CPLD & FPGA Architectures and Applications

Tuesday, April 11, 2023 Dr.Y.Narasimha Murthy Ph.D

CPLD & FPGA ARCHITECTURES AND

APPLICATIONS

Dr . Y . NARASIMHA MURTHY. Ph.DSRI SAIBABA NATIONAL COLLEGE

(AUTONOMOUS)

[email protected]


PROLOGUE• The popular digital ICs like TTL or CMOS have

fixed functionality and the user has no option to change or modify their functionality .i.e they work according to the design given by the manufacturer.

• To get the user expected functionality from these ICs designers started thinking of a methodology by which the functionality of an IC can be modified or changed.

• This introduced the idea of using Fuses in ICs very soon gained momentum.


Contd..

• This is the motivation for the invention of programmable devices and was realized in early 70s with the design of PLD by Ron Cline from Signetics .

• The method of changing or modifying the functionality of an IC using the Fuses was appreciated and gained momentum soon.

• This method of blowing a Fuse between two contacts or keeping the Fuse intact was done by using a software and hence these devices were called Programmable Logic Devices(PLDs).


Contd..

• Programmable Read-Only Memories(PROM) were the first programmable logic devices to achieve widespread use in digital systems.

• PROM allowed the chip vendor, to store code in the device using a simple and relatively inexpensive desktop programmer.

• This new device was called a programmable read only memory (PROM).


Contd..• The process of storing the code in the PROM is

called programming, or “burning” the PROM. PROMs, like ROMs, retain their contents even after power has been turned off.

• The PROMs were initially intended for storing code and constant data but design engineers found them useful for implementing logic also.

• The engineers could program state machine logic into a PROM, creating what is called “micro-coded” state machines.


A simple PROM CELL


Contd..

• A PROM can be constructed with an array of fuses and transistors as shown in the previous slide. The fuses are like household fuses that consist of a wire that breaks connection when a large amount of current goes through it.

• To program a one-bit cell as a logic one or zero, the fuse for that cell is selectively burned out or left connected.


Contd..

• Eventually, erasable PROMs were developed which allowed users to program,erase, and reprogram the devices using an inexpensive, desktop programmer.

• Typically, PROMs now refer to devices that cannot be erased after being programmed. Erasable PROMS include erasable programmable read only memories (EPROMs) that are programmed by applying high-voltage electrical signals and erased by flooding the devices with UV light


Contd…• Electrically erasable programmable read only

memories (EEPROMs) are programmed and erased by applying high voltages to the device.

• Flash EPROMs are programmed and erased electrically and have sections that can be erased electrically in a short time and independently of other sections within the device.


Contd..• PROMs are excellent for implementing any kind of

combinational logic with a limited number of inputs and outputs. Each output can be any combinatorial function of the inputs, no matter how complex.

• The problem with PROMs is that they tend to be extremely slow.

• Even today , access times are of the order of 40 nano-seconds or more

• Hence they are not useful for high speed applications .


Classification of PLDs• The classification of PLDs is given below.


Simple Programmable Logic Device [SPLD]

• As the name suggests SPLD has a simple architecture. PROM is a best example for SPLD.

• SPLD is capable of implementing hundreds of gates and normally programmed by the user by using inexpensive programmers.

• The main limitation of SPLDs is their low logic capacities due to the restricted nature of AND-OR planes.


BASIC CIRCUIT OF PLD


Contd…


Contd..

• There are three main types of SPLD architectures

• (i).Programmable logic array (PLA), ii).Programmable array logic (PAL) , and

(iii).sGeneric array of logic (GAL)


Configuration of SPLDs


Contd…• Two of the most popular SPLDs are the PALs

produced by Advanced Micro Devices (AMD) known as the 16R8 and 22V10.

• Both of these devices are industry standards and are widely second-sourced by various companies.

• The name “16R8” means that the PAL has a maximum of 16 inputs (there are 8 dedicated inputs and 8 input/outputs), and a maximum of 8 outputs.


Contd…

• The “R” refers to the type of outputs provided by the PAL and means that each output is “registered” by a D flip-flop.

• Similarly, the “22V10” has a maximum of 22 inputs and 10 outputs. Here, the “V” means each output is versatile and can be configured in various ways, some configurations registered and some not.


Contd..

• Another widely used and second sourced SPLD is the Altera Classic EP610.

• This device is similar in complexity to PALs, but it offers more flexibility in the way that outputs are produced and has larger AND- and OR- planes.

• In the EP610, outputs can be registered and the flip-flops are configurable as any of D, T, JK, or SR.


PLA- Programmable logic array • The PLA consists of two programmable planes

AND and OR . The AND plane consists of programmable interconnect along with AND gates.

