Using the Meeting Graph Framework to Minimise Kernel Loop Unrolling for Scheduled Loops

A considerable part of program execution time is consumed by loops, so that loop optimization is highly effective especially for the innermost loops of a program. Software pipelining and loop unrolling are known methods for loop optimization. Software pipelining is advantageous in that the code becomes only slightly longer. This method, however, is difficult to apply if the loop includes branching when the parallelism is limited. On the other hand, loop unrolling, while being free of such limitations, suffers from a number of drawbacks. In particular the code size grows substantially and it is difficult to determine the optimal number of body replications. In order to solve these problems, it seems important to combine software pipelining with loop unrolling so as to utilize the advantages of both techniques while paying due regard to properties of programs under consideration and to the machine resources available. This paper describes a method for applying optimal loop unrolling and effective software pipelining to achieve this goal. Program characteristics obtained by means of an extended PDG (program dependence graph) are taken into consideration as well as machine resources. © 1998 Scripta Technica, Syst Comp Jpn, 29(9): 62–73, 1998

We address the problem of generating compact code from software pipelined loops. Although software pipelining is a powerful technique to extract fine-grain parallelism, it generates lifetime intervals spanning multiple loop iterations. These intervals require periodic register allocation (also called variable expansion), which in turn yields a code generation challenge. We are looking for the minimal unrolling factor enabling the periodic register allocation of software pipelined kernels. This challenge is generally addressed through one of: (1) hardware support in the form of rotating register files, which solve the unrolling problem but are expensive in hardware; (2) register renaming by inserting register moves, which increase the number of operations in the loop, and may damage the schedule of the software pipeline and reduce throughput; (3) post-pass loop unrolling that does not compromise throughput but often leads to impractical code growth. The latter approach relies on the proof that MAXLIVE registers (maximal number of values simultaneously alive) are sufficient for periodic register allocation (Eisenbeis et al. in PACT ’95: Proceedings of the IFIP WG10.3 working conference on Parallel Architectures and Compilation Techniques, pages 264–267, Manchester, UK, 1995; Hendren et al. in CC ’92: Proceedings of the 4th International Conference on Compiler Construction, pages 176–191, London, UK, 1992). However, the best existing heuristic for controlling this code growth—modulo variable expansion (Lam in SIGPLAN Not 23(7):318–328, 1988)—may not apply the correct amount of loop unrolling to guarantee that MAXLIVE registers are enough, which may result in register spills Eisenbeis et al. in PACT ’95: Proceedings of the IFIP WG10.3 working conference on Parallel Architectures and Compilation Techniques, pages 264–267, Manchester, UK, 1995. This paper presents our research results on the open problem of minimal loop unrolling, allowing a software-only code generation that does not trade the optimality of the initiation interval (II) for the compactness of the generated code. Our novel idea is to use the remaining free registers after periodic register allocation to relax the constraints on register reuse. The problem of minimal loop unrolling arises either before or after software pipelining, either with a single or with multiple register types (classes). We provide a formal problem definition for each scenario, and we propose and study a dedicated algorithm for each problem. Our solutions are implemented within an industrial-strength compiler for a VLIW embedded processor from STMicroelectronics, and validated on multiple benchmarks suites.

Software oriented techniques to hide memory latency in superscalar and superpipelined machines include loop unrolling, software pipelining, and software cache prefetching. Issuing the data fetch request prior to actual need for data allows overlap of accessing with useful computations. Loop unrolling and software pipelining do not necessitate microarchitecture or instruction set architecture changes, whereas software controlled prefetching does. While studies on the benefits of the individual techniques have been done, no study evaluates all of these techniques within a consistent framework. This paper attempts to remedy this by providing a comparative evaluation of the features and benefits of the techniques. Loop, unrolling and static scheduling of loads is seen to produce significant improvement in performance at lower latencies. Software pipelining is observed to be better than software controlled prefetching at lower latencies, but at higher latencies, software prefetching outperforms software pipelining. Aggressive prefetching beyond conditional branches can detrimentally affect performance by increasing the memory bandwidth requirements and bus traffic. >

