Optimal Memory Allocation and Scheduling for DMA Data Transfers under the LET Paradigm
Event Type
Research Manuscript
Virtual Programs
Hosted in Virtual Platform
Time-Critical System Design
Embedded Systems
DescriptionThe Logical Execution Time (LET) paradigm in increasingly used to achieve predictable communications in multicore automotive applications. Direct Memory Access (DMA) engines can perform the data copies that are needed in LET with improved parallelism and reduced overheads. However, each DMA transfer requires contiguous memory areas, and the performance strongly depends on the allocation in memory of the variables.
This paper proposes a protocol to perform LET communications with a DMA overcoming the limitations of CPU-driven transfers. Furthermore, it presents an optimal memory allocation scheme using mixed-integer linear programming. Experimental results are reported to compare the performance of different approaches.