• The OR plane consists of programmable interconnect along with OR gates.

• Each of the inputs can be connected to an AND gate with any of the other inputs by connecting the crossover point of the vertical and horizontal interconnect lines in the AND gate programmable interconnect.


Contd..• Initially, the crossover points are not electrically

connected, but configuring the PLA will connect particular cross over points together.

• The AND gate is seen with a single line to the input. This view is by convention, but this also means that any of the inputs (vertical lines) can be connected. Hence, for four PLA inputs, the AND gate also has four inputs. The single output from each of the AND gates is applied to an OR gate programmable inter connect.


Contd..


PROGRAMMABLE ARRAY LOGIC (PAL) • The first programmable device was the

programmable array logic (PAL) developed by Monolithic Memories Inc(MMI).

• The Programmable Array Logic or PAL is similar to PLA, but in a PAL device only AND gates are programmable. The OR array is fixed by the manufacturer.

• This makes PAL devices easier to program and less expensive than PLA. On the other hand, since the OR array is fixed, it is less flexible than a PLA device.


Schematic Representation- PAL


Block diagram of PAL


PAL contd..

• The PAL device. has n input lines which are fed to buffers/inverters.

• Buffers/inverters are connected to inputs of AND gates through programmable links. Outputs of AND gates are then fed to the OR array with fixed connections


GAL-Generic Array Logic

• PAL and PLA devices are one-time programmable (OTP) based on PROM, so the PAL or PLA configuration cannot be changed after it has been configured.

• This limitation means that the configured device would have to be discarded and a new device configured. The GAL, although similar to the PAL architecture, uses EEPROM and can be reconfigured.


Contd…• The Generic Array Logic (GAL) device was

invented by Lattice Semiconductor. • The GAL was an improvement on the PAL

because one device was able to take the place of many PAL devices or could even have functionality not covered by the original range. Its primary benefit, however, was that it was erasable and re-programmable making prototyping and design changes easier for engineers.


Complex Programmable Logic Devices (CPLDs)

• CPLDs were pioneered by Altera, first in their family of chips called Classic EPLDs, and then

in three additional series, called MAX 5000, MAX 7000 and MAX 9000.

• The CPLD is the complex programmable Logic Device which is more complex than the SPLD.

• This is build on SPLD architecture and creates a much larger design. Consequently, the SPLD can be used to integrate the functions of a number of discrete digital ICs into a single device and the CPLD can be used to integrate the functions of a number of SPLDs into a single device.


Contd..

• CPLD architecture is based on a small number of logic blocks and a global programmable interconnect.

• Instead of relying on a programming unit to configure chip , it is advantageous to be able to perform the programming while the chip is still attached to its circuit board.

• This method of programming is known is called In-System programming (ISP). It is not usually provided for PLAs (or) PALs , but it is available for the more sophisticated chips known as Complex programmable logic device.


Architecture of CPLD


Contd…

• The CPLD consists of a number of logic blocks or functional blocks, each of which contains a macrocell and either a PLA or PAL circuit arrangement.

• In the diagram eight logic blocks are shown. The building block of the CPLD is the macro-cell, which contains logic implementing disjunctive normal form expressions and more specialized logic operations.


Contd..• The macro cell provides additional circuitry

to accommodate registered or nonregistered outputs, along with signal polarity control.

• Polarity control provides an output that is a true signal or a complement of the true signal.

• The actual number of logic blocks within a CPLD varies ,the more logic blocks available, the larger the design that can be configured.


Contd..• In the center of the design is a global

programmable interconnect.• This interconnect allows connections to the

logic block macrocells and the I/O cell arrays (the digital I/O cells of the CPLD connecting to the pins of the CPLD package).

• The programmable interconnect is usually based on either array-based interconnect or multiplexer-based interconnect


CPLD Architecture


Contd..• Multiplexer-based interconnect uses digital

multiplexers connected to each of the macrocell inputs within the logic blocks.

• Specific signals within the programmable interconnect are connected to specific inputs of the multiplexers.

• It would not be practical to connect all internal signals within the programmable interconnect to the inputs of all multiplexers due to size and speed of operation considerations.


FIELD PROGRAMMABLE GATE ARRAYS

• The concept of FPGA was emerged in 1985 with the XC2064TM FPGA family from Xilinx .

• The “FPGA is an integrated circuit that contains many (64 to over 10,000) identical logic cells that can be viewed as standard components.”

• The individual cells are interconnected by a matrix of wires and programmable switches.


Contd..• Unlike CPLDs (Complex Programmable Logic

Devices) FPGAs contain neither AND nor OR planes.