A well-known code transformation for improving the run-time performance of a program is loop unrolling. The most obvious benefit of unrolling a loop is that the transformed loop usually requires fewer instruction executions than the original loop. The reduction in instruction executions comes from two sources: the number of branch instructions executed is reduced, and the control variable is modified fewer times. In addition, for architectures with features designed to exploit instruction-level parallelism, loop unrolling can expose greater levels of instruction-level parallelism. Loop unrolling is an effective code transformation often improving the execution performance of programs that spend much of their execution time in loops by 10 to 30 percent. Possibly because of the effectiveness of a simple application of loop unrolling, it has not been studied as extensively as other code improvements such as register allocation or common subexpression elimination. The result is that many compilers employ simplistic loop unrolling algorithms that miss many opportunities for improving run-time performance. This paper describes how aggressive loop unrolling is done in a retargetable optimizing compiler. Using a set of 32 benchmark programs, the effectiveness of this more aggressive approach to loop unrolling is evaluated. The results show that aggressive loop unrolling can yield additional performance increase of 10 to 20 percent over the simple, naive approaches employed by many production compilers.

Software pipelining is a powerful technique to expose fine-grain parallelism, but it results in variables staying alive across more than one kernel iteration. It requires periodic register allocation and is challenging for code generation: the lack of a reliable solution currently restricts the applicability of software pipelining. The classical software solution that does not alter the computation throughput consists in unrolling the loop a posteriori [11], [10]. However, the resulting unrolling degree is often unacceptable and may reach absurd levels. Alternatively, loop unrolling can be avoided thanks to software register renaming. This is achieved through the insertion of move operations, but this may increase the initiation interval (II) which nullifies the benefits of software pipelining. This article aims at tightly controling the post-pass loop unrolling necessary to generate code. We study the potential of live range splitting to reduce kernel loop unrolling, introducing additional move instructions without inscreasing the II. We provide a complete formalisation of the problem, an algorithm, and extensive experiments. Our algorithm yields low unrolling degrees in most cases - with no increase of the II.

This paper studies two compilation techniques for enhancing scalar performance in high-speed scientific processors: software pipelining and loop unrolling. We study the impact of the architecture (size of the register file) and of the hardware (size of instruction buffer) on the efficiency of loop unrolling. We also develop a methodology for classifying software pipelining techniques. For loop unrolling, a straightforward scheduling algorithm is shown to produce near-optimal results when not inhibited by recurrences or memory hazards. Our study indicates that the performance produced with a modified CRAY-1S scalar architecture and a code scheduler utilizing loop unrolling is comparable to the performance achieved by the CRAY-1S with a vector unit and the CFT vectorizing compiler. Finally, we show that the combination of loop unrolling and dynamic software pipelining, as implemented by a decoupled computer, substantially outperforms the vector CRAY-1S.

This paper studies two compilation techniques for enhancing scalar performance in high-speed scientific processors: software pipelining and loop unrolling. We study the impact of the architecture (size of the register file) and of the hardware (size of instruction buffer) on the efficiency of loop unrolling. We also develop a methodology for classifying software pipelining techniques. For loop unrolling, a straightforward scheduling algorithm is shown to produce near-optimal results when not inhibited by recurrences or memory hazards. Software pipelining requires less hardware but also achieves less speedup. Finally, we show that the performance produced with a modified CRAY-1S scalar architecture and a code scheduler utilizing loop unrolling is comparable to the performance achieved by the CRAY-1S with a vector unit and the CFT vectorizing compiler.

This paper studies two compilation techniques for enhancing scalar performance in high-speed scientific processors: software pipelining and loop unrolling. We study the impact of the architecture (size of the register file) and of the hardware (size of instruction buffer) on the efficiency of loop unrolling. We also develop a methodology for classifying software pipelining techniques. For loop unrolling, a straightforward scheduling algorithm is shown to produce near-optimal results when not inhibited by recurrences or memory hazards. Software pipelining requires less hardware but also achieves less speedup. Finally, we show that the performance produced with a modified CRAY-1S scalar architecture and a code scheduler utilizing loop unrolling is comparable to the performance achieved by the CRAY-1S with a vector unit and the CFT vectorizing compiler.

This paper studies two compilation techniques for enhancing scalar performance in high-speed scientific processors: software pipelining and loop unrolling. We study the impact of the architecture (size of the register file) and of the hardware (size of instruction buffer) on the efficiency of loop unrolling. We also develop a methodology for classifying software pipelining techniques. For loop unrolling, a straightforward scheduling algorithm is shown to produce near-optimal results when not inhibited by recurrences or memory hazards. Software pipelining requires less hardware but also achieves less speedup. Finally, we show that the performance produced with a modified CRAY-1S scalar architecture and a code scheduler utilizing loop unrolling is comparable to the performance achieved by the CRAY-1S with a vector unit and the CFT vectorizing compiler.

