Mahsa Vahidi and Alex Orailoglu. La Jolla CA of alternatives needs to be explored to obtain the

Size: px
Start display at page:

Download "Mahsa Vahidi and Alex Orailoglu. La Jolla CA of alternatives needs to be explored to obtain the"

Transcription

1 Metric-Based Transformations for Self Testable VLSI Designs with High Test Concurrency Mahsa Vahidi and Alex Orailoglu Department of Computer Science and Engineering University of California, San Diego La Jolla CA Abstract We propose an approach for improving the testability of a design under BIST methodology through behavioral restructuring. Our results show that the proposed transformations help reduce the number of required test sessions. 1 Introduction With the constant increase in chip sizes, access paths for testing chip modules have become increasingly scarce. Traditional test approaches are rapidly becoming major cost factors for high-frequency, submicron VLSI designs. Techniques, such as design for testability, make chips testable by placing test structures on parts of the circuit that are less accessible and hard to test, thus alleviating the impact of rapidly increasing test costs. It has been projected that the inclusion of design for testability techniques will increase from 25% to 90% during the next dozen years. This change will also result in a dramatic decrease in the number of required test vectors. While traditional test approaches capture increasingly a larger share of the overall design cost, traditional design approaches themselves have been having trouble in meeting the design challenges posed by large, high-density chips. Consequently, methodologies such as high-level synthesis have been instrumental in alleviating design costs by raising the abstraction level of design entry. High level synthesis addresses the design and implementation of digital circuits given a behavioral description. The function to be performed by the chip is given in a high level language, such as VHDL, along with a set of design constraints. The recent trend in moving from structural to system level design and the increasing importance of expediting the design and re-design cycles has driven an increased number of semiconductor companies to adopt synthesis as part of their design methodology. Anovel problem posed by the higher design entry levels necessitated by high level synthesis is the uniqueness of the specication. Frequently, a designer can express the same design specication in multiple behavioral forms, while a single behavioral description may map to multiple intermediate forms. This multiplicity of forms, even though semantically equivalent, This work was supported by the Semiconductor Research Corporation under contract number DJ-538. nonetheless, can lead to radically dierent implementations with current synthesis tools. Each of these representations has an associated, and frequently differing, time, area and test overhead. This rich set of alternatives needs to be explored to obtain the most suitable representation for the given design constraints. Traditionally, area and speed have been the most important constraints in the process of high level synthesis. However, due to the increased complexityof VLSI chips, design for testability constraints have recently become an important part of high level synthesis. As tradeo options increase, the evaluation of all possible alternatives becomes impossible without the use of an analysis tool. It is essential that methods for addressing tradeos, including test considerations, be addressed at the highest levels of design entry, where frequently the resultant impact of a design decision is the largest. In this paper we present a new approach onex- ploring test options at behavioral level. Our approach uses transformations to restructure behavioral design descriptions with the goal of improving the design's testability. Testability improvements in high level synthesis are aimed at reducing the time and area overhead associated with incorporating test structures. Test application time is a constraint that directly affects the production cost by lengthening the time required for testing. In this paper, we illustrate the ecacy of our transformational approach to test cost alleviation, by applying it to the problem of synthesis of self-testable designs. BIST is a design for testability approach wherein testability isachieved by performing test pattern generation and test result evaluation on chip. The specication restructurings are performed on an intermediate dataow graph format of the design which is transparent to the designers, but more suitable for performing algebraic transformations. Such transformational approaches are supported through testability metrics, previously developed, which capture the test time overhead measures, associated with BIST [11]. The metrics and the corresponding test synthesis model are described in the denition section. In the results section, we construct specic benchmark examples to illustrate the eect of transformations on test time issues in a conceptual manner. Furthermore, experimental results are obtained on standard benchmarks to show the coarse grain eects of restructuring on a design's testability.

2 2 Previous Research The recent advances in high level synthesis have provided an aid for designers to expedite the design cycle by automating the design process. Design constraints such as area and speed tradeos have been extensively studied in synthesis [5]. The increasing importance of generating testable designs has raised the need to incorporate testability constraints into high level synthesis. An overview of recent work in synthesis for testability can be found in [1]. Various metrics have been developed to help make focused decisions on performing testability improvements on a design. Thearling and Abraham have studied observability and controllability testability measures at the functional level [10]. Randomness and transparency metrics have helped explore the tradeos between test time and test hardware overhead at system level [3]. In [11], a set of metrics for measuring testability in terms of test time and concurrency are proposed. Transformations have been used to explore constraints such as register optimization [8], power minimization [2], and ecient resource utilization [9]. Transformations have also been applied in high level synthesis of fault-tolerant ASICs [7]. Dey proposes transformations for reducing the partial scan cost of test synthesis [4]. In this approach, test area minimization is achieved by using hot potato transformation techniques to minimize the number of scan registers needed to break CDFG loops and to avoid the formation of other types of loops during assignment. 3 Denitions 3.1 Testability Model The work in this paper intends to evaluate testability issues involved in incorporating Parallel BIST methodology. During BIST operations, the registers in a design are recongured as either linear feedback registers () or signature analysis registers (). We briey describe some features of our model in this section; a complete denition can be found in [6]. A test path is a collection of paths which are chosen in a manner to allow access to the ports of all modules. Two test paths can be performed in one test session as long as there is no conict between them. A conict can occur if the two paths share a register which is used as in one path and in the other. It can also happen when two sets of test vectors need to go through one at the same time. A conguration conict is illustrated in gure 1. In order to achieve high concurrency, the number of test sessions must be minimized. This can be done by reducing the number of conicts among test paths. A dataow graph (DFG) represents the data dependencies between operations, while a datapath contains the data dependencies between assigned modules. The test path will contain false paths when it is not a subgraph of the dataow graph. Our test model enables the possible use of false paths since the degree of freedom they provide in choosing test paths can result in reduced test time. Test controller optimization can be used to reduce this additional cost. R 1 R 2 R 3 R 4 TEST PATH 1 M 1 A 1 R 1 R 2 R 3 R 4 M 1 A 1 R 5 R 5 A 2 R 6 TEST PATH 2 TEST PATH 1 R 7 A 2 R 6 TEST PATH 2 Figure 1: conguration conict, Conict is eliminated due to more fan out at and the two test paths can be executed concurrently 3.2 Test Synthesis Model Research for synthesis of self-testable designs is still at early stages of development and consequently no methodology has as of yet been rmly established. In order to provide an appropriate context, we outline a specic test synthesis methodology in this section, for which the applicability of our test metrics [6] has been proven. Figure 2 illustrates the organization of this framework. The test synthesis framework con- Behavioral Description Dataflow Graph Generator Dataflow Graph Behavioral Restructuring DFG With Improved Testability Test Register Insertion Scheduling BIST Datapath Behavioral Synthesis RTL Datapath Test Scheduling Binding Structural Synthesis Figure 2: Test synthesis framework Test Path Definition BIST Test Plan sists of many modules which work towards producing a testable design and its BIST datapath and test plan. The module which performs Behavioral Restructuring is the topic of this paper and transforms chip level design into semantically equivalent forms which are more suitable for satisfying the given test constraints. Such restructuring is followed by the behavioral synthesis process which includes two modules. Dataow Scheduling and Binding determine the RTL interconnection structure of the design. Once the microarchitectural description has been dened, the test synthesis of the design is completed by performing structural synthesis. Structural synthesis includes three modules: Test Register Selection, Test Path Denition and Test Scheduling. 3.3 Metrics In this section, we briey review the testability metrics which form the basis of our restructuring techniques. These three metrics are dened as conict, coverage and correlation [11]. The metrics are formulated at the behavioral level and transition smoothly to all lower abstraction levels of synthesis. The behavioral formulation of the metrics is utilized in this work to address the needs of behavioral transformations. Since scheduling and