• The FPGA architecture consists of configurable logic blocks, configurable I/O blocks, and programmable interconnect.

• Also, there will be clock circuitry for driving the clock signals to each logic block, and additional logic resources such as ALUs, memory, and decoders may be available.


Contd..• The two basic types of programmable

elements for an FPGA are Static RAM and anti-fuses.

• Each logic block in an FPGA has a small number of inputs and one output.

• A look up table (LUT) is the most commonly used type of logic block used within FPGAs.

• There are two types of FPGAs.(i) SRAM based FPGAs and (ii) Anti-fuse technology based(OTP).


FPGA-Architecture


Contd..

Every FPGA consists of the following elements.

• Configurable logic blocks(CLBs)• Configurable input output blocks(IOBs)• Two layer metal network of vertical and

horizontal lines for interconnecting the CLBS. Which are called Programmable Interconnects.


XILINX Logic Cell Array(LCA)• LCA is the novel architectural feature

introduced by XILINX in the year 1985 for their FPGA devices. It is almost like a proprietary or trade mark property of XILINX implemented for FPGA devices.

• The XILINX LCA architecture consists of three major Components. They are

(i) Configurable Logic Blocks (CLBs)

(ii) Input/Output Blocks (lOBs) and

(iii)Programmable Interconnect.


Contd…

• In addition, configuration memory is used to hold the configuration program bits which control the configuration of CLRM, IOBs and interconnect.

• This LCA architecture consists of an interior matrix of logic blocks and a surrounding ring of I/O interface blocks.

• Interconnect resources occupy the channels between the rows and columns of logic blocks and between the logic blocks and I/O blocks. Like a microprocessor the LCA is a program driven logic device.


Contd..

• The functions of the LCA’s configurable logic blocks and I/O blocks and their interconnection are controlled by a configuration program stored in an on-chip memory.

• The configuration program is loaded automatically from an external memory on power-up or on command, or is programmed by a microprocessor as part of system initialization


Contd..As shown below diagram the configuration memory consists of a distributed array of static memory cells.During configuration the cell is written through the data line and is read through the data line during read back operation


LCA-Architecture

• The core of the LCA is a matrix of identical Configurable Blocks (CLBs).Each CLB contains programmable combinational logic and storage registers.

• The combinational logic section of of the block is capable of implementing any Boolean function of its input variables.

• The registers can be loaded from the combinational logic or directly from a CLB input the register outputs can be inputs to the combinational logic via an internal feedback path


Block Diagram


Contd..

• The periphery of the Logic Cell Array is made up of user programmable input/output blocks (IOBs).

• Each block can be programmed independently to be an input ,an output or bi-directional pin with three state control. Inputs can be programmed to recognize either TTL or CMOS thresholds.

• Each IOB also includes flip-flops that can be used to buffer inputs and outputs.


Programmable Interconnect

• In FPGAs three types of metal resources are provided to fulfill various network interconnect requirements. They are

• General Purpose Interconnect

• Direct Connection

• Long lines (multiplexed busses and wide AND gates)


General Purpose Interconnect

• It consists of a grid of five horizontal and five vertical metal segments located between the rows and columns of logic and IOBs.

• Each segment is the height or width of a logic block.

• Switching matrices join the ends of these segments and allow programmed interconnections between the metal grid segments of adjoining rows and columns.


Contd..


Contd...• The switches of an un-programmed device are all

non-conducting. • The connections through the switch matrix may be

established by the automatic routing or by selecting the desired pairs of matrix pins to be connected or disconnected.

• The interconnect buffers are available to propagate signals in either direction on a given general interconnect segment.

• These bidirectional (bidi) buffers are found adjacent to the switching matrices, above and to the right.


Direct Interconnect

• Direct interconnect provides the most efficient implementation of networks between adjacent CLBs or I/O Blocks. Signals routed from block to block using the direct interconnect exhibit minimum interconnect propagation and use no general interconnect resources.


Contd..


Contd…

• Direct interconnect should be used to maximize the speed of high-performance portions of logic.

• Where logic blocks are adjacent to IOBs, direct connect is provided alternately to the IOB inputs (I) and outputs (O) on all four edges of the die.

• The right edge provides additional direct connects from CLB outputs to adjacent IOBs.


Long lines

• The Long lines bypass the switch matrices and are intended primarily for signals that must travel a long distance, or must have minimum skew among multiple destinations.

• Long lines, run vertically and horizontally the height or width of the interconnect area.

• Each interconnection column has three vertical Long lines, and each interconnection row has two horizontal Long lines.


Contd..


