Open Collections

Accelerating in-system debug of high-level synthesis generated circuits on field-programmable gate arrays using incremental compilation techniques

University of British Columbia

High-Level Synthesis (HLS) has emerged as a promising technology that allows designers to create a digital hardware circuit using a high-level language like C, allowing even software developers to obtain the benefits of hardware implementation. HLS will only be successful if it is accompanied by a suitable debug ecosystem. There are existing debugging methodologies based on software simulation, however, these are not suitable for finding bugs which occur only during the actual execution of the circuit. Recent efforts have presented in-system debug techniques which allow a designer to debug an implementation, running on a Field-Programmable Gate Array (FPGA) at its actual speed, in the context of the original source code. These techniques typically add instrumentation to store a history of all user variables in a design on-chip. To maximize the effectiveness of the limited on-chip memory and to simplify the debug instrumentation logic, it is desirable to store only selected user variables. Unfortunately, this may lead to multiple debug runs. In existing frameworks, changing the variables to be stored between runs changes the debug instrumentation circuitry. This requires a complete recompilation of the design before reprogramming it on an FPGA. In this thesis, we quantify the benefits of recording fewer variables and solve the problem of lengthy full compilations in each debug run using incremental compilation techniques present in the commercial FPGA CAD tools. We propose two promising debug flows that use this technology to reduce the debug turn-around time for an in-system debug framework. The first flow, in which the user circuit and instrumentation are co-optimized during compilation, gives the fastest debug clock speeds but suffers in user circuit performance once the debug instrumentation is removed. In the second flow, the optimization of the user circuit is sacrosanct. It is placed and routed first without having any constraints and the debug instrumentation is added later leading to the fastest user circuit clock speeds, but performance suffers slightly during debug. Using either flow, we achieve 40% reduction in debug turn-around times, on average.

Text

2017-08-10

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/62550

Electrical and Computer Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/