Xilinx - Design Reuse Methodology for ASIC and FPGA Designers.pdf

More documents

Recommendations

Info

2Table of Contents1 Introduction ......................................................................................... 31.1 System-on-a-Reprogrammable Chip ....................................................................31.2 Why Use an FPGA?.............................................................................................41.2.1 ASIC vs. FPGA Design Flows .....................................................................41.3 A Common Design Reuse Strategy ......................................................................62 System Level Reuse Issues for FPGAs ............................................... 81.4 Definitions & Acronyms......................................................................................72.1 System Synthesis and Timing Issues ....................................................................82.1.1 Synchronous vs. Asynchronous Design Style ...............................................82.1.3 System Clocking and Clock Distribution......................................................92.2 Memory and Memory Interface..........................................................................122.2.1 On-Chip Memory..........................................................................................122.2.2 Interfacing to Large Memory Blocks..........................................................133 Coding and Synthesis Tips................................................................ 162.3 External Operability (I/O Standards)..................................................................143.1 Abundance of Registers .....................................................................................163.1.1 Duplicating Registers .................................................................................163.1.2 Partitioning at Register Boundary...............................................................183.1.3 One-Hot State Machines.............................................................................183.1.4 Pipelining...................................................................................................183.2 Case and IF-Then-Else.......................................................................................203.3 Critical Path Optimization..................................................................................233.4 Tristate vs. Mux Buses.......................................................................................244 Verification Strategy ......................................................................... 263.5 Arithmetic Functions..........................................................................................244.1 HDL Simulation and Testbench .........................................................................264.2 Static Timing .....................................................................................................264.3 Formal Verification............................................................................................27
31 IntroductionFPGAs have changed dramatically since Xilinx first introduced them just 15 years ago. In thepast, FPGA were primarily used for prototyping and lower volume applications; custom ASICswere used for high volume, cost sensitive designs. FPGAs had also been too expensive and tooslow for many applications, let alone for System Level Integration (SLI). Plus, the developmenttools were often difficult to learn and lacked the features found in ASIC development systems.Now, this has all changed.Silicon technology has progressed to allow chips with tens of millions of transistors. This notonly promises new levels of integration onto a single chip, but also allows more features andcapabilities in reprogrammable technology. With today’s deep sub-micron technology, it ispossible to deliver over 2 - million usable system gates in a FPGA. In addition, the averageASIC design operating at 30 – 50MHz can be implemented in a FPGA using the same RTLsynthesis design methodology as ASICs. By the year 2004, the state-of-the-art FPGA will exceed10 million system gates, allowing for multimillion gates FPGAs operating at speeds surpassing300 MHz. Many designs, which previously could only achieve speed and cost-of-density goals inASICs, are converting to much more flexible and productive reprogrammable solutions.The availability of FPGAs in the 1-million system gate range has started a shift of SoC designstowards using Reprogrammable FPGAs, thereby starting a new era of System-on-a-Reprogrammable-Chip (SoRC). For a 1-million system gate SoRC design, an engineer designing100 gates/day would require a hypothetical 42 years to complete, at a cost of $6 million. Clearly,immense productivity gains are needed to make million gate designs commercially viable, andSoRC based on today’s million logic gate FPGAs, and tomorrow’s 10-million logic gate FPGAsis a promising solution.SoRC is no different from SoC in that it requires leveraging existing intellectual property (IP) toimprove designer productivity. Reusable IP is essential to constructing bug-free multimilliongatedesigns in a reasonable amount of time. Without reuse, the electronics industry will simplynot be able to keep pace with the challenge of delivering the “better, faster, cheaper” devices thatconsumers expect.With the availability of new FPGA architectures designed for system level integration and FPGAdesign tools that are compatible with ASIC design methodologies, it is now possible to employ asimilar if not identical design reuse methodology for ASICs and FPGAs. For design teams thathave longed to eliminate NRE costs and improve time-to-market without climbing the learningcurve to become FPGA experts, this design reuse methodology adds a new level of freedom.However, this methodology is not without its limitations. This paper examines the designs thatcan take advantage of an identical ASIC and FPGA design reuse methodology and the simplemodifications that can be made to the RTL code to enhance performance and reuse.1.1 System-on-a-Reprogrammable ChipTo define SoRC, let us first start with a general definition of System-on-a-Chip (SoC). Most ofthe industry agrees that SoC is the incorporation of an entire system onto one chip. Dataquest’s1995 definition included a compute engine (microprocessor, microcontroller or digital signalprocessor), at least 100K of user gates and significant on-chip memory.
Page 1: Xilinx Design Reuse Methodologyfor
Page 5 and 6: 5ASIC Design FlowFPGA Design FlowSp
Page 7 and 8: 7effectively on SoRC designs and by
Page 9 and 10: 9XSI page 2-28 - X4975Figure 2 - D
Page 11 and 12: 11Neither a PLL nor a DLL can be us
Page 13 and 14: 13RAM can be used for status regist
Page 15 and 16: 15Table 1 - Example of different I/
Page 17 and 18: 17entri_enclk[63:0]in[63:0]out64 lo
Page 19 and 20: 19Guideline - We recommend careful
Page 21 and 22: 21Example 6 - VHDL example of ineff
Page 23 and 24: 232’b000 : Z = C;2’b001 : Z = D
Page 25 and 26: Guideline - Refer to the synthesis
Page 27: 274.3 Formal VerificationFormal ver

Xilinx - Design Reuse Methodology for ASIC and FPGA Designers.pdf

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?