A special-purpose compiler for look-up table and code generation for function evaluation

Zhang, Yuanrui; Deng, Lanping; Yedlapalli, Praveen; Muralidhara, Sai Prashanth; Zhao, Huijun; Kandemir, Mahmut; Chakrabarti, Chaitali; Pitsianis, Nikos P.; Sun, Xiaobai

doi:10.1109/date.2010.5456978

Cited by 2 publications

(1 citation statement)

References 16 publications

(27 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Currently, compiler-based math functions are utilized. Zhang et al proposed a special-purpose compiler that is capable of generating LUTs that use Taylor series interpolants based on accuracy and memory constraints [44]. Anand et al [45] proposed a domain-specific language (DSL) for libm functions tuned for Cell/B.E.…”

Section: A Mathematical Functionsmentioning

confidence: 99%

SIMD-Constrained Lookup Table for Accelerating Variable-Weighted Convolution on x86/64 CPUs

Naganawa,

Kamei,

Kanetaka

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Convolution is the inner product of the neighborhood signal and weights and plays a fundamental role in image processing; thus, acceleration of convolution is essential. Among convolutions, variable-weighted convolution is used in adaptive filters and edge-preserving smoothing to realize various applications. Some weights are replaced with lookup tables (LUTs) to accelerate these filters. LUT reference is a classical acceleration method. However, the difference between the growth rate in computing speed and memory I/O speed has limited the scope of utilization of LUT references. Speedup would be possible if registers could be used as LUTs, but their small size makes them difficult to utilize. Therefore, this study proposes a downsampling method to fit LUTs into SIMD registers, which are relatively large and an efficient reference method for register-LUTs. Experimental results show that the proposed method can reproduce an accuracy in PSNR of 65.52 (+25.11) dB, while a simple full-size LUT in the register size can only reproduce 40.41 dB. Using a wider register width, the PSNR was 78.63 (+38.22) dB with AVX-512 and 84.5 (+44.09) dB with bfloat16. The fastest proposed method was on average 4.82/3.72 times faster than direct vector computing, 2.99/3.10 times faster than vector addressing, and 3.79/7.80 times faster than scalar addressing on the AVX2/AVX-512 computers while exceeding the display limit of 60 dB for 8-bit displays. Taking into account these speed/accuracy trade-offs, the performance of the proposed method was superior. This paper shows that LUT references can be realized with small SIMD registers in convolution. The proposed method is expected to be extended to adaptive filters, convolutional neural networks, and other image processing applications by accelerating the approximation with this register-LUT. Our code is available at https://fukushimalab.github.io/registerLUT4conv/.

show abstract

Section: A Mathematical Functionsmentioning

confidence: 99%

SIMD-Constrained Lookup Table for Accelerating Variable-Weighted Convolution on x86/64 CPUs

Naganawa,

Kamei,

Kanetaka

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

Optimizing Expression Selection for Lookup Table Program Transformation

Wilcox

Strout

Bieman

2012

2012 IEEE 12th International Working Conference on Source Code Analysis and Manipulation

View full text Add to dashboard Cite

Abstract-Scientific programmers can speed up function evaluation by precomputing and storing function results in lookup table (LUTs), thereby replacing costly evaluation code with an inexpensive memory access. A code transform that replaces computation with LUT code can improve performance, however, accuracy is reduced because of error inherent in reconstructing values from LUT data. LUT transforms are commonly used to approximate expensive elementary functions. The current practice is for software developers to (1) manually identify expressions that can benefit from a LUT transform, (2) modify the code by hand to implement the LUT transform, and (3) run experiments to determine if the resulting error is within application requirements. This approach reduces productivity, obfuscates code, and limits programmer control over accuracy and performance. We propose source code analysis and program transformation to substantially automate the application of LUT transforms. Our approach uses a novel optimization algorithm that selects Pareto optimal sets of expressions that benefit most from LUT transformation, based on error and performance estimates. We demonstrate our methodology with the Mesa tool, which achieves speedups of 1.4-6.9× on scientific codes while managing introduced error. Our tool makes the programmer more productive and improves the chances of finding an effective solution.

show abstract

A special-purpose compiler for look-up table and code generation for function evaluation

Cited by 2 publications

References 16 publications

SIMD-Constrained Lookup Table for Accelerating Variable-Weighted Convolution on x86/64 CPUs

SIMD-Constrained Lookup Table for Accelerating Variable-Weighted Convolution on x86/64 CPUs

Optimizing Expression Selection for Lookup Table Program Transformation

Contact Info

Product

Resources

About