SCoP Detection

In the efforts to run Tadashi (Vatai et al., 2025) on real world applications we developed scop_detector (and the convenience wrapper scops_in_dir) to help identify potential apps and/or SCoPs within them.

SCoP detection in Tadashi

Tadashi extracts SCoPs from source files using PET (Verdoolaege & Grosser, 2012). The entry point of PET is the pet_transform_C_source() function, which processes each SCoP by executing a callback function it passed to pet_transform_C_source() as a parameter.

autodetect feature (in tadashi and in general)

scop_detecor similarly uses pet_transform_C_source(), but the callback, instead of transforming the scop, prints it as is with commented out scop/endscop pragmas (with an additional comment which indicates the index of the SCoP and makes it easier to locate the scops, like this

...
// #pragma scop // [N] //////////////////
<SCoP>
// #pragma endscop // [N] //////////////////

This allows the user to just uncomment the pragmas for Tadashi to fin the SCoPs. However, care must be take: if you uncomment the scop/endscop pragmas with index 0 and 2 (but leave SCoP with index 1 commented out), this will result in Tadashi skipping SCoP 1 and messing up the index of the third SCoP and SCoP with index 2 (in the comment generated by scop_detector) will have index 1 in Tadashi.

(setq display-line-numbers-type ‘visual)

References

Tadashi: Enabling AI-Based Automated Code Generation With Guaranteed Correctness

Emil Vatai, Aleksandr Drozd, Ivan R. Ivanov, and 3 more authors

Jun 2025

Comment: Submitted to SC25

Abs DOI

Frameworks and domain-specific languages for auto-generating code have traditionally depended on human experts to implement rigorous methods ensuring the legality of code transformations. Recently, machine learning (ML) has gained traction for generating code optimized for specific hardware targets. However, ML approaches-particularly black-box neural networks-offer no guarantees on the correctness or legality of the transformations they produce. To address this gap, we introduce Tadashi, an end-to-end system that leverages the polyhedral model to support researchers in curating datasets critical for ML-based code generation. Tadashi provides an end-to-end system capable of applying, verifying, and evaluating candidate transformations on polyhedral schedules with both reliability and practicality. We formally prove that Tadashi guarantees the legality of generated transformations, demonstrate its low runtime overhead, and showcase its broad applicability. Tadashi available at https://github.com/vatai/tadashi/.
Polyhedral Extraction Tool

Sven Verdoolaege and Tobias Grosser

Jun 2012

Abs

We present a new library for extracting a polyhedral model from C source. The library is based on clang, the LLVM C frontend, and isl, a library for manipulating quasi-affine sets and relations. The use of clang for parsing the C code brings advanced diagnostics and full support for C99. The use of isl allows for an easy construction and a powerful and compact representation of the polyhedral model. Besides allowing arbitrary piecewise quasi-affine index expressions and conditions, the library also supports some data dependent constructs and has special treatment for unsigned integers. The library has been successfully used to obtain polyhedral models for use in an equivalence checker, a tool for constructing polyhedral process networks, a parallelizer targeting GPUs and an interactive polyhedral environment.

Enjoy Reading This Article?

Here are some more articles you might like to read next: