Rapid circuit-specific inlining tuning for FPGA high-level synthesis
Bibliographic record
Abstract
Assumptions about the underlying architecture of the target hardware is typically what dictates the behavior of compiler optimizations. Nevertheless, modern high-level synthesis (HLS) tools that target field-programmable gate arrays (FPGAs) are still using the same optimization passes that were developed and tuned for general purpose processors. This paper examines the effect of the inlining pass on HLS-generated hardware, focusing on the circuit area and clock cycles metrics. An iterative search method to create a custom inliner tailored to each benchmark for each specific metric is proposed and evaluated. The quality of the results generated is analyzed and the effect of the coefficients used for making the inline decisions are also separately investigated. Furthermore, a novel compiler cache is proposed, enabling the rapid evaluation of new inlining logic. Results show that a circuit-specific inliner is able to generate circuits with either 6% fewer LEs, 15% fewer clock cycles or 11% smaller LEs * clock cycle product when compared to LLVM's default approach. Moreover, our inliner achieved a speedup of 23x when compared to LLVM performing the same task without .the compiler cache.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".