2018 21st Euromicro Conference on Digital System Design (DSD) 2018
DOI: 10.1109/dsd.2018.00042
|View full text |Cite
|
Sign up to set email alerts
|

Embedded Operating System Optimization through Floating to Fixed Point Compiler Transformation

Abstract: Architectures targeted at embedded systems often have limited floating point computation capabilities, and in many cases do not provide any hardware support. In this work, we propose a self-contained compiler transformation pass implemented within LLVM to perform floating point to fixed point conversion. This pass is used to optimize the scheduler of the MIOSIX 1 embedded real-time operating system. We compare the proposed approach with the original floating point implementation, a handtuned fixed point one, a… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
12
0

Year Published

2019
2019
2023
2023

Publication Types

Select...
6
2
1

Relationship

3
6

Authors

Journals

citations
Cited by 15 publications
(12 citation statements)
references
References 13 publications
0
12
0
Order By: Relevance
“…Taking the Hi3559 series of chips as an example, the inner NNIE core calculates the fixed-point data much faster than the floating-point numbers. Therefore, the data need to complete the transformation from floating-point number to fixed-point number before entering the computer core [ 16 ].…”
Section: Core Acceleration Methodsmentioning
confidence: 99%
“…Taking the Hi3559 series of chips as an example, the inner NNIE core calculates the fixed-point data much faster than the floating-point numbers. Therefore, the data need to complete the transformation from floating-point number to fixed-point number before entering the computer core [ 16 ].…”
Section: Core Acceleration Methodsmentioning
confidence: 99%
“…Proposals have been made at the HW level to take advantage of inherent perceptual limitations, redundant data, or reduced precision input [15], as well as to reduce system costs or improve power efficiency [16]. At the same time, works on floating-point to fixed-point conversion tools [17], [18] allow us to trade-off the algorithm exactness for a more efficient implementation. Finally, new data types are emerging including BFloat16 [19] and Posit [20].…”
Section: Programming Models and Toolchainsmentioning
confidence: 99%
“…Due to the structure of the input dataset, it is not possible to meaningfully differentiate the input classes through requirements (1) through (4). More in detail, requirements (1) and (4) are always satisfied, as every single triangle intersection test is independent and of a predetermined constant size.…”
Section: Jmeintmentioning
confidence: 99%
“…taffo is a precision tuning assistant that has already been employed in the embedded systems [4] and in the approximate computing domain [8] to convert floating point code into fixed point. More specifically, the taffo toolchain supplies the code analysis, the code manipulation, and the code validation software components.…”
mentioning
confidence: 99%