Optimal integrated code generation for VLIW architectures

Keßler, Christoph; Bednarski, Andrzej

doi:10.1002/cpe.1012

Cited by 10 publications

(4 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Optimization techniques. IP is the most widely used optimization technique with the exception of Scholz and Eckstein [85] using PBQP, Bashford and Leupers [9] using CP, and Kessler and Bednarski [56] using DP. For combinatorial instruction scheduling in isolation a broader set of techniques has been applied, including the use of special-purpose branch-and-bound techniques [24].…”

Section: Related Approachesmentioning

confidence: 99%

Combinatorial Register Allocation and Instruction Scheduling

Lozano

Carlsson

Blindell

et al. 2019

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

This paper introduces a combinatorial optimization approach to register allocation and instruction scheduling, two central compiler problems. Combinatorial optimization has the potential to solve these problems optimally and to exploit processor-specific features readily. Our approach is the first to leverage this potential in practice: it captures the complete set of program transformations used in state-of-the-art compilers, scales to mediumsized functions of up to 1000 instructions, and generates executable code. This level of practicality is reached by using constraint programming, a particularly suitable combinatorial optimization technique. Unison, the implementation of our approach, is open source, used in industry, and integrated with the LLVM toolchain.An extensive evaluation confirms that Unison generates better code than LLVM while scaling to mediumsized functions. The evaluation uses systematically selected benchmarks from MediaBench and SPEC CPU2006 and different processor architectures (Hexagon, ARM, MIPS). Mean estimated speedup ranges from 1.1% to 10% and mean code size reduction ranges from 1.3% to 3.8% for the different architectures. A significant part of this improvement is due to the integrated nature of the approach. Executing the generated code on Hexagon confirms that the estimated speedup results in actual speedup. Given a fixed time limit, Unison solves optimally functions of up to 946 instructions, nearly an order of magnitude larger than previous approaches.The results show that our combinatorial approach can be applied in practice to trade compilation time for code quality beyond the usual compiler optimization levels, identify improvement opportunities in heuristic algorithms, and fully exploit processor-specific features. This paper is partially based on preliminary work presented at the Principles and Practice of Constraint Programming (2012) [20]; Languages, Compilers, and Tools for Embedded Systems (2014) [21]; and Compiler Construction (2016) [22] conferences. Compared to the preliminary work, this paper is completely restructured and rewritten, completes the combinatorial model with rematerialization, proposes extensions to capture additional program transformations and processor-specific features, and contributes a more exhaustive evaluation. Additions to the evaluation include more benchmarks and processors, evidence of the fundamental benefit of the integrated approach, an in-depth study of scalability, and actual execution measurements.

show abstract

Section: Related Approachesmentioning

confidence: 99%

Combinatorial Register Allocation and Instruction Scheduling

Lozano

Carlsson

Blindell

et al. 2019

ACM Trans. Program. Lang. Syst.

View full text Add to dashboard Cite

show abstract

“…In previous work [8], we developed a dynamic programming approach and implemented it in our retargetable framework called OPTIMIST [9]. However, there may be further general problem solving strategies that could likewise be applied to the integrated code generation problem.…”

Section: Introductionmentioning

confidence: 99%

“…The remainder of this paper is organized as follows: After introducing some notation, we provide in Section 3 the ILP formulation for fully integrated code generation for VLIW processors. For a description of the DP approach of OPTIMIST, we refer to a recent article [8]. Section 4 evaluates the DP approach against the ILP approach, and draws some conclusions.…”

Section: Introductionmentioning

confidence: 99%

Optimal Integrated VLIW Code Generation with Integer Linear Programming

Bednarski

Keßler

2006

Euro-Par 2006 Parallel Processing

Self Cite

View full text Add to dashboard Cite

We give an Integer Linear Programming (ILP) solution that fully integrates all steps of code generation, i.e. instruction selection, register allocation and instruction scheduling, on the basic block level for VLIW processors.In earlier work, we contributed a dynamic programming (DP) based method for optimal integrated code generation, implemented in our retargetable code generator OPTIMIST. In this paper we give first results to evaluate and compare our ILP formulation with our DP method on a VLIW processor. We also demonstrate how to precondition the ILP model by a heuristic relaxation of the DP method to improve ILP optimization time.

show abstract

“…Therefore, compiling techniques are needed to translate high-level programming languages into VLIW assembly for STA. In the area of compiler, extensive work has been done, e.g., a detailed introduction of a dynamic programming method at basic blocks level is given by Kessler and Bednarski [13] to generate optimal VLIW assembly. Unfortunately, the compiled code by a large amount of existing compilers may be manifold larger and slower than hand crafted code.…”

Section: Code Generation For Sta Architecturementioning

confidence: 99%