Contd…• Two additional Long lines are located adjacent to

the outer sets of switching matrices. • Long lines can be driven by a logic block or IOB

output on a column-by-column basis. • This capability provides a common low skew

control or clock line within each column of logic blocks.

• Isolation buffers are provided at each input to a Long line and are enabled automatically by the development system when a connection is made.


Technology Mapping for FPGA

• The high functionality of FPGA logic blocks presents new challenges for logic synthesis. So,the technology mapping provides a solution for FPGAs that use lookup tables to implement combinational logic.

• Technology mapping is a process of transforming a technology independent Boolean network into a technology dependent network.

• For example a K input lookup table (LUT) is a digital memory that can implement any Boolean function of K variables


Contd..• Technology mapping is the logic synthesis task

that is directly concerned with selecting the circuit elements used to implement the optimized circuit.

• Previous approaches to technology mapping have focused on using circuit elements from a limited set of simple gates.

• However such approaches are inappropriate for complex logic blocks where each logic block can implement a large number of functions


Library-Based Technology Mapping

• In library based mapping, gates or components are selected from a technology library to implement a circuit.

• Hence it is also referred to as library binding. So, this method generates a technology mapping for a given Boolean network using a characterized cell library with the objective of cost optimization or delay optimization


Contd..

• In this method the set of available circuit elements is represented as a library of functions and the construction of the optimized circuit is divided into three sub problems

• (i). Decomposition, (ii). Matching and (iii) Covering.• The original network is first decomposed into a canonical

representation that uses limited fan in NAND nodes. • This decomposition guarantees that there will be no nodes

in the network that are too large to be implemented by any library element provided the library includes NAND gates that reach the fan in limit.


contd..

• After decomposition the network is partitioned into a forest of trees The optimal sub circuit covering each tree is constructed and finally the circuit covering the entire network is assembled from these sub circuits.

• To form the forest of trees, the decomposed network is partitioned at fan out nodes into a set of single output sub networks.


Contd..

• The major obstacle to applying library-based technology mapping to LUT circuits is the large number of different functions that a K-input LUT can implement.

• The function implemented by a K-input LUT is determined by the values stored in its 2K memory bits. Since each bit can independently be either 0 or 1, there are 22K

different Boolean functions of K- variables.


contd..• The major obstacle to applying library-based

technology mapping to LUT circuits is the large number of different functions that a K-input LUT can implement.

• The function implemented by a K-input LUT is determined by the values stored in its 2K memory bits. Since each bit can independently be either 0 or 1, there are 22K

different Boolean functions of K- variables


Contd..• For values of K greater than 3 the library required to

represent a K-input LUT becomes very large. • The size of the library can be reduced by noting that

some patterns are equivalent after a. permutation of inputs.

• The inversion of outputs or inputs, which is trivially accomplished with a LUT, can also produce equivalent ‘patterns.

• Another alternative is to use a partial library tuned to take advantage of the network structure likely to be produced by technology independent logic optimization.


LUT-based Technology Mapping• The limitations of earlier technology

mapping approaches paved the way for the development of technology mapping that deals specially with LUT circuits.

• The first LUT based technology mappers appeared in 90s. and later improved for optimized delay performance of LUT circuits by minimizing the number of levels of LUT in the final circuit.


Contd..• In LUT based FPGAs (example XILINX FPGAs)

the building blocks are LUTs and Flip-Flops. • In an LUT based FPGA chip the basic

programmable logic block is a K-input Look Up Table.(K-LUT) which can implement any Boolean function of up to K- variables.

• The technology mapping in LUT based FPGA designs is to cover a general Boolean Network using K-LUTs to obtain functionally equivalent K-LUT network.


Contd..• The main objectives in LUT mapping are

(i).Cost optimal mapping i.e Minimizing the number of LUTs and Minimizing the number of CLBs

(ii) Delay optimal mapping i.e Minimizing the number of LUT levels and Minimizing the delays (including routing delays)

(iii).Maximizing the routability of the mapping schemes.

• The LUT based technology can be implemented using two types of algorithms .They are

• (a).The Area Algorithm and (b).The delay algorithm


MULTIPLEXER BASED TECHNOLOGY MAPPING

• This Multiplexer based technology mapping is used in ACTEL FPGAs and in recent Xilinx VIRTEX 6 FPGA devices .

• Because their logic block architectures are MUX based.

• In Actel based FPGAs ,the size of the Multiplexers is small and suitable to achieve the objective of area optimization and minimum delays.


Contd..• Circuits usually contain a large number of

multiplexers (MUXes).• This is mainly true for circuits that are automatically

synthesized from high-level descriptions. • MUXes exist in the data-paths of circuits, where they

are used to route operands to operators. Also, the control logic is frequently specified as a CASE statement in HDL descriptions.