3 binding information is not available at these stages of synthesis, probability of common bindings (PCB) values model the probability that two operations will share a hardware unit in the nal datapath and are used as a basis for all the metrics. The concept behind the conict metric was shown in gure 1. Multiple occurrences of a component, within a concurrent test session, can introduce con- icts in component utilization during testing and result in a slower test application time. The conict metric is a measure of the amount of conict in a graph if all components were to be tested in one test session. The coverage metric represents test application time by looking at the design from a path-oriented perspective. Since coverage values are computed on the basis of a single test session assumption, high coverage values throughout the design denote few test sessions while low coverage values denote increased test sessions. While coverage and conict are aimed at solving the problems of test time minimization by minimizing the conicts among test paths, the correlation metric is developed to improve problem areas in the design which can cause increased test time due to a need for more test patterns. Correlation is caused when both inputs to a module are receiving random patterns originating from the same source and can either happen through false paths or be embedded in the structure of the ow graph. 4 Transformations The proposed transformations include associativity, distributivity, duplication, commutativity and strength reduction. The above transformations are divided into local and global categories based on their overall eect on the design's testability metrics. Global transformations such as associativity impact the mobility of a wide range of operations in the graph and can achieve higher optimization. Local transformations such as commutativity and distributivity follow the global transformations and are useful for ne grain improvements. They have the advantage of being simple and less computationally intensive. Transformations can frequently result in adverse eects on other design attributes. Detailed objective functions can be used to help resolve such conicts between design attributes. 4.1 Commutativity Commutativity is a local transformation which helps improve the conict and correlation metrics. This transformation is especially eective in cases where the problem lies in the interconnect structure of the dataow graph. We use the simple example of gure 3 to illustrate how commutativity reduces con- ict. Applying commutativity on node 1 eliminates the conguration conict which was caused by and registers. While two test sessions are needed to test the structure in gure 3, only one test session is needed for testing the structure in gure 3. Commutativity allows us to explore dierent interconnect options and nd the one which leads to the least amount of conict. 1 2 C-Step1 Commutativity 1 2 C-Step1 C-Step2 C-Step2 Test Path 1 Test Path Test Path 2 Figure 3: Applying commutativity on DFG reduces the number of test paths Commutativity helps improve correlation in the same fashion. This problem is eliminated by using commutativity toswap the correlating edges on one of the respective nodes. As a result, the correlating edges will be forced to go into the same input port in the resulting data path. It can be observed from gure 3 that commutativity reduces the interconnect and the number of required muxes in this example. Nonetheless, it is easy to observe that cases exist where the commutative transformation can lead to more complex interconnect. 4.2 Duplication Another transformation which helps improve both coverage and conict metrics is duplication. The amount ofcoverage that each node receives is dependent on its accessibility to the registers. This accessibility can be increased by performing duplication on low coverage nodes and providing them with more paths to reach the registers. Creation of these alternate paths can help increase the coverage of a node which has little or no access to an register. As a result, conict is also reduced on the modules which have a limited access to test paths. Duplication reduces correlation by increasing the length of the path from the source of correlation to the reconvergent fanout. Duplication can result in an increase in area overhead, especially for designs with high levels of hardware utilization. 4.3 Distributivity The distributivity transformation is used to eliminate reconvergent fanouts that are inherent in the structure of the owgraph. The eects of this transformation on correlation are illustrated in gure 4 where causes degradation of test data at. As can be seen in gure 4b, correlation is obviated as a result of the application of the distributive transformation. The fashion in which distributivity is applied for correlation problems decreases hardware requirements and area overhead. However, an additional clock cycle may be required, as shown in the above example.

4 A * * B Figure 4: Correlation is eliminated by applying distributivity 4.4 Associativity Associativity is a global transformation which affects the critical path of a design by changing the overall height of the dataow graph. As a result, the mobility and input paths of some operations will be altered in the dataow graph. Associativity is used to reduce the length of the paths with low coverage values and reduced access to registers. Applying associativity in this manner also reduces the folding of operations that result in correlation or conict problems. In addition, reducing the length of the critical paths of a design usually improves the performance of a design by sacricing its area. Associativity-induced owgraph restructurings help equalize the amount ofcoverage and conict in the graph and eliminate the problem areas which result in conicting test paths and increased number of test sessions. An example of how associativity can improve testability is shown in gures 6 and 7 and is described in the results section. 4.5 Strength Reduction Once operations of the same type are bound to the same hardware unit, their inputs can potentially compete for test path access and cause conict, coverage and correlation problems. Strength reduction can help improve these testability problems by changing the node operation types in a dataow graph and eliminating their folding probability. However, since this transformation can only be applied to certain structures (which typically happen infrequently) in design specication, its eect on testability improvement is not expected to be as signicant as the aforementioned transformations. 5 Results Our results are presented in two parts. We rst illustrate the eects of the proposed transformations by using small examples described in detail. This ne grain approach should make it easier for the reader to view restructuring eects. The examples at this level show how transformations aect the conict, coverage and correlation measures of a design. Furthermore, such detailed examples aord the opportunity tomo- tivate the benecial eects of such transformations on the actual testability of the projected designs, rather than on the metric values only. We follow this up with extensive results, cumulatively presented, applied on standard high-level synthesis benchmarks to show how our testability metrics are eected by the restructur- A * B ing process. These benchmarks include the dierential equation and the AR lter. In order to outline the eects of the proposed transformations, we derive results that show the improvements separately on each of the three proposed testability metrics. A greedy heuristic is employed in all three cases to derive results that outline the bene- ts of the proposed transformations. The transformations are guided by the criticality of these metric values and are selectively applied on the ow graph. Since these measures are a reection of test time, the transformations work towards producing an alternative design specication with less test time overhead. Global transformations are applied rst (since their far reaching eects would tend to inuence unpredicatably the improvements of the local transformations, otherwise) and local transformations follow for negrain improvements. This ordering of the transformations enables us to improve the areas in the graph which were either negatively aected by the global transformations or ignored by them. We start by rst discussing ne-grained design examples. Figure 5 illustrates the eect of applying commutativity tonode1 of the design. The metric values for gures 5 and 5 are shown in table 1. The folding of the multiply nodes causes the add and subtract operations to compete for the multiplier's input for access to registers. Reconvergent fanout probabilities for these nodes are due to the fact that even binding decisions can not prevent the multiply module from receiving the adder's output data from both input ports in the nal datapath as shown in gure 5. The metric values for the transformed design (table 1) show afavorable decrease in conict and correlation and a similarly favorable increase in coverage. In the transformed version, the add operations have a high probability of folding into the same physical module and they will not compete for test register access. Commutativity also places all the add nodes at the right input of the multiply nodes and eliminates the probability of reconvergent fanout at the output of the multipliers. *2 Correlation because of I8 I *3 <1 Conflict *2-1 1 I8 *3 3 2 Conflict and Correlation are improved I9 Commutativity Figure 5: Conict and correlation are reduced atthe multiply nodes after restructuring Figures 6 and 7 illustrate the eects of applying associativity. A conguration conict at the <1

