High Level Synthesis (HLS) tools enable application domain experts to implement applications and algorithms on FPGAs. The majority of present FPGA applications is following a stream processing model which is almost entirely implemented statically and not exploiting the benefits enabled by partial reconfiguration. In this paper, we propose a generic approach for implementing and using partial reconfiguration through an HLS design flow for Maxeler platforms. Our flow extracts HLS generated HDL code from the Maxeler compilation process in order to implement a static FPGA infrastructure as well as run-time reconfigurable stream processing modules. As a distinct feature, our infrastructure can accommodate multiple partial modules in a pipeline daisy-chained manner, which aligns directly to Maxeler's dataflow programming paradigm. The benefits of the proposed flow are demonstrated by a case study of a dynamically reconfigurable video processing pipeline delivering 6.4GB/s throughput.

No abstract

Making full use of the capabilities of the FPGA as an accelerator is difficult for non hardware experts, especially if partial reconfiguration is to be employed. One of the issues that arise is to physically implement modules into bounding boxes of minimum size for improving fragmentation cost and reconfiguration time. In this paper we present a method which automates the modules designing step, fulfilling module resource requirements and architectural FPGA constraints. We present a case study that shows how our automatic module implementation flow can be used to generate run-time reconfigurable bitstreams that are suited for stitching together processing pipelines directly from a Maxeler MaxJ HLS specification. This takes into consideration design alternatives, fragmentation, and routing failure mitigation strategies. ModelOur goal is to build a design flow that can be used by non FPGA experts to take advantage of partial dynamic reconfiguration. For this, we assume that an expert must first design a

No abstract