• MUXs arise as a result of a direct translation of CASE statements in HDLs into a logic-level description


Contd..• The main objective behind this Mux based

technology mapping is ,describing a combinational circuit in terms of Boolean equations and realize it using minimum number of basic blocks of the target Mux based architecture and minimizing the delay on the critical path.

• In this algorithm an appropriate base function ,a library of cells and a set of pattern graphs are selected .


Contd…• The advantages of MUX based technology

mapping are it generates optimal mappings, which are often much better than those produced by conventional heuristic techniques.

• Moderately large circuits can be mapped optimally in a small amount of time. Very large circuits can be mapped near-optimally by partitioning the circuits and mapping each partition individually


Programming Technologies

• There are a number of programming technologies that have been used for reconfigurable architectures.

• Each of these technologies have different characteristics and have significant effect on the programmable architecture.

Some of the well-known technologies are

(i).SRAM Based Programming Technology (ii).Flash Programming Technology(EEPROM) , and (iii) Anti-fuse based Programming Technology


SRAM-Based Programming Technology

• Static memory cells are the basic cells used for SRAM-based FPGAs.

• Most commercial vendors like XILINX, Lattice and Altera etc.use static memory (SRAM) based programming technology in their devices.

• These devices use static memory cells which are divided throughout the FPGA to provide configurability.


Contd..• There are two primary uses for the SRAM cells.

Most of them are used to set the select lines to multiplexers that steer interconnect signals.

• The majority of the remaining SRAM cells are used to store the data in the lookup-tables (LUTs) that are typically used in SRAM-based FPGAs to implement logic functions.

• Historically, SRAM cells were used to control the tri-state buffers and simple pass transistors that were also used for programmable interconnect.


• SRAM-based programming technology has become the dominant approach for FPGAs because of its re-programmability and the use of standard CMOS process technology and therefore leading to increased integration, higher speed and lower dynamic power consumption of new process with smaller geometry.


Contd..

• There are however a number of drawbacks associated with SRAM-based programming technology.

• For example an SRAM cell requires 6 transistors which makes this technology costly in terms of area compared to other programming technologies.

• Further SRAM cells are volatile in nature and external devices are required to permanently store the configuration data.

• These external devices add to the cost and area overhead of SRAM-based FPGAs.


Flash Programming Technology

• An important alternative to the SRAM-based programming technology is the use of flash or EEPROM based programming technology. This technology inject charge onto a gate that “floats” above the transistor.

• This approach is used in flash or EEPROM memory cells. These cells are non-volatile; they do not lose information when the device is powered down.

• With modern IC fabrication processes, it has become possible to use the floating gate cells directly as switches.


Contd..

• Flash memory cells, in particular, are now used because of their improved area efficiency.

• The widespread use of flash memory cells for non-volatile memory chips ensures that flash manufacturing processes will benefit from steady decreases in process geometries.

• Flash-based programming technology offers several advantages. For example, this programming technology is nonvolatile in nature.


• Flash-based programming technology is also more area efficient than SRAM-based programming technology.

• Flash-based programming technology has its own disadvantages also.

• Unlike SRAM-based programming technology, flash based devices cannot be reconfigured/reprogrammed an infinite number of times.

• Also, flash-based technology uses non-standard CMOS process.


Contd..• This flash-based programming technology

offers several unique advantages, most importantly non-volatility.

• This feature eliminates the need for the external resources required to store and load configuration data when SRAM-based programming technology is used.

• Additionally, a flash-based device can function immediately upon power-up instead of having to wait for the loading of configuration data.


Contd..

• The flash approach is more area efficient than SRAM-based technology which requires up to six transistors to implement the programmable storage.

• The programming circuitry, such as the high and low voltage buffers needed to program the cell, contributes an area overhead not present in SRAM-based devices.


Contd..• In devices from Altera, Xilinx and Lattice,

on-chip flash memory is used to provide non-volatile storage while SRAM cells are still used to control the programmable elements in the design.

• This addresses the problems associated with the volatility of pure-SRAM approaches, such as the cost of additional storage devices or the possibility of configuration data interception, while maintaining the infinite re-configurability of SRAM-based devices


Anti-fuse Programming Technology

• An alternative to SRAM and floating gate-based technologies is anti fuse programming technology.

• This technology is based on structures which exhibit very high-resistance under normal circumstances but can be programmably “blown” (in reality, connected) to create a low resistance link.


Contd..

• An anti-fuse is a two terminal device with an unprogrammed state presenting a very high resistance between its terminals.

