arXiv subjects

Roland Leißa

Publications and source records attributed to Roland Leißa.

6 recordsLinked to original sources

arXiv2026

SSA without Dominance for Higher-Order Programs

Dominance is a fundamental concept in compilers based on static single assignment (SSA) form. It underpins a wide range of analyses and transformations and defines a core property of SSA: every use must be dominated by its definition. We argue that this reliance on dominance has become increasingly problematic -- both in terms of precision and applicability to modern higher-order languages. First, control flow overapproximates data flow, which makes dominance-based analyses inherently imprecise. Second, dominance is well-defined only for first-order control-flow graphs (CFGs). More critically, higher-order programs violate the assumptions underlying SSA and classic CFGs: without an explicit CFG, the very notion that all uses of a variable must be dominated by its definition loses meaning. We propose an alternative foundation based on free variables. In this view, $ϕ$-functions and function parameters directly express data dependencies, enabling analyses traditionally built on dominance while improving precision and naturally extending to higher-order programs. We further present an efficient technique for maintaining free-variable sets in a mutable intermediate representation (IR). For analyses requiring additional structure, we introduce the nesting tree -- a relaxed analogue of the dominator tree constructed from variable dependencies rather than control flow. Our benchmarks demonstrate that the algorithms and data structures presented in this paper scale log-linearly with program size in practice.

cs.PL↗

arXiv2024

MimIR: An Extensible and Type-Safe Intermediate Representation for the DSL Age

Traditional compilers, designed for optimizing low-level code, fall short when dealing with modern, computation-heavy applications like image processing, machine learning, or numerical simulations. Optimizations should understand the primitive operations of the specific application domain and thus happen on that level. Domain-specific languages (DSLs) fulfill these requirements. However, DSL compilers reinvent the wheel over and over again as standard optimizations, code generators, and general infrastructure & boilerplate code must be reimplemented for each DSL compiler. This paper presents MimIR, an extensible, higher-order intermediate representation. At its core, MimIR is a pure type system and, hence, a form of a typed lambda calculus. Developers can declare the signatures of new (domain-specific) operations, called "axioms". An axiom can be the declaration of a function, a type constructor, or any other entity with a possibly polymorphic, polytypic, and/or dependent type. This way, developers can extend MimIR at any low or high level and bundle them in a "plugin". Plugins extend the compiler and take care of optimizing and lowering the plugins' axioms. We show the expressiveness and effectiveness of MimIR in three case studies: Low-level plugins that operate at the same level of abstraction as LLVM, a regular-expression matching plugin, and plugins for linear algebra and automatic differentiation. We show that in all three studies, MimIR produces code that has state-of-the-art performance.

cs.PL↗

arXiv2022

AnySeq/GPU: A Novel Approach for Faster Sequence Alignment on GPUs

In recent years, the rapidly increasing number of reads produced by next-generation sequencing (NGS) technologies has driven the demand for efficient implementations of sequence alignments in bioinformatics. However, current state-of-the-art approaches are not able to leverage the massively parallel processing capabilities of modern GPUs with close-to-peak performance. We present AnySeq/GPU-a sequence alignment library that augments the AnySeq1 library with a novel approach for accelerating dynamic programming (DP) alignment on GPUs by minimizing memory accesses using warp shuffles and half-precision arithmetic. Our implementation is based on the AnyDSL compiler framework which allows for convenient zero-cost abstractions through guaranteed partial evaluation. We show that our approach achieves over 80% of the peak performance on both NVIDIA and AMD GPUs thereby outperforming the GPU-based alignment libraries AnySeq1, GASAL2, ADEPT, and NVBIO by a factor of at least 3.6 while achieving a median speedup of 19.2x over these tools across different alignment scenarios and sequence lengths when running on the same hardware. This leads to throughputs of up to 1.7 TCUPS (tera cell updates per second) on an NVIDIA GV100, up to 3.3 TCUPS with half-precision arithmetic on a single NVIDIA A100, and up to 3.8 TCUPS on an AMD MI100.

cs.DC↗

arXiv2021

FLOWER: A comprehensive dataflow compiler for high-level synthesis

FPGAs have found their way into data centers as accelerator cards, making reconfigurable computing more accessible for high-performance applications. At the same time, new high-level synthesis compilers like Xilinx Vitis and runtime libraries such as XRT attract software programmers into the reconfigurable domain. While software programmers are familiar with task-level and data-parallel programming, FPGAs often require different types of parallelism. For example, data-driven parallelism is mandatory to obtain satisfactory hardware designs for pipelined dataflow architectures. However, software programmers are often not acquainted with dataflow architectures - resulting in poor hardware designs. In this work we present FLOWER, a comprehensive compiler infrastructure that provides automatic canonical transformations for high-level synthesis from a domain-specific library. This allows programmers to focus on algorithm implementations rather than low-level optimizations for dataflow architectures. We show that FLOWER allows to synthesize efficient implementations for high-performance streaming applications targeting System-on-Chip and FPGA accelerator cards, in the context of image processing and computer vision.

cs.AR↗

arXiv2020

tinyMD: A Portable and Scalable Implementation for Pairwise Interactions Simulations

This paper investigates the suitability of the AnyDSL partial evaluation framework to implement tinyMD: an efficient, scalable, and portable simulation of pairwise interactions among particles. We compare tinyMD with the miniMD proxy application that scales very well on parallel supercomputers. We discuss the differences between both implementations and contrast miniMD's performance for single-node CPU and GPU targets, as well as its scalability on SuperMUC-NG and Piz Daint supercomputers. Additionaly, we demonstrate tinyMD's flexibility by coupling it with the waLBerla multi-physics framework. This allow us to execute tinyMD simulations using the load-balancing mechanism implemented in waLBerla.

cs.PF↗

arXiv2020

AnySeq: A High Performance Sequence Alignment Library based on Partial Evaluation

Sequence alignments are fundamental to bioinformatics which has resulted in a variety of optimized implementations. Unfortunately, the vast majority of them are hand-tuned and specific to certain architectures and execution models. This not only makes them challenging to understand and extend, but also difficult to port to other platforms. We present AnySeq - a novel library for computing different types of pairwise alignments of DNA sequences. Our approach combines high performance with an intuitively understandable implementation, which is achieved through the concept of partial evaluation. Using the AnyDSL compiler framework, AnySeq enables the compilation of algorithmic variants that are highly optimized for specific usage scenarios and hardware targets with a single, uniform codebase. The resulting domain-specific library thus allows the variation of alignment parameters (such as alignment type, scoring scheme, and traceback vs.~plain score) by simple function composition rather than metaprogramming techniques which are often hard to understand. Our implementation supports multithreading and SIMD vectorization on CPUs, CUDA-enabled GPUs, and FPGAs. AnySeq is at most 7% slower and in many cases faster (up to 12%) than state-of-the art manually optimized alignment libraries on CPUs (SeqAn) and on GPUs (NVBio).

cs.DC↗