Code generation for complex integer loops within the context of a VLIW architecture, has to date, been handled by several disparate methodologies. We provide an empirical study to characterize what a typical complex integer loop is and propose a general solution that optimally modifies the key control dependencies in common integer loops. This single algorithm, integrates several software techniques (assuming key architectural features) in order to provide for varying degrees of nested complex control flow. A number of techniques, including loop peeling, loop unrolling, software pipelining, if-conversion, and procedure inlining are combined cohesively to make the best transformation decisions, for a typical integer loop, before scheduling and register allocation. Optimal fusion and distribution decisions are assumed.

Register allocation is a much studied problem. A particularly important context for optimizing register allocation is within loops, since a significant fraction of the execution time of programs is often inside loop code. A variety of algorithms have been proposed in the past for register allocation, but the complexity of the problem has resulted in a decoupling of several important aspects, including loop unrolling, register promotion, and instruction reordering. In this paper, we develop an approach to register allocation and promotion in a unified optimization framework that simultaneously considers the impact of loop unrolling and instruction scheduling. This is done via a novel instruction tiling approach where instructions within a loop are represented along one dimension and innermost loop iterations along the other dimension. By exploiting the regularity along the loop dimension, and imposing essential dependence based constraints on intra-tile execution order, the problem of optimizing register pressure is cast in a constraint programming formalism. Experimental results are provided from thousands of innermost loops extracted from the SPEC benchmarks, demonstrating improvements over the current state-of-the-art.

International audience

Recently loop unrolling has been shown in a new light from the superscalar architectural point of view. In this paper, we show that in addition to superscalar effect and scalar replacement effect, loop unrolling can hide memory latency, and that the combination of those effects improve the performance of loop unrolling. A major contribution of this paper is that the analysis is done symbolically and quantitatively. Although they have been known as major reasons that affect the performance of loop unrolling, no quantitative approach has not been tried. Our analysis can make clear the behaviour of superscalar functions and memory latency hiding in loop unrolling.

An iterative modulo scheduling is very important for compilers targeting high performance multi-issue digital signal processors. This is because these processors are often severely limited by idle state functional units and thus the reduced idle units can have a positively significant impact on their performance. However, complex instructions, which are used in most recent DSPs such as mac, usually increase data dependence complexity, and such complex dependencies that exist in signal processing applications often restrict modulo scheduling freedom and therefore, become a limiting factor of the iterative modulo scheduler. In this work, we propose a technique that efficiently reselects instructions of an application loop code considering dependence complexity, which directly resolve the dependence constraint. That is specifically featured for accelerating software pipelining performance by minimizing length of intrinsic cyclic dependencies. To take advantage of this feature, few existing compilers support a loop unrolling based dependence relaxing technique, but only use them for some limited cases. This is mainly because the loop unrolling typically occurs an overhead of huge code size increment, and the iterative modulo scheduling with relaxed dependence techniques for general cases is an NP-hard problem that necessitates complex assignments of registers and functional units. Our technique uses a heuristic to efficiently handle this problem in pre-stage of iterative modulo scheduling without loop unrolling.

Integrating register allocation and software pipelining of loops is an active research area. We focus on techniques that precondition the dependence graph before software pipelining in order to ensure that no register spill instructions are inserted by the register allocator in the software pipelined loop. If spilling is not necessary for the input code, preconditioning techniques insert dependence arcs so that the maximum register pressure MAXLIVE achieved by any loop schedule is below the number of available registers, without hurting the initiation interval if possible. When a solution exists, a spill-free software pipeline is guaranteed to exist. Existing preconditioning techniques consider one register type (register class) at a time [Deschinkel and Touati 2008]. In this article, we extend preconditioning techniques so that multiple register types are considered simultaneously. First, we generalize the existing theory of register pressure minimization for cyclic scheduling. Second, we implement our method inside the production compiler of the ST2xx VLIW family, and we demonstrate its efficiency on industry benchmarks (FFMPEG, MEDIABENCH, SPEC2000, SPEC2006). We demonstrate a high spill reduction rate without a significant initiation interval loss.