5 Node Original Transformed Name DFG DFG Con/Cov / Cor Con/Cov / Cor /0.42/ /1.00/ /0.99/ /0.99/ /1.00/ /1.00/ /1.00/ /1.00/ /1.00/ /1.00/0.00 < /1.00/ /1.00/ /1.00/ /1.00/ /1.00/ /0.99/0.00 AVG 0.28/0.93/ /0.99/0.15 Table 1: Comparison of metrics for gure 5 inputs of the subtracter results in a minimum of two test sessions. The binding and scheduling steps in later stages of synthesis can not prevent this conict if only one subtracter module is to be utilized. Due to this conict, the ADD1 module can not be covered in the rst test path and will need a second test session to be added as shown in gure 6. Applying associativity helps eliminate this conict by altering the input paths of the subtracter and redirecting the output of the ADD1 module. The corresponding metric values are shown in table 2. As is to be expected, restructuring also helps reduce the number of required clock cycles while the area overhead remains the same. 1 R1 I 1 I 2 I 6 I 4 I 3 I /1 R R1 I 1 I 2 0 ADD1 I 6 I 4 I 3 I 7 Figure 7: Performing associativity eliminates con- ict, Datapath requires one test session natives becomes impossible without a computational model. In this section, we show the eects of our transformations on three standard benchmarks. Figure 8 shows the dierential equation dataow graph. After *X4 -X7 -X10 X3 <X6 O6 *X9 X11 1 ADD2 *X4 -X7 -X10 SUB 0 X3 <X6 O6 *X9 DIV 0 X / R1 0 0 ADD2 ADD1 Conflict Figure 6: DFG before restructuring contains con- ict, The corresponding datapath requires two test sessions denoted by the dashed and bold paths Node Original Transformed Name DFG DFG Con/Cov Con/Cov / / / / / / / / / /1.00 =1 0.00/ /1.00 AVG 0.33/ /1.00 Table 2: Comparison of metrics for gures 6 and 7 As we move to larger circuits, tradeo options increase drastically and evaluating all possible alter- SUB DIV Figure 8: Dierential Equation before restructuring, Conict is reduced after transformations applying transformations (gure 8), the metric values for all operations either improved or remained the same. After calculating the metrics for this graph, we can see that node X2 has the highest probability of conict in the graph. This is primarily due to the large number of nodes that can be folded onto X2 and having with a fanout of 1 as one of its input registers. Since does not source any other node, it has to be tested through the module that gets bound to X2 and therefore has a high probability of causing a conguration conict. This is illustrated in gure 9. Similarly, nodex5 has an input Fanout (3) (1) (4) (2) (3) (4) Figure 9: Possible data path for nodes X1 and X2 before restructuring, Same datapath after restructuring has much less conict which can only be tested through the multiplier module which X5 maps to. (1)

6 Figure 8 shows the same dataow graph after applying transformations to optimize for test application time. The amount of conict in the design is signicantly reduced by restructuring the graph. An example of this improvement is illustrated in gure 9 where the datapath resulting from folding X1 and X2 is shown before and after applying associativity. As we can see, the probability ofhaving a con- guration conict is eliminated at the left input port of the module and reduced at the right input. M2 Figure 10: Restructuring eliminates reconvergence Correlation was reduced on nodes X8 and X5 for similar reasons. Figure 10 illustrates the reconvergent fanout at the output of the module in one possible data path that is created by the folding of X8 and X1. The correlation of 's pseudorandom vectors at the input ports of can cause a degradation in test data results. Applying commutativity on node X8 eliminates the probability of creating a reconvergent test path through false paths. The resultant data path after applying commutativity is shown in gure 10. We utilize the AR lter to illustrate how restructuring techniques can help improve coverage throughout a dataow graph. The original dataow graph for the AR lter is shown in gure 11. Associativity was applied in order to improve the coverage in low coverage areas. Figure 11 shows the transformed graph. The coverage values for all highlighted nodes were improved after restructuring the graph. These improvements were done by shortening the paths which contain low coverage values in dierent parts of the graph. Low coverage nodes are the bottleneck operations in the design since their coverage is inversely related to the number of test sessions required. Equalizing the coverage values throughout the graph helps improve the overall accessibility of operations to test registers *3 *4 8 * *20 *21 *22 25 * M2 7 * M2 *2 *3 *4 8 6 * *20 *22 Figure 11: Restructuring increases coverage throughout the AR lter The results for conict and correlation improvements are summarized in tables 3 and 4 respectively. As we can see, the restructuring techniques have resulted in overall improved testability in both benchmarks. Benchmark Original Transformed DIFF EQUATION AR FILTER Table 3: The eects of restructuring on conict Benchmark Original Transformed DIFF EQUATION AR FILTER Table 4: The eects of restructuring on correlation 6 Conclusions We have presented an approach for improving the testability of a design at the behavioral level. A set of testability metrics are used to pinpoint testability problems with an emphasis on test application time and concurrency. Testability improvements were made by using transformations to restructure the behavioral representation and generate semantically equivalent designs with reduced test time overhead. It is essential, if high level synthesis is to achieve wide commercial acceptance, that design aids which help a designer traverse the rich space of semantically equivalent specications be provided, especially for design constraints such as testability which are consuming an ever increasing fraction of design costs. References [1] L. Avra and E.J. McCluskey. High-Level Synthesis of Testable Designs: An Overview of University Systems. In ITC, Test Synthesis Seminar, pages 1{8, [2] A.P. Chandrakasan, M. Potkonjak, J. Rabaey, and R.W. Broderson. HYPER-LP: A System for Power Minimization Using Architectural Transformations. In ICCAD, pages 300{303, [3] S. Chiu and C. A. Papachristou. A Design for Testability Scheme with Applications to Data Path Synthesis. In DAC, pages 271{277, [4] S. Dey and M. Potkonjak. Transforming Behavioral Specications To Facilitate Synthesis of Testable Designs. In ITC, pages 184{193, [5] D.D. Gajski, N.D. Dutt, A.C. Wu, and S.Y. Lin. High-Level Synthesis: Introduction to Chip and System Design. Kluwer Academic Publishers, [6] I. G. Harris and A. Orailoglu. SYNCBIST: SYNthesis for Concurrent Built-In Self-Testability. InICCD, pages 101{ 104, [7] R. Karri and A. Orailoglu. Transformation-Based High- Level Synthesis of Fault-Tolerant ASICS. In DAC, pages 662{665, [8] R. Karri and A. Orailoglu. Transformation-Based Register Optimization in High-Level Synthesis. In The Asilomar Conference on Signals, Systems, and Computers, pages 894{898, [9] M. Potkonjak and J. Rabaey. Optimizing Resource Utilization using Transformations. In ICCAD, pages 88{91, [10] K. Thearling and J. Abraham. An Easily Computed Functional Level Testability Measure. In ITC, pages 381{390, [11] M. Vahidi and A. Orailoglu. Testability Metrics for Synthesis of Self-Testable Designs and Eective Test Plans. In IEEE VLSI Test Symposium, pages 170{175, 1995.

Type T1: force false. Type T2: force true. Type T3: complement. Type T4: load

Type T1: force false. Type T2: force true. Type T3: complement. Type T4: load Testability Insertion in Behavioral Descriptions Frank F. Hsu Elizabeth M. Rudnick Janak H. Patel Center for Reliable & High-Performance Computing University of Illinois, Urbana, IL Abstract A new synthesis-for-testability

More information

Gate Level Fault Diagnosis in Scan-Based BIST

Gate Level Fault Diagnosis in Scan-Based BIST Gate Level Fault Diagnosis in Scan-Based BIST Ismet Bayraktaroglu Computer Science & Engineering Department University of California, San Diego La Jolla, CA 92093 ibayrakt@csucsdedu Alex Orailoglu Computer

More information

High-Level Test Synthesis. Tianruo Yang and Zebo Peng. Department of Computer and Information Science

High-Level Test Synthesis. Tianruo Yang and Zebo Peng. Department of Computer and Information Science An Ecient Algorithm to Integrate Scheduling and Allocation in High-Level Synthesis Tianruo Yang and Zebo Peng Department of Computer and Information Science Linkoping University, S-581 83, Linkoping, Sweden

More information

Incorporating the Controller Eects During Register Transfer Level. Synthesis. Champaka Ramachandran and Fadi J. Kurdahi

Incorporating the Controller Eects During Register Transfer Level. Synthesis. Champaka Ramachandran and Fadi J. Kurdahi Incorporating the Controller Eects During Register Transfer Level Synthesis Champaka Ramachandran and Fadi J. Kurdahi Department of Electrical & Computer Engineering, University of California, Irvine,

More information

COE 561 Digital System Design & Synthesis Introduction

COE 561 Digital System Design & Synthesis Introduction 1 COE 561 Digital System Design & Synthesis Introduction Dr. Aiman H. El-Maleh Computer Engineering Department King Fahd University of Petroleum & Minerals Outline Course Topics Microelectronics Design

