Many commercial processors now offer the possibility of extending their instruction set for a specific application - that is, to introduce customized functional units. There is a need to develop algorithms that decide automatically, from high-level application code, which operations are to be carried out in the customized extensions. A few algorithms exist but are severely limited in the type of operation clusters they can choose and hence reduce significantly the effectiveness of specialization. In this paper, we introduce a more general algorithm which selects maximal-speedup convex subgraphs of the application dataflow graph under fundamental microarchitectural constraints, and which improves significantly on the state of the art.

/pdf/automatic-application-specific-instruction-set-extensions-42n5dobbjd.pdf

Automatic application-specific instruction-set extensions under microarchitectural constraints

Application-specific extensions to the computational capabilities of a processor provide an efficient mechanism to meet the growing performance and power demands of embedded applications. Hardware, in the form of new function units (or co-processors), and the corresponding instructions, are added to a baseline processor to meet the critical computational demands of a target application. The central challenge with this approach is the large degree of human effort required to identify and create the custom hardware units, as well as porting the application to the extended processor. In this paper, we present the design of a system to automate the instruction set customization process. A dataflow graph design space exploration engine efficiently identifies profitable computation subgraphs from which to create custom hardware, without artificially constraining their size or shape. The system also contains a compiler subgraph matching framework that identifies opportunities to exploit and generalize the hardware to support more computation graphs. We demonstrate the effectiveness of this system across a range of application domains and study the applicability of the custom hardware across the domain.

/pdf/processor-acceleration-through-automated-instruction-set-3xpcfdl41h.pdf

Processor acceleration through automated instruction set customization

In embedded computing, cost, power, and performance constraints call for the design of specialized processors, rather than for the use of the existing off-the-shelf solutions. While the design of these application-specific CPUs could be tackled from scratch, a cheaper and more effective option is that of extending the existing processors and toolchains. Extensibility is indeed a feature now offered in real designs, e.g., by processors such as Tensilica Xtensa [T. R. Halfhill, Microprocess Rep., 2003], ARC ARCtangent [T. R. Halfhill, Microprocess Rep., 2000], STMicroelectronics ST200 [P. Faraboschi, G. Brown, J. A. Fisher, G. Desoli, and F. Homewood, Proc. 27th Annu. Int. Symp. Computer Architecture, 2000, p. 203], and MIPS CorExtend [T. R. Halfhill, Microprocess Rep., 2003]. While all these processors provide development environments with simulation capabilities for evaluating efficiently hand-crafted solutions, the tools to identify automatically the best processor configuration for a given application are less common. In particular, solutions to choose specialized instruction-set extensions (ISEs) have been investigated in the past years but are still seldom part of commercial toolchains. This paper provides a formal methodology and a set of algorithms that help address the problem. It proposes exact algorithms to derive optimal ISEs; exact identification of a single ISE is applicable to basic blocks of up to 1500 assembler-like instructions. This paper also introduces approximate methods that can process basic blocks of larger size. Results show that the described algorithms find solutions close to those that a designer would obtain by a detailed study of the application code. Both heuristic and exact algorithms find ISEs able to speed up unextended processors up to 5.0x. State-of-the-art comparisons show that the presented algorithms outperform existing ones by up to 2.6x

/pdf/exact-and-approximate-algorithms-for-the-extension-of-3kfoqw2pls.pdf

Exact and approximate algorithms for the extension of embedded processor instruction sets

This book provides broad and comprehensive coverage of the entire EDA flow. EDA/VLSI practitioners and researchers in need of fluency in an "adjacent" field will find this an invaluable reference to the basic EDA concepts, principles, data structures, algorithms, and architectures for the design, verification, and test of VLSI circuits. Anyone who needs to learn the concepts, principles, data structures, algorithms, and architectures of the EDA flow will benefit from this book.


Covers complete spectrum of the EDA flow, from ESL design modeling to logic/test synthesis, verification, physical design, and test - helps EDA newcomers to get "up-and-running" quickly 
Includes comprehensive coverage of EDA concepts, principles, data structures, algorithms, and architectures - helps all readers improve their VLSI design competence 
Contains latest advancements not yet available in other books, including Test compression, ESL design modeling, large-scale floorplanning, placement, routing, synthesis of clock and power/ground networks - helps readers to design/develop testable chips or products 
Includes industry best-practices wherever appropriate in most chapters - helps readers avoid costly mistakes



Table of Contents


Chapter 1: Introduction Chapter 2: Fundamentals of CMOS Design Chapter 3: Design for Testability Chapter 4: Fundamentals of Algorithms Chapter 5: Electronic System-Level Design and High-Level Synthesis Chapter 6: Logic Synthesis in a Nutshell Chapter 7: Test Synthesis Chapter 8: Logic and Circuit Simulation Chapter 9:?Functional Verification Chapter 10: Floorplanning Chapter 11: Placement Chapter 12: Global and Detailed Routing Chapter 13: Synthesis of Clock and Power/Ground Networks Chapter 14: Fault Simulation and Test Generation.

