WCET-aware data selection and allocation for scratchpad memory

Abstract

In embedded systems, SPM (scratchpad memory) is an attractive alternative to cache memory due to its lower energy consumption and higher predictability of program execution. This paper studies the problem of placing variables of a program into an SPM such that its WCET (worst-case execution time) is minimized. We propose an efficient dynamic approach that comprises two novel heuristics. The first heuristic iteratively selects a most beneficial variable as an SPM resident candidate based on its impact on the k longest paths of the program. The second heuristic incrementally allocates each SPM resident candidate to the SPM based on graph coloring and acyclic graph orientation. We have evaluated our approach by comparing with an ILP-based approach and a longestpath-based greedy approach using the eight benchmarks selected from Powerstone and Mälardalen WCET Benchmark suites under three different SPM configurations. Our approach achieves up to 21% and 43% improvements in WCET reduction over the ILPbased approach and the greedy approach, respectively.

Scratchpad Memory aware task scheduling with minimum number of preemptions on a single processor

“…The approaches proposed in [18,19] Next, we propose a new approach for determining the maxi mum size of the SPM for each task based on our previous work on allocating variables of a single task to SPM [23]. For each variable Vi, we define a benefit vector benefit( Vi) as follows.…”

“…For more details on selecting a most beneficial variable and allocating it to SPM, we refer to [23] .…”

Data Allocation for Embedded Systems with Hybrid On-Chip Scratchpad and Caches

“…In the past decade, many studies have been done to exploit optimizing the energy consumption ( [16], [18], [7], [8]) and performance ( [13], [15], [20], [6], [19]) by using SPM in embedded systems.…”

“…Some researches focused on worst case execution time with improved timing predictability of the application by using SPM ( [5], [15], [19]). Deverge et al [5] dynamically allocated both static and automatic data to SPM in order to reduce a single task's WCET.…”

“…Suhendra et al [15] proposed SPM allocation techniques for data memory to minimize a task's WCET. Wan et al [19] studied the problem of placing variables of a program into an SPM to minimize WCET. Verma et al [17] proposed Cache Aware Scratchpad Allocation (CASA) algorithm to use the SPM for storing instruction, they generated traces and then created the conflict graph to represent conflict cache misses.…”