More information

OUT. + * * + + * * + c1 c2. c4 c5 D2 OUT

OUT. + * * + + * * + c1 c2. c4 c5 D2  OUT Techniques for Functional Test Pattern Execution Inki Hong and Miodrag Potkonjak UCLA Computer Science Department Los Angeles, CA 90095-596 USA Abstract Functional debugging of application specic integrated

More information

A Controller Testability Analysis and Enhancement Technique

A Controller Testability Analysis and Enhancement Technique A Controller Testability Analysis and Enhancement Technique Xinli Gu Erik Larsson, Krzysztof Kuchinski and Zebo Peng Synopsys, Inc. Dept. of Computer and Information Science 700 E. Middlefield Road Linköping

More information

1 Introduction Data format converters (DFCs) are used to permute the data from one format to another in signal processing and image processing applica

1 Introduction Data format converters (DFCs) are used to permute the data from one format to another in signal processing and image processing applica A New Register Allocation Scheme for Low Power Data Format Converters Kala Srivatsan, Chaitali Chakrabarti Lori E. Lucke Department of Electrical Engineering Minnetronix, Inc. Arizona State University

More information

Achilles: A High-Level Synthesis System. for Asynchronous Circuits. Jordi Cortadella, Rosa M. Badia, Enric Pastor, and Abelardo Pardo y

Achilles: A High-Level Synthesis System. for Asynchronous Circuits. Jordi Cortadella, Rosa M. Badia, Enric Pastor, and Abelardo Pardo y Achilles: A High-Level Synthesis System for Asynchronous Circuits Jordi Cortadella, Rosa M. Badia, Enric Pastor, and Abelardo Pardo y Dept. of Computer Architecture Universitat Politecnica de Catalunya

More information

PPS : A Pipeline Path-based Scheduler. 46, Avenue Felix Viallet, Grenoble Cedex, France.

PPS : A Pipeline Path-based Scheduler. 46, Avenue Felix Viallet, Grenoble Cedex, France. : A Pipeline Path-based Scheduler Maher Rahmouni Ahmed A. Jerraya Laboratoire TIMA/lNPG,, Avenue Felix Viallet, 80 Grenoble Cedex, France Email:rahmouni@verdon.imag.fr Abstract This paper presents a scheduling

More information

Area. A max. f(t) f(t *) A min

Area. A max. f(t) f(t *) A min Abstract Toward a Practical Methodology for Completely Characterizing the Optimal Design Space One of the most compelling reasons for developing highlevel synthesis systems has been the desire to quickly

More information

PARAS: System-Level Concurrent Partitioning and Scheduling. University of Wisconsin. Madison, WI

PARAS: System-Level Concurrent Partitioning and Scheduling. University of Wisconsin. Madison, WI PARAS: System-Level Concurrent Partitioning and Scheduling Wing Hang Wong and Rajiv Jain Department of Electrical and Computer Engineering University of Wisconsin Madison, WI 53706 http://polya.ece.wisc.edu/~rajiv/home.html

More information

1488 IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, VOL. 16, NO. 12, DECEMBER 1997

1488 IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, VOL. 16, NO. 12, DECEMBER 1997 1488 IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, VOL. 16, NO. 12, DECEMBER 1997 Nonscan Design-for-Testability Techniques Using RT-Level Design Information Sujit Dey,

More information

Submitted for TAU97 Abstract Many attempts have been made to combine some form of retiming with combinational

Submitted for TAU97 Abstract Many attempts have been made to combine some form of retiming with combinational Experiments in the Iterative Application of Resynthesis and Retiming Soha Hassoun and Carl Ebeling Department of Computer Science and Engineering University ofwashington, Seattle, WA fsoha,ebelingg@cs.washington.edu

More information

Basic Block. Inputs. K input. N outputs. I inputs MUX. Clock. Input Multiplexors

Basic Block. Inputs. K input. N outputs. I inputs MUX. Clock. Input Multiplexors RPack: Rability-Driven packing for cluster-based FPGAs E. Bozorgzadeh S. Ogrenci-Memik M. Sarrafzadeh Computer Science Department Department ofece Computer Science Department UCLA Northwestern University

More information

Datapath Allocation. Zoltan Baruch. Computer Science Department, Technical University of Cluj-Napoca

Datapath Allocation. Zoltan Baruch. Computer Science Department, Technical University of Cluj-Napoca Datapath Allocation Zoltan Baruch Computer Science Department, Technical University of Cluj-Napoca e-mail: baruch@utcluj.ro Abstract. The datapath allocation is one of the basic operations executed in

More information

FILTER SYNTHESIS USING FINE-GRAIN DATA-FLOW GRAPHS. Waqas Akram, Cirrus Logic Inc., Austin, Texas

FILTER SYNTHESIS USING FINE-GRAIN DATA-FLOW GRAPHS. Waqas Akram, Cirrus Logic Inc., Austin, Texas FILTER SYNTHESIS USING FINE-GRAIN DATA-FLOW GRAPHS Waqas Akram, Cirrus Logic Inc., Austin, Texas Abstract: This project is concerned with finding ways to synthesize hardware-efficient digital filters given

More information

7 DESIGN FOR TESTABILITY II: FROM HIGH LEVEL PERSPECTIVE

7 DESIGN FOR TESTABILITY II: FROM HIGH LEVEL PERSPECTIVE 122 Advances in Microelectronics 7 DESIGN FOR TESTABILITY II: FROM HIGH LEVEL PERSPETIVE hia Yee Ooi Norlina Paraman 7.1 ONTEXT The advantage of a top-down design flow, specifying design a high abstraction

More information

An Algorithm for the Allocation of Functional Units from. Realistic RT Component Libraries. Department of Information and Computer Science

An Algorithm for the Allocation of Functional Units from. Realistic RT Component Libraries. Department of Information and Computer Science An Algorithm for the Allocation of Functional Units from Realistic RT Component Libraries Roger Ang rang@ics.uci.edu Nikil Dutt dutt@ics.uci.edu Department of Information and Computer Science University

More information

Design and Synthesis for Test

Design and Synthesis for Test TDTS 80 Lecture 6 Design and Synthesis for Test Zebo Peng Embedded Systems Laboratory IDA, Linköping University Testing and its Current Practice To meet user s quality requirements. Testing aims at the

More information

Self Test Built-in Plan for Data-Path Functional Units

Self Test Built-in Plan for Data-Path Functional Units Self Test Built-in Plan for Data-Path Functional Units J. A. Quilici Gonzalez*, José Roberto de A. Amazonas**, Marius Strum***, Wang Jiang Chau**** Laboratório de Microeletrônica da EPUSP Av. Luciano Gualberto

More information

Area. f(t) f(t *) A min. T min. T T * T max Latency

Area. f(t) f(t *) A min. T min. T T * T max Latency Ecient Optimal Design Space Characterization Methodologies STEPHEN A. BLYTHE Rensselaer Polytechnic Institute and ROBERT A. WALKER Kent State University 1 One of the primary advantages of a high-level

More information

High-level Variable Selection for Partial-Scan Implementation

High-level Variable Selection for Partial-Scan Implementation High-level Variable Selection for Partial-Scan Implementation FrankF.Hsu JanakH.Patel Center for Reliable & High-Performance Computing University of Illinois, Urbana, IL Abstract In this paper, we propose

More information

A taxonomy of race. D. P. Helmbold, C. E. McDowell. September 28, University of California, Santa Cruz. Santa Cruz, CA

A taxonomy of race. D. P. Helmbold, C. E. McDowell. September 28, University of California, Santa Cruz. Santa Cruz, CA A taxonomy of race conditions. D. P. Helmbold, C. E. McDowell UCSC-CRL-94-34 September 28, 1994 Board of Studies in Computer and Information Sciences University of California, Santa Cruz Santa Cruz, CA