Electronic Design Automation: Synthesis, Verification, and Test

Extensible processors allow addition of application-specific custom instructions to the core instruction set architecture. However, it is computationally expensive to automatically select the optimal set of custom instructions. Therefore, heuristic techniques are often employed to quickly search the design space. In this paper, we present an efficient algorithm for exact enumeration of all possible candidate instructions given the dataflow graph (DFG) corresponding to a code fragment. Even though this is similar to the "subgraph enumeration" problem (which is exponential), we find that most subgraphs are not feasible candidates for various reasons. In fact, the number of candidates is quite small compared to the size of the DFG. Compared to previous approaches, our technique achieves orders of magnitude speedup in enumerating these candidate custom instructions for very large DFGs.

/pdf/scalable-custom-instructions-identification-for-instruction-2cbeqjida5.pdf

Scalable custom instructions identification for instruction-set extensible processors

This paper studies the use of a reconfigurable architecture platform for embedded control applications aimed at improving real time performance. The hw/sw codesign methodology from POLIS is used. It starts from high-level specifications, optimizes an intermediate model of computation (Extended Finite State Machines) and derives both hardware and software, based on performance constraints. We study a particular architecture platform, which consists of a general purpose processor core, augmented with a reconfigurable function unit and data-path to improve run time performance. A new mapping flow and algorithms to partition hardware and software are proposed to generate implementations that best utilize this architecture. Encouraging preliminary results are shown for automotive electronic control examples.

/pdf/hw-sw-partitioning-and-code-generation-of-embedded-control-nb9waa6t29.pdf

HW/SW partitioning and code generation of embedded control applications on a reconfigurable architecture platform

Early power analysis for systems-on-chip (SoC) is crucial for determining the appropriate packaging and cost. This early analysis commonly relies on evaluating power formulas for all cores for multiple configurations of voltage, frequency, technology and application parameters, which is a tedious and error-prone process. This work presents a methodology and algorithms for automating the power analysis of SoCs. Given the power state machines for individual cores, this work defines the product power state machine for the whole SoC and uses formal symbolic simulation algorithms for traversing and computing the minimum and maximum power dissipated by sets of power states in the SoC.

/pdf/state-based-power-analysis-for-systems-on-chip-3se9bg9c29.pdf

State-based power analysis for systems-on-chip

A program called MVSIS (Multi-Valued Sequential Interactive Synthesis) has been developed which optimizes multi-level multi-valued (MV) networks. We describe what such a network is and the capabilities contained in MVSIS. MVSIS is modeled after SIS (Sequential Interactive Synthesis), which synthesizes binary multi-level networks, but the logic network of MVSIS is such that all variables can be multi-valued, each with its own range. Included in MVSIS are almost all the technology-independent transformations of SIS for combinational and sequential logic synthesis as well as transformations specific to multi-valued nodes, such as "merge", "pair-decode" and "encode". MVSIS can read and write BLIF-MV and BLIF (Berkeley Logic Interchange Format) files which describe MV networks and binary networks respectively.

/pdf/optimization-of-multi-valued-multi-level-networks-1dl7ijabhn.pdf

Optimization of multi-valued multi-level networks

We address optimizing multi-valued (MV) logic functions in a multi-level combinational logic network. Each node in the network, called an MV-node, has multi-valued inputs and single multi-valued output. The notion of don't cares used in binary logic is generalized to multi-valued logic. It contains two types of flexibility: incomplete specification and non-determinism. We generalize the computation of observability don't cares for a multi-valued function node. Methods are given to compute (a) the maximum set of observability don't cares, and (b) the compatible set of observability don't cares for each MV-node. We give a recursive image computation to transform the don't cares into the space of local inputs of the node to be minimized. The methods are applied to some experimental multi-valued networks, and demonstrate reduction in the size of the tables that represent multi-valued logic functions.

/pdf/don-t-cares-and-multi-valued-logic-network-minimization-zq1fl4i1xi.pdf

Don't cares and multi-valued logic network minimization

This paper addresses the problem of automatic generation of implementation software from high-level functional specifications in the context of embedded system on chip designs. Software design complexity for embedded systems has increased so much that a high-level functional programming paradigm need to be adopted for formal verifiability, maintainability and short time-to-market. We propose a framework for efficiently generating implementation software from a synchronous state machine specification for embedded control systems. The framework is generic enough to allow hardware/software partition for a given architecture platform. It is demonstrated that the logic optimization and simulation techniques can be combined to produce fast execution code for such embedded systems. Specifically, we propose a framework for software synthesis from multi-valued logic, including fast evaluation of logic functions, and scheduling techniques for node execution. Experiments are performed to show the initial results of our algorithms in this framework.

/pdf/software-synthesis-from-synchronous-specifications-using-4o3xck32di.pdf

Yunjian Jiang

Papers

HW/SW partitioning and code generation of embedded control applications on a reconfigurable architecture platform

State-based power analysis for systems-on-chip

Optimization of multi-valued multi-level networks

Don't cares and multi-valued logic network minimization

Software synthesis from synchronous specifications using logic simulation techniques