Hyunuk Jung

Conversion of reference C code to dataflow model: H.264 encoder case study

Model-based design is widely accepted in developing complex embedded system under intense time-to-market pressure. While it promises improved design productivity, the main bottleneck lies not in the design methodology but in constructing the initial algorithm representation in the specified model. It is particularly true if a complicated multimedia application is given in the form of a sequential reference C code. In this paper we propose a systematic procedure for converting a sequential C code to a dataflow specification that has been widely used in many design environments for DSP systems. The proposed technique is successfully applied to H.264 encoder algorithm as a case study

Optimized RTL Code Generation from Coarse-Grain Dataflow Specification for Fast HW/SW Cosynthesis

This paper presents a new methodology of automatic RTL code generation from coarse-grain dataflow specification for fast HW/SW cosynthesis. A node in a coarse-grain dataflow specification represents a functional block such as FIR and DCT and an arc may deliver multiple data samples per block invocation, which complicates the problem and distinguishes it from behavioral synthesis problem. Given optimized HW library blocks for dataflow nodes, we aim to generate the RTL codes for the entire hardware system including glue logics such as buffer and MUX, and the central controller. In the proposed design methodology, a dataflow graph can be mapped to various hardware structures by changing the resource allocation and schedule information. It simplifies the management of the area/performance tradeoff in hardware design and widens the design space of hardware implementation of a dataflow graph. We also support Fractional Rate Dataflow (FRDF) specification for more efficient hardware implementation. To overcome the additional hardware area overhead in the synthesized architecture, we propose two techniques reducing buffer overhead. Through experiments with some real examples, the usefulness of the proposed technique is demonstrated.

A quantitative analysis of performance benefits of 3D die stacking on mobile and embedded SoC

3D stacked DRAM improves peak memory performance. However, its effective performance is often limited by the constraints of row-to-row activation delay (tRRD), four active bank window (tFAW), etc. In this paper, we present a quantitative analysis of the performance impact of such constraints. In order to resolve the problem, we propose balancing the budget of DRAM row activation across DRAM channels. In the proposed method, an inter-memory controller coordinator receives the current demand of row activation from memory controllers and re-distributes the budget to the memory controllers in order to improve DRAM performance. Experimental results show that sharing the budget of row activation between memory channels can give average 4.72% improvement in the utilization of 3D stacked DRAM.