More information

Hardware Design Environments. Dr. Mahdi Abbasi Computer Engineering Department Bu-Ali Sina University

Hardware Design Environments. Dr. Mahdi Abbasi Computer Engineering Department Bu-Ali Sina University Hardware Design Environments Dr. Mahdi Abbasi Computer Engineering Department Bu-Ali Sina University Outline Welcome to COE 405 Digital System Design Design Domains and Levels of Abstractions Synthesis

More information

B2 if cs < cs_max then cs := cs + 1 cs := 1 ra

B2 if cs < cs_max then cs := cs + 1 cs := 1 ra Register Transfer Level VHDL Models without Clocks Matthias Mutz (MMutz@sican{bs.de) SICAN Braunschweig GmbH, Digital IC Center D{38106 Braunschweig, GERMANY Abstract Several hardware compilers on the

More information

RTL Scan Design for Skewed-Load At-Speed Test under Power Constraints

RTL Scan Design for Skewed-Load At-Speed Test under Power Constraints RTL Scan Design for Skewed-Load At-Speed Test under Power Constraints Ho Fai Ko and Nicola Nicolici Department of Electrical and Computer Engineering McMaster University, Hamilton, ON, L8S 4K1, Canada

More information

Behavioral Array Mapping into Multiport Memories Targeting Low Power 3

Behavioral Array Mapping into Multiport Memories Targeting Low Power 3 Behavioral Array Mapping into Multiport Memories Targeting Low Power 3 Preeti Ranjan Panda and Nikil D. Dutt Department of Information and Computer Science University of California, Irvine, CA 92697-3425,

More information

(RC) utilize CAD tools to perform the technology mapping of a extensive amount of time is spent for compilation by the CAD

(RC) utilize CAD tools to perform the technology mapping of a extensive amount of time is spent for compilation by the CAD Domain Specic Mapping for Solving Graph Problems on Recongurable Devices? Andreas Dandalis, Alessandro Mei??, and Viktor K. Prasanna University of Southern California fdandalis, prasanna, ameig@halcyon.usc.edu

More information

Exploiting Off-Line Hierarchical Test Paths in Module Diagnosis and On-Line Test

Exploiting Off-Line Hierarchical Test Paths in Module Diagnosis and On-Line Test Exploiting Off-Line Hierarchical Paths in Diagnosis and On-Line lu Reliable Systems Synthesis Lab Computer Science & Engineering Department University of California San Diego 9500 Gilman Drive MC-0114

More information

Conclusions and Future Work. We introduce a new method for dealing with the shortage of quality benchmark circuits

Conclusions and Future Work. We introduce a new method for dealing with the shortage of quality benchmark circuits Chapter 7 Conclusions and Future Work 7.1 Thesis Summary. In this thesis we make new inroads into the understanding of digital circuits as graphs. We introduce a new method for dealing with the shortage

More information

Design Procedure Based on VHDL Language Transformations

Design Procedure Based on VHDL Language Transformations VLSI Design, 2002 Vol. 14 (4), pp. 349 354 Design Procedure Based on VHDL Language Transformations LÁSZLÓ VARGA, GÁBOR HOSSZÚ* and FERENC KOVÁCS Department of Electron Devices, Technical University of

More information

Abstract A SCALABLE, PARALLEL, AND RECONFIGURABLE DATAPATH ARCHITECTURE

Abstract A SCALABLE, PARALLEL, AND RECONFIGURABLE DATAPATH ARCHITECTURE A SCALABLE, PARALLEL, AND RECONFIGURABLE DATAPATH ARCHITECTURE Reiner W. Hartenstein, Rainer Kress, Helmut Reinig University of Kaiserslautern Erwin-Schrödinger-Straße, D-67663 Kaiserslautern, Germany

More information

Network. Department of Statistics. University of California, Berkeley. January, Abstract

Network. Department of Statistics. University of California, Berkeley. January, Abstract Parallelizing CART Using a Workstation Network Phil Spector Leo Breiman Department of Statistics University of California, Berkeley January, 1995 Abstract The CART (Classication and Regression Trees) program,

More information

scription of the circuit using a high-level design language (HDL) such as VHDL or Verilog. The behavior of this description is checked using simulatio

scription of the circuit using a high-level design language (HDL) such as VHDL or Verilog. The behavior of this description is checked using simulatio BooleDozer : Logic synthesis for ASICs L. Stok D. S. Kung D. Brand A. D. Drumm A. J. Sullivan L. N. Reddy N. Hieter D. J. Geiger H. Chao P. J. Osler February 6, 1996 Logic synthesis is the process of automatically

More information

100-hour Design Cycle: A Test Case. keeping the specication close to the conceptualization. 2. Use of standard languages for input specications.

100-hour Design Cycle: A Test Case. keeping the specication close to the conceptualization. 2. Use of standard languages for input specications. 100-hour Design Cycle: A Test Case Daniel D. Gajski, Loganath Ramachandran, Peter Fung 3, Sanjiv Narayan 1 and Frank Vahid 2 University of California, Irvine, CA 3 Matsushita Electric Works, Research and

More information

Compilation Issues for High Performance Computers: A Comparative. Overview of a General Model and the Unied Model. Brian J.

Compilation Issues for High Performance Computers: A Comparative. Overview of a General Model and the Unied Model. Brian J. Compilation Issues for High Performance Computers: A Comparative Overview of a General Model and the Unied Model Abstract This paper presents a comparison of two models suitable for use in a compiler for

More information

Effects of Technology Mapping on Fault Detection Coverage in Reprogrammable FPGAs

Effects of Technology Mapping on Fault Detection Coverage in Reprogrammable FPGAs Syracuse University SURFACE Electrical Engineering and Computer Science College of Engineering and Computer Science 1995 Effects of Technology Mapping on Fault Detection Coverage in Reprogrammable FPGAs

More information

SPARK: A Parallelizing High-Level Synthesis Framework

SPARK: A Parallelizing High-Level Synthesis Framework SPARK: A Parallelizing High-Level Synthesis Framework Sumit Gupta Rajesh Gupta, Nikil Dutt, Alex Nicolau Center for Embedded Computer Systems University of California, Irvine and San Diego http://www.cecs.uci.edu/~spark

More information

Managing Dynamic Reconfiguration Overhead in Systems-on-a-Chip Design Using Reconfigurable Datapaths and Optimized Interconnection Networks

Managing Dynamic Reconfiguration Overhead in Systems-on-a-Chip Design Using Reconfigurable Datapaths and Optimized Interconnection Networks Managing Dynamic Reconfiguration Overhead in Systems-on-a-Chip Design Using Reconfigurable Datapaths and Optimized Interconnection Networks Zhining Huang, Sharad Malik Electrical Engineering Department

More information

Cycle-accurate RTL Modeling with Multi-Cycled and Pipelined Components

Cycle-accurate RTL Modeling with Multi-Cycled and Pipelined Components Cycle-accurate RTL Modeling with Multi-Cycled and Pipelined Components Rainer Dömer, Andreas Gerstlauer, Dongwan Shin Technical Report CECS-04-19 July 22, 2004 Center for Embedded Computer Systems University

More information

Solve the Data Flow Problem

Solve the Data Flow Problem Gaining Condence in Distributed Systems Gleb Naumovich, Lori A. Clarke, and Leon J. Osterweil University of Massachusetts, Amherst Computer Science Department University of Massachusetts Amherst, Massachusetts

More information

A Module Diagnosis and Design-for-Debug Methodology Based on Hierarchical Test Paths

A Module Diagnosis and Design-for-Debug Methodology Based on Hierarchical Test Paths A Diagnosis and Design-for-Debug Methodology ased on Hierarchical Test s

More information

Dierential-Linear Cryptanalysis of Serpent? Haifa 32000, Israel. Haifa 32000, Israel