• When a high voltage (from 11 to 20 volts, depending on the type of anti-fuse) is applied across its terminals the anti-fuse will “blow” and create a low resistance link.

• This link is permanent.


• Programming an anti-fuse requires extra circuitry to deliver the high programming voltage and a relatively high current of 5 mA or more.

• This is done in through fairly sizable pass transistors to provide addressing to each anti-fuse. Anti-fuse technology is used in the FPGA’s from Actel , Quick logic , and Cross point


Contd..• A major advantage of the anti-fuse is its

small size, little more than the cross-section of two metal wires.

• But this advantage is limited by the large size of the necessary programming transistors, which handle large currents, and the inclusion of isolation transistors that are sometimes needed to protect low voltage transistors from high programming voltages.


Contd..

• A second major advantage of an anti-fuse is its relatively low series resistance.

• The on-resistance of the ONO anti-fuse is 300 to500 ohms, while the amorphous silicon anti-fuse is 50 to100 ohms.

• Additionally, the parasitic capacitance of an un programmed amorphous anti-fuse is significantly lower than for other programming technologies


Contd..• The limitations of this technology are , this

technology does not make use of standard CMOS process.

• Also, anti-fuse programming technology based devices cannot be reprogrammed.

• The ideal technology should be re-programmable, non-volatile, and that uses a standard CMOS process.

• But it is clear that none of the above technologies satisfy these conditions


Comparison of Programming Technologies

Inspites of all the advantages and disadvantages, the SRAM-based programming technology is the most widely used programming technology. The main reason is its use of standard CMOS process .Due to this reason it is expected that this technology will continue to dominate the other two programming technologies


XILINX XC3000 FPGA Device

• Xilinx introduced the first FPGA family, called the XC2000 series, in 1984 and next offered three more series of FPGAs namely XC3000, XC4000, and XC5000 etc.

• XC3000 series of FPGA devices were introduced in 1985 by XILINX Inc.

• This was the most successful family of FPGAs. The XC3000 archtecture includes enhancements to the XC2000 architecture to improve performance ,density and usability.


Contd..

• The XC3000 Family covers a range of nominal device densities from 2,000 to 9,000 gates, practically achievable densities from 1,000 to 6,000 gates with up to 144 user-definable I/Os.

• The XC3000 Configurable Logic block is substantially larger than XC2000 and Each of the lookup tables has four inputs and requires 16 bits of configuration memory.

• There are now four distinct families within the XC3000 Series of FPGA devices


XC3000 Family of Devices

The basic LCA (Logic Cell Array) of XC3000 consists of three components .They are Programmable I/O Blocks , Configurable Logic Block and Programmable Interconnect. In addition to this a small amount of configurable memory is also present


Programmable I/O Block • The I/O Block of the XC3000 is more complex

than the XC2000 , IOB. The important addition in this is a flip-flop in the out-put path

• By registering the data in IOB ,the clock to-out- time does ot include interconnect delays.

• Each user-configurable IOB provides an interface between the external package pin of the device and the internal user logic. Each IOB includes both registered and direct input paths


Programmable I/O Block


Contd..• Each IOB includes input and output storage

elements and I/O options selected by configuration memory cells.

• A choice of two clocks is available on each die edge. The polarity of each clock line (not each flip-flop or latch) is programmable.

• Each input circuit also provides input clamping diodes to provide electrostatic protection, and circuits to inhibit latch-up produced by input currents.


Configurable Logic Block(CLB)

• The XC3000 CLB is substantially larger than the XC2000 CLB.

• Each of the look-up tables has four inputs rather than three and hence requires sixteen bits of configuration memory rather than eight.

• The lookup tables can be combined with a multiplexer to produce any function of five inputs and some functions of up to seven inputs


• The XC3000 CLB has two flip-flops ,to ensure that all combinational logic can be followed by a pipelining flip-flop.

• The register rich CLB allows the XC3000 to implement state intensive applications and heavily pipe lined designs efficiently.

• Each CLB has a combinatorial logic section, two flip-flops, and an internal control section. The CLB has five logic inputs (A, B, C, D and E)


XC3000 CLB


Contd..• Data input for the flip-flops within a CLB is

supplied from the function F or G outputs of the combinatorial logic, or the block input, DI.

• Both flip-flops in each CLB share the asynchronous RD which, when enabled , is dominant over clocked inputs.

• All flip-flops are reset by the active-Low chip input, RESET, or during the configuration process.


Programmable Interconnect

• Programmable-interconnection resources in the Field Programmable Gate Array provide routing paths to connect inputs and outputs of the IOBs and CLBs into logic networks.

• Interconnections between blocks are composed of a two-layer grid of metal segments.

