in a high-level array access must be checked for bounds violations. The first two challenges have been addressed in past work. In this paper, we address the third challenge of optimizing the overhead of array bounds checks by developing a novel region-based interprocedural array bounds analysis to automatically identify redundant checks. Elimination of redundant checks reduces the runtime overhead of bounds checks, and also enables further optimization by removing constraints that arise from precise exception semantics. We have implemented an array bounds check elimination algorithm that inserts special annotations that are recognized by a modified JVM. We also introduce array views, a high-level construct that improves productivity by allowing the programmer to access the underlying array through multiple views. We describe a technique for optimizing away the overhead of many common cases of array views in X10. Our experiments show that eliminating bounds checks using the results of the analysis described in this paper improves the performance of our benchmarks by up to 22% over JIT compilation.

Accelerating Habanero-Java programs with OpenCL generation

“…As discussed in Section 3.3, we instead use HJ array-views [13] to provide multi-dimensional syntax backed by a one-dimensional array. In Figure 3, line 3 shows the declaration of a one-dimensional array.…”

Legion: Expressing locality and independence with logical regions

“…However, once data and tasks have been placed they are fixed, which mandates that data movement be explicitly managed by user level code or implicitly by the compiler [16]. Recently X10 has introduced regions into the compiler's intermediate representation [17]. Unlike Legion, regions in X10 are not visible to the programmer but are inferred from high level arrays through static analysis.…”

Regent

“…Another related class of work is represented by PGAS and related models, such as UPC [14,2], Titanium [35], Chapel [16,15], X10 [18,17,25] and HPX [26]. Regent is similar to PGAS programming models in so far as pointers created in Regent are valid everywhere in the machine.…”