Dierential-Linear Cryptanalysis of Serpent? Haifa 32000, Israel. Haifa 32000, Israel Dierential-Linear Cryptanalysis of Serpent Eli Biham, 1 Orr Dunkelman, 1 Nathan Keller 2 1 Computer Science Department, Technion. Haifa 32000, Israel fbiham,orrdg@cs.technion.ac.il 2 Mathematics Department,

More information

ED&TC 97 on CD-ROM Permission to make digital/hard copy of part or all of this work for personal or classroom use is granted without fee provided

ED&TC 97 on CD-ROM Permission to make digital/hard copy of part or all of this work for personal or classroom use is granted without fee provided Accurate High Level Datapath Power Estimation James E. Crenshaw and Majid Sarrafzadeh Department of Electrical and Computer Engineering Northwestern University, Evanston, IL 60208 Abstract The cubic switching

More information

Consistent Logical Checkpointing. Nitin H. Vaidya. Texas A&M University. Phone: Fax:

Consistent Logical Checkpointing. Nitin H. Vaidya. Texas A&M University. Phone: Fax: Consistent Logical Checkpointing Nitin H. Vaidya Department of Computer Science Texas A&M University College Station, TX 77843-3112 hone: 409-845-0512 Fax: 409-847-8578 E-mail: vaidya@cs.tamu.edu Technical

More information

Exploring Logic Block Granularity for Regular Fabrics

Exploring Logic Block Granularity for Regular Fabrics 1530-1591/04 $20.00 (c) 2004 IEEE Exploring Logic Block Granularity for Regular Fabrics A. Koorapaty, V. Kheterpal, P. Gopalakrishnan, M. Fu, L. Pileggi {aneeshk, vkheterp, pgopalak, mfu, pileggi}@ece.cmu.edu

More information

A New Optimal State Assignment Technique for Partial Scan Designs

A New Optimal State Assignment Technique for Partial Scan Designs A New Optimal State Assignment Technique for Partial Scan Designs Sungju Park, Saeyang Yang and Sangwook Cho The state assignment for a finite state machine greatly affects the delay, area, and testabilities

More information

160 M. Nadjarbashi, S.M. Fakhraie and A. Kaviani Figure 2. LUTB structure. each block-level track can be arbitrarily connected to each of 16 4-LUT inp

160 M. Nadjarbashi, S.M. Fakhraie and A. Kaviani Figure 2. LUTB structure. each block-level track can be arbitrarily connected to each of 16 4-LUT inp Scientia Iranica, Vol. 11, No. 3, pp 159{164 c Sharif University of Technology, July 2004 On Routing Architecture for Hybrid FPGA M. Nadjarbashi, S.M. Fakhraie 1 and A. Kaviani 2 In this paper, the routing

More information

Combining MBP-Speculative Computation and Loop Pipelining. in High-Level Synthesis. Technical University of Braunschweig. Braunschweig, Germany

Combining MBP-Speculative Computation and Loop Pipelining. in High-Level Synthesis. Technical University of Braunschweig. Braunschweig, Germany Combining MBP-Speculative Computation and Loop Pipelining in High-Level Synthesis U. Holtmann, R. Ernst Technical University of Braunschweig Braunschweig, Germany Abstract Frequent control dependencies

More information

Coordinated Resource Optimization in Behavioral Synthesis

Coordinated Resource Optimization in Behavioral Synthesis Coordinated Resource Optimization in Behavioral Synthesis Jason Cong Bin Liu Junjuan Xu Computer Science Department, University of California, Los Angeles Email: {cong, bliu, irene.xu}@cs.ucla.edu Abstract

More information

Computer Technology Institute. Patras, Greece. In this paper we present a user{friendly framework and a

Computer Technology Institute. Patras, Greece. In this paper we present a user{friendly framework and a MEASURING SOFTWARE COMPLEXITY USING SOFTWARE METRICS 1 2 Xenos M., Tsalidis C., Christodoulakis D. Computer Technology Institute Patras, Greece In this paper we present a user{friendly framework and a

More information

A tabu search based memetic algorithm for the max-mean dispersion problem

A tabu search based memetic algorithm for the max-mean dispersion problem A tabu search based memetic algorithm for the max-mean dispersion problem Xiangjing Lai a and Jin-Kao Hao a,b, a LERIA, Université d'angers, 2 Bd Lavoisier, 49045 Angers, France b Institut Universitaire

More information

Digital Design Methodology (Revisited) Design Methodology: Big Picture

Digital Design Methodology (Revisited) Design Methodology: Big Picture Digital Design Methodology (Revisited) Design Methodology Design Specification Verification Synthesis Technology Options Full Custom VLSI Standard Cell ASIC FPGA CS 150 Fall 2005 - Lec #25 Design Methodology

More information

INTERFACE SYNTHESIS. process A. process A1 variable MEM : intarray ;. process A2. process A1 variable MEM : intarray ; procedure send_ch1(...

INTERFACE SYNTHESIS. process A. process A1 variable MEM : intarray ;. process A2. process A1 variable MEM : intarray ; procedure send_ch1(... Protocol Generation for Communication Channels y Sanjiv Narayan Daniel D. Gajski Viewlogic Systems Inc. Dept. of Computer Science Marlboro, MA 01752 Univ. of California, Irvine, CA 92717 Abstract System-level

More information

a) Generic Representation of Multilevel Optimization

a) Generic Representation of Multilevel Optimization Chapter 6 OPTIMIZATION FOR RECONFIGURABLE SYSTEMS USING HIERARCHICALABSTRACTION Elaheh Bozorgzadeh UCLA Computer Science Department elib@cs.ucla.edu Adam Kaplan UCLA Computer Science Department kaplan@cs.ucla.edu

More information

Dept. of Computer Science, Keio University. Dept. of Information and Computer Science, Kanagawa Institute of Technology

Dept. of Computer Science, Keio University. Dept. of Information and Computer Science, Kanagawa Institute of Technology HOSMII: A Virtual Hardware Integrated with Yuichiro Shibata, 1 Hidenori Miyazaki, 1 Xiao-ping Ling, 2 and Hideharu Amano 1 1 Dept. of Computer Science, Keio University 2 Dept. of Information and Computer

More information

International Journal of Foundations of Computer Science c World Scientic Publishing Company DFT TECHNIQUES FOR SIZE ESTIMATION OF DATABASE JOIN OPERA

International Journal of Foundations of Computer Science c World Scientic Publishing Company DFT TECHNIQUES FOR SIZE ESTIMATION OF DATABASE JOIN OPERA International Journal of Foundations of Computer Science c World Scientic Publishing Company DFT TECHNIQUES FOR SIZE ESTIMATION OF DATABASE JOIN OPERATIONS KAM_IL SARAC, OMER E GEC_IO GLU, AMR EL ABBADI

More information

Don't Cares in Multi-Level Network Optimization. Hamid Savoj. Abstract

Don't Cares in Multi-Level Network Optimization. Hamid Savoj. Abstract Don't Cares in Multi-Level Network Optimization Hamid Savoj University of California Berkeley, California Department of Electrical Engineering and Computer Sciences Abstract An important factor in the

More information

CHAPTER 1 INTRODUCTION

CHAPTER 1 INTRODUCTION CHAPTER 1 INTRODUCTION Rapid advances in integrated circuit technology have made it possible to fabricate digital circuits with large number of devices on a single chip. The advantages of integrated circuits

More information

Implementing Logic in FPGA Memory Arrays: Heterogeneous Memory Architectures

Implementing Logic in FPGA Memory Arrays: Heterogeneous Memory Architectures Implementing Logic in FPGA Memory Arrays: Heterogeneous Memory Architectures Steven J.E. Wilton Department of Electrical and Computer Engineering University of British Columbia Vancouver, BC, Canada, V6T

More information

Digital Design Methodology

Digital Design Methodology Digital Design Methodology Prof. Soo-Ik Chae Digital System Designs and Practices Using Verilog HDL and FPGAs @ 2008, John Wiley 1-1 Digital Design Methodology (Added) Design Methodology Design Specification

More information

Abstract. provide substantial improvements in performance on a per application basis. We have used architectural customization

Abstract. provide substantial improvements in performance on a per application basis. We have used architectural customization Architectural Adaptation in MORPH Rajesh K. Gupta a Andrew Chien b a Information and Computer Science, University of California, Irvine, CA 92697. b Computer Science and Engg., University of California,

More information

Addressing Verification Bottlenecks of Fully Synthesized Processor Cores using Equivalence Checkers

Addressing Verification Bottlenecks of Fully Synthesized Processor Cores using Equivalence Checkers Addressing Verification Bottlenecks of Fully Synthesized Processor Cores using Equivalence Checkers Subash Chandar G (g-chandar1@ti.com), Vaideeswaran S (vaidee@ti.com) DSP Design, Texas Instruments India

More information

Document Image Restoration Using Binary Morphological Filters. Jisheng Liang, Robert M. Haralick. Seattle, Washington Ihsin T.