• Specially designed pass transistors, each controlled by a configuration bit, form programmable interconnect points (PIPs) and switching matrices used to implement the necessary connections between selected metal segments and block pins.


Contd..• The XC3000 interconnect structure has five general

interconnect lines both vertically and horizontally .• In addition each CLB has direct connections to

adjacent CLBs both vertically and horizontally.• Three types of metal resources are provided to

accommodate various network interconnect requirements.

• General Purpose Interconnect

• Direct Connection

•Long lines (multiplexed busses and wide AND gates)


XC3000 Interconnect


XILINX XC4000 FPGA Device

• The XC4000 was designed to improve performance and gate density for large designs.

• Several dedicated features were added to the general purpose logic features of XC3000 , resulting an interesting combination of special -purpose and general purpose functions.

• The XC4000 family was designed using placement and routing tools to evaluate architectural decisions.


The basic building blocks in the XC4000 family

• Look-up tables for implementation of logic functions.• A designer can use a fumction generator to implement

any Boolen function of a given number of inputs by pre-loading the memory with the bit pattern corresponding to the truth table of the function.

• All functions of a function generator have the timing ,the time to look-up results in the memory.

• Therefore ,the inputs to the function generator are fully interchangeable by simple rearrangement of the bits in the look-up table.


Contd..• A Programmable Interconnect Point(PIP) is a

pass transistor controlled by a memory cell.

The PIP is the basic unit of configurable interconnect mechanism.

• The wire segments on each side of the transistor are connected depending on the value in the memory cell.

• The pass transistor introduces resistance into the interconnected paths and hence delay occurs.


Advanced Features of the XC4000 FPGAs

• CLBs can be used as on-chip RAM• Fast carry chain for highspeed implementation

of arithmetic• Boundary scan compatibility (JTAG)• Wide decode logic,More global clocks• Faster placement and routing algorithms• Scaled routing resources.


Configurable Logic Block (CLB)

• The XC4000 CLB is similar to the XC3000CLB. It contains three lookup tables and two flip-flops.(F,G &H)

• The two primary look-up tables F & G implement any function of four variables.

• These two results can be brought out of the block independently or they can be combined with another input in the H –look up table to make any function of five inputs or some function of up to nine inputs.


Contd..

• The XC3000 can implement arithmetic with sum in one look-up table and carry in another look-up table.

• The XC4000 CLB can implement arithmetic in this way also,but as the speed of the arithmetic operation is dominated by the speed of the carry chain ,the XC4000 CLB includes dedicated high speed carry logic.


Block Diagram-CLB


XC4000 I/O BLOCK• The signals to be output from the chip can be

registered before output and enabled by a separate control signal.

• Outputs can be optionally pulled up or down and the output driver can be configured with either fast or or slow slew rate.

• Inputs from the pad can be brought into the interior of the chip directly ,registered or both to facilitate multiplexed bus interfaces


Contd..• The XC4000IOB includes boundary scan logic

compatible with the ANSI EEE1149.1 (JTAG) boundary scan standard.

• The boundary scan can check internal logic or external logic.

• Scan operation can take place before and after the FPGA is programmed and do not interfere with the operation of the part.


Interconnect Structure

• The XC4000 interconnect is arranged in horizontal and vertical channels.

• Each channel contains some number of short wire segments that span a single CLB (the number of segments in each channel depends on the specific part number), longer segments that span two CLBs, and very long segments that span the entire length or width of the chip.

• Programmable switches are available to connect the inputs and outputs of the CLBs to the wire segments, or to connect one wire segment to another..


Contd..

The figure below shows only the wire segments in a horizontal channel, and does not show the vertical routing channels, the CLB inputs and outputs, or the routing switches


Contd..

• The salient feature about the Xilinx interconnect is that signals must pass through switches to reach one CLB from another, and the total number of switches traversed depends on the particular set of wire segments used.

• Thus, speed-performance of an implemented circuit depends in part on how the wire segments are allocated to individual signals by CAD tools.


Actel FPGAs

• In contrast to XILINX FPGAs the devices manufactured by Actel are based on anti fuse

technology. • Actel offers three main families .They are :

Act 1, Act 2, and Act 3. • Actel devices are based on a structure similar

to traditional gate arrays; the logic blocks are arranged in rows and there are horizontal routing channels between adjacent rows.


LOGIC BLOCK –ACTEL FPGA


Contd..• The logic blocks in the Actel devices are

relatively small in comparison to the LUT based ones. , and are based on multiplexers.

• It comprises an AND and OR gate that are connected to a multiplexer based circuit block.

• The multiplexer circuit is arranged such that, in combination with the two logic gates, a very wide range of functions can be realized in a single logic block.


