Latent Idiom Recognition for a Minimalist Functional Array Language Using Equality Saturation

Accelerating programs is typically done by recognizing code idioms matching high-performance libraries or hardware interfaces. However, recognizing such idioms automatically is challenging. The idiom recognition machinery is difficult to write and requires expert knowledge. In addition, slight varia...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings / International Symposium on Code Generation and Optimization pp. 270 - 282
Main Authors: Van Der Cruysse, Jonathan, Dubach, Christophe
Format: Conference Proceeding
Language:English
Published: IEEE 02.03.2024
Subjects:
array programming
Codes
equality saturation
functional programming
Kernel
Libraries
Machinery
Optimization
pattern matching
Robustness
Semantics
ISSN:2643-2838
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Accelerating programs is typically done by recognizing code idioms matching high-performance libraries or hardware interfaces. However, recognizing such idioms automatically is challenging. The idiom recognition machinery is difficult to write and requires expert knowledge. In addition, slight variations in the input program might hide the idiom and defeat the recognizer. This paper advocates for the use of a minimalist functional array language supporting a small, but expressive, set of operators. The minimalist design leads to a tiny sets of rewrite rules, which encode the language semantics. Crucially, the same minimalist language is also used to encode idioms. This removes the need for hand-crafted analysis passes, or for having to learn a complex domain-specific language to define the idioms. Coupled with equality saturation, this approach is able to match the core functions from the BLAS and PyTorch libraries on a set of computational kernels. Compared to reference C kernel implementations, the approach produces a geometric mean speedup of 1.46× for C programs using BLAS, when generating such programs from the high-level minimalist language.
ISSN:2643-2838
DOI:10.1109/CGO57630.2024.10444879