Document Image Restoration Using Binary Morphological Filters. Jisheng Liang, Robert M. Haralick. Seattle, Washington Ihsin T. Document Image Restoration Using Binary Morphological Filters Jisheng Liang, Robert M. Haralick University of Washington, Department of Electrical Engineering Seattle, Washington 98195 Ihsin T. Phillips

More information

Lecture 9 - Matrix Multiplication Equivalences and Spectral Graph Theory 1

Lecture 9 - Matrix Multiplication Equivalences and Spectral Graph Theory 1 CME 305: Discrete Mathematics and Algorithms Instructor: Professor Aaron Sidford (sidford@stanfordedu) February 6, 2018 Lecture 9 - Matrix Multiplication Equivalences and Spectral Graph Theory 1 In the

More information

is developed which describe the mean values of various system parameters. These equations have circular dependencies and must be solved iteratively. T

is developed which describe the mean values of various system parameters. These equations have circular dependencies and must be solved iteratively. T A Mean Value Analysis Multiprocessor Model Incorporating Superscalar Processors and Latency Tolerating Techniques 1 David H. Albonesi Israel Koren Department of Electrical and Computer Engineering University

More information

INTERNATIONAL JOURNAL OF PROFESSIONAL ENGINEERING STUDIES Volume VII /Issue 2 / OCT 2016

INTERNATIONAL JOURNAL OF PROFESSIONAL ENGINEERING STUDIES Volume VII /Issue 2 / OCT 2016 NEW VLSI ARCHITECTURE FOR EXPLOITING CARRY- SAVE ARITHMETIC USING VERILOG HDL B.Anusha 1 Ch.Ramesh 2 shivajeehul@gmail.com 1 chintala12271@rediffmail.com 2 1 PG Scholar, Dept of ECE, Ganapathy Engineering

More information

PARALLEL PERFORMANCE DIRECTED TECHNOLOGY MAPPING FOR FPGA. Laurent Lemarchand. Informatique. ea 2215, D pt. ubo University{ bp 809

PARALLEL PERFORMANCE DIRECTED TECHNOLOGY MAPPING FOR FPGA. Laurent Lemarchand. Informatique. ea 2215, D pt. ubo University{ bp 809 PARALLEL PERFORMANCE DIRECTED TECHNOLOGY MAPPING FOR FPGA Laurent Lemarchand Informatique ubo University{ bp 809 f-29285, Brest { France lemarch@univ-brest.fr ea 2215, D pt ABSTRACT An ecient distributed

More information

On Test Generation for Transition Faults with Minimized Peak Power Dissipation

On Test Generation for Transition Faults with Minimized Peak Power Dissipation 30.3 On Test Generation for Transition Faults with Minimized Peak Power Dissipation Wei Li Sudhakar M. Reddy Irith Pomeranz 2 Dept. of ECE School of ECE Univ. of Iowa Purdue University Iowa City, IA 52242

More information

reasonable to store in a software implementation, it is likely to be a signicant burden in a low-cost hardware implementation. We describe in this pap

reasonable to store in a software implementation, it is likely to be a signicant burden in a low-cost hardware implementation. We describe in this pap Storage-Ecient Finite Field Basis Conversion Burton S. Kaliski Jr. 1 and Yiqun Lisa Yin 2 RSA Laboratories 1 20 Crosby Drive, Bedford, MA 01730. burt@rsa.com 2 2955 Campus Drive, San Mateo, CA 94402. yiqun@rsa.com

More information

6 DESIGN FOR TESTABILITY I: FROM FULL SCAN TO PARTIAL SCAN

6 DESIGN FOR TESTABILITY I: FROM FULL SCAN TO PARTIAL SCAN 94 Advances in Microelectronics 6 DESIGN FOR TESTABILITY I: FROM FULL SCAN TO PARTIAL SCAN Chia Yee Ooi 6.1 CONTEXT It is important to check whether the manufactured circuit has physical defects or not.

More information

Eect of fan-out on the Performance of a. Single-message cancellation scheme. Atul Prakash (Contact Author) Gwo-baw Wu. Seema Jetli

Eect of fan-out on the Performance of a. Single-message cancellation scheme. Atul Prakash (Contact Author) Gwo-baw Wu. Seema Jetli Eect of fan-out on the Performance of a Single-message cancellation scheme Atul Prakash (Contact Author) Gwo-baw Wu Seema Jetli Department of Electrical Engineering and Computer Science University of Michigan,

More information

Architectural Design and Analysis of a VLIW Processor. Arthur Abnous and Nader Bagherzadeh. Department of Electrical and Computer Engineering

Architectural Design and Analysis of a VLIW Processor. Arthur Abnous and Nader Bagherzadeh. Department of Electrical and Computer Engineering Architectural Design and Analysis of a VLIW Processor Arthur Abnous and Nader Bagherzadeh Department of Electrical and Computer Engineering University of California, Irvine Irvine, CA 92717 Phone: (714)

More information

IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, VOL. 18, NO. 9, SEPTEMBER

IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, VOL. 18, NO. 9, SEPTEMBER IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, VOL. 18, NO. 9, SEPTEMBER 1999 1327 Testability Analysis and Test-Point Insertion in RTL VHDL Specifications for Scan-Based

More information

Area. f(t) f(t *) A min. T min. T T* T max Latency

Area. f(t) f(t *) A min. T min. T T* T max Latency This material is based upon work supported by the National Science Foundation under Grant No. MIP-9423953 while the authors were with the Department of Computer Science at Rensselaer Polytechnic Institute.

More information

Self-checking combination and sequential networks design

Self-checking combination and sequential networks design Self-checking combination and sequential networks design Tatjana Nikolić Faculty of Electronic Engineering Nis, Serbia Outline Introduction Reliable systems Concurrent error detection Self-checking logic

More information

Egemen Tanin, Tahsin M. Kurc, Cevdet Aykanat, Bulent Ozguc. Abstract. Direct Volume Rendering (DVR) is a powerful technique for

Egemen Tanin, Tahsin M. Kurc, Cevdet Aykanat, Bulent Ozguc. Abstract. Direct Volume Rendering (DVR) is a powerful technique for Comparison of Two Image-Space Subdivision Algorithms for Direct Volume Rendering on Distributed-Memory Multicomputers Egemen Tanin, Tahsin M. Kurc, Cevdet Aykanat, Bulent Ozguc Dept. of Computer Eng. and

More information

Unit 2: High-Level Synthesis

Unit 2: High-Level Synthesis Course contents Unit 2: High-Level Synthesis Hardware modeling Data flow Scheduling/allocation/assignment Reading Chapter 11 Unit 2 1 High-Level Synthesis (HLS) Hardware-description language (HDL) synthesis

More information

Experience in Critical Path Selection For Deep Sub-Micron Delay Test and Timing Validation

Experience in Critical Path Selection For Deep Sub-Micron Delay Test and Timing Validation Experience in Critical Path Selection For Deep Sub-Micron Delay Test and Timing Validation Jing-Jia Liou, Li-C. Wang, Angela Krstic, and Kwang-Ting Cheng University of California at Santa Barbara Abstract

More information

Outline. Computer Science 331. Information Hiding. What This Lecture is About. Data Structures, Abstract Data Types, and Their Implementations

Outline. Computer Science 331. Information Hiding. What This Lecture is About. Data Structures, Abstract Data Types, and Their Implementations Outline Computer Science 331 Data Structures, Abstract Data Types, and Their Implementations Mike Jacobson 1 Overview 2 ADTs as Interfaces Department of Computer Science University of Calgary Lecture #8

More information

Software Synthesis. using Timed Decision Tables. Sumit Gupta. Rajesh Gupta. January Abstract

Software Synthesis. using Timed Decision Tables. Sumit Gupta. Rajesh Gupta.  January Abstract . Software Synthesis using Timed Decision Tables Sumit Gupta Rajesh Gupta http://www.ics.uci.edu/iesag Technical Report #99-01 Department of Information and Computer Science University of California, Irvine,

More information

Power Efficient Register Assignment

Power Efficient Register Assignment Power Efficient Register Assignment Jui-Ming Chang and Massoud Pedram CENG Technical Report 95-03 Department of Electrical Engineering - Systems University of Southern Los Angeles, California 90089-2562

More information

Hot Potato Techniques in High Level Synthesis

Hot Potato Techniques in High Level Synthesis Hot Potato Techniques in High Level Synthesis Miodrag Potkonjak Sujit Dey C&C Research Laboratories, NEC USA, 4 Independence Way, Princeton, NJ 08540 ABSTRACT This paper introduces hot potato high level

More information

AMS Behavioral Modeling

AMS Behavioral Modeling CHAPTER 3 AMS Behavioral Modeling Ronald S. Vogelsong, Ph.D. Overview Analog designers have for many decades developed their design using a Bottom-Up design flow. First, they would gain the necessary understanding

More information

Testing Embedded Cores Using Partial Isolation Rings

Testing Embedded Cores Using Partial Isolation Rings Testing Embedded Cores Using Partial Isolation Rings Nur A. Touba and Bahram Pouya Computer Engineering Research Center Department of Electrical and Computer Engineering University of Texas, Austin, TX

More information

Efficient Static Timing Analysis Using a Unified Framework for False Paths and Multi-Cycle Paths

Efficient Static Timing Analysis Using a Unified Framework for False Paths and Multi-Cycle Paths Efficient Static Timing Analysis Using a Unified Framework for False Paths and Multi-Cycle Paths Shuo Zhou, Bo Yao, Hongyu Chen, Yi Zhu and Chung-Kuan Cheng University of California at San Diego La Jolla,

More information

signature i-1 signature i instruction j j+1 branch adjustment value "if - path" initial value signature i signature j instruction exit signature j+1

signature i-1 signature i instruction j j+1 branch adjustment value if - path initial value signature i signature j instruction exit signature j+1 CONTROL FLOW MONITORING FOR A TIME-TRIGGERED COMMUNICATION CONTROLLER Thomas M. Galla 1, Michael Sprachmann 2, Andreas Steininger 1 and Christopher Temple 1 Abstract A novel control ow monitoring scheme

More information

High Level Synthesis

High Level Synthesis High Level Synthesis Design Representation Intermediate representation essential for efficient processing. Input HDL behavioral descriptions translated into some canonical intermediate representation.

More information

Digital VLSI Testing Prof. Santanu Chattopadhyay Department of Electronics and EC Engineering India Institute of Technology, Kharagpur.

Digital VLSI Testing Prof. Santanu Chattopadhyay Department of Electronics and EC Engineering India Institute of Technology, Kharagpur. Digital VLSI Testing Prof. Santanu Chattopadhyay Department of Electronics and EC Engineering India Institute of Technology, Kharagpur Lecture 05 DFT Next we will look into the topic design for testability,

More information

On the Use of Fuzzy Techniques for Partial Scan Insertion Based on the Testability metrics

On the Use of Fuzzy Techniques for Partial Scan Insertion Based on the Testability metrics On the Use of Fuzzy Techniques for Partial Scan Insertion Based on the Testability metrics Caro Lucas,Naghmeh Karimi Electrical and Computer Engineering University of Tehran Abstract: The chief goal of

More information

CIM: Component Isolation and Monitoring for System-Level Verification

CIM: Component Isolation and Monitoring for System-Level Verification C: Component Isolation and Monitoring for System-Level Verification Glenn Freytag and Ravi Shankar Center for Systems Integration, Florida Atlantic University, Boca Raton, FL {glenn, ravi}@cse.fau.edu

More information

Selecting PLLs for ASIC Applications Requires Tradeoffs

Selecting PLLs for ASIC Applications Requires Tradeoffs Selecting PLLs for ASIC Applications Requires Tradeoffs John G. Maneatis, Ph.., President, True Circuits, Inc. Los Altos, California October 7, 2004 Phase-Locked Loops (PLLs) are commonly used to perform

More information

EECS 579: Built-in Self-Test 3. Regular Circuits

EECS 579: Built-in Self-Test 3. Regular Circuits EECS 579: Built-in Self-Test 3 Outline Implementing BIST by regularization Adder ALU RAM Commercial BIST approaches LOCSD STUMPS CSTP Case Study Bosch AE11 microcontroller John P. Hayes University of Michigan

More information

University of California, Berkeley. Midterm II. You are allowed to use a calculator and one 8.5" x 1" double-sided page of notes.

University of California, Berkeley. Midterm II. You are allowed to use a calculator and one 8.5 x 1 double-sided page of notes. University of California, Berkeley College of Engineering Computer Science Division EECS Fall 1997 D.A. Patterson Midterm II October 19, 1997 CS152 Computer Architecture and Engineering You are allowed

More information

CREATIVE ASSERTION AND CONSTRAINT METHODS FOR FORMAL DESIGN VERIFICATION

CREATIVE ASSERTION AND CONSTRAINT METHODS FOR FORMAL DESIGN VERIFICATION CREATIVE ASSERTION AND CONSTRAINT METHODS FOR FORMAL DESIGN VERIFICATION Joseph Richards SGI, High Performance Systems Development Mountain View, CA richards@sgi.com Abstract The challenges involved in

More information

Modeling Arbitrator Delay-Area Dependencies in Customizable Instruction Set Processors

Modeling Arbitrator Delay-Area Dependencies in Customizable Instruction Set Processors Modeling Arbitrator Delay-Area Dependencies in Customizable Instruction Set Processors Siew-Kei Lam Centre for High Performance Embedded Systems, Nanyang Technological University, Singapore (assklam@ntu.edu.sg)

More information

Chapter 14 Performance and Processor Design

Chapter 14 Performance and Processor Design Chapter 14 Performance and Processor Design Outline 14.1 Introduction 14.2 Important Trends Affecting Performance Issues 14.3 Why Performance Monitoring and Evaluation are Needed 14.4 Performance Measures

More information

Research on outlier intrusion detection technologybased on data mining

Research on outlier intrusion detection technologybased on data mining Acta Technica 62 (2017), No. 4A, 635640 c 2017 Institute of Thermomechanics CAS, v.v.i. Research on outlier intrusion detection technologybased on data mining Liang zhu 1, 2 Abstract. With the rapid development

More information

An Evaluation of Information Retrieval Accuracy. with Simulated OCR Output. K. Taghva z, and J. Borsack z. University of Massachusetts, Amherst

An Evaluation of Information Retrieval Accuracy. with Simulated OCR Output. K. Taghva z, and J. Borsack z. University of Massachusetts, Amherst An Evaluation of Information Retrieval Accuracy with Simulated OCR Output W.B. Croft y, S.M. Harding y, K. Taghva z, and J. Borsack z y Computer Science Department University of Massachusetts, Amherst

More information