Contd..

• Actel’s interconnect is organized in horizontal routing channels.

• The channels consist of wire segments of various lengths with anti-fuses to connect logic blocks to wire segments or one wire to another.

• Also, Actel chips have vertical wires that overlay the logic blocks, for signal paths that span multiple rows.

• In terms of speed-performance, it is evident that Actel chips are not fully predictable, because the number of anti-fuses traversed by a signal depends on how the wire segments are allocated during circuit implementation by CAD tools.


Quicklogic pASIC FPGAs • The Quicklogic is the main competitor for Actel in

anti-fuse -based FPGAs . • It produces two families of devices, called pASIC

and pASIC-2. The pASIC-2 is an enhanced version of pASIC.

• The pASIC, consists of a regular two-dimensional array of blocks called pASIC Logic Blocks (pLBs).

• The logic capacities of first generation of Quick Logic FPGAs is between 48 and 380pLBs,or 500 to 4000 equivalent MPGAs gates.s


Contd..

As shown in figure below pASIC has similarities to other FPGAs i.e the overall structure is array-based like Xilinx FPGAs, and logic blocks use multiplexers similar to Actel FPGAs, and the interconnect consists of only long- lines like in Altera FLEX 8000.


Contd..• pASIC’s multiplexer-based logic block is shown in below

figure. It is more complex than Actel’s Logic Module, with more inputs and wide (6-input) AND-gates on the multiplexer select lines. Every logic block also contains a flip- flops.


Altera FLEX 8000 and FLEX 10000 FPGAs

• The first FPGA chips from Aletra were simple arrays of logic cells ,which are relatively simple logic elements (LEs),each element comprising of a three input look-up table (LUT ) to generate logic functions ,a single configurable flip-flop and multiplexers for routing the signals and selecting clocks.

• The logic cells were connected by switch boxes instead of fixed interconnect. The general architecture of Altera’s FPGAs is shown in the next slide.


Architecture of ALTERA FPGA


• There are two high performance FPGA series called FLEX series.

• Altera’s FLEX 8000 series consists of a three-level hierarchy similar to CPLDs.

• However, the lowest level of the hierarchy consists of a set of lookup tables, rather than an SPLD like block, and so the FLEX 8000 is categorized here as an FPGA.


Contd..

• The architecture of FLEX 8000 is shown in next slide.

• The basic logic block, called a Logic Element (LE) contains a four-input LUT, a flip-flop, and special-purpose carry circuitry for arithmetic circuits (similar to Xilinx XC 4000).

• The LE also includes cascade circuitry that allows for efficient implementation of wide AND functions


Architecture of Altera FLEX 8000 FPGA


contd..• A major difference between FLEX 8000 and

Xilinx chips is that Fast Track consists of only long lines. This makes the FLEX 8000 easy for CAD tools to automatically configure.

• All Fast-Track wires horizontal wires are identical, and so interconnect delays in the FLEX 8000 are more predictable than FPGAs that employ many smaller length segments because there are fewer programmable switches in the longer paths.


contd..• Predictability is furthered aided by the fact that

connections between horizontal and vertical lines pass through active buffers.

• The FLEX 8000 architecture has been extended in the state-of-the-art FLEX 10000 family.

• FLEX 10000 offers all of the features of FLEX 8000, with the addition of variable-sized blocks of SRAM, called Embedded Array Blocks (EABs) which shows that each row in a FLEX 10000 chip has an EAB on one end.


Concurrent Logic FPGA Device • The manufacturer Concurrent Logic offers the

CFA6006 FPGA device ,which is based on two dimensional array of identical blocks ,where each block is symmetrical on its four sides.

• The array holds 3136 of such blocks ,providing a total logic capacity of about 5000 equivalent gates.

• Connections are formed using multiplexers that are configured by a static RAM programming technology.


Contd..

• The structure of the Concurrent Logic Block is shown in the next slide.It comprises of user configurable multiplexers, basic gates and a D type flip-flop .

• The concurrent FPGA is especially suitable for register-intensive and arithmetic applications since the logic block can easily implement a half-adder and a register bit.


Structure of the Concurrent Logic Block


Crosspoint Solutions FPGAs• The crosspoint FPGAs are different from other

FPGAs because it is configurable at the transistor level as aoposed to logic block level in other FPGAs.

• Basically the architecture consists of rows of transistor pairs ,where the rows are separated by horizontal wiring segments .

• Veritical wiring segments are also available ,for connection among the rows


Contd..• Each transistor row comprises two lines of

series connected transistors ,with one line being NMOS and the other PMOS .

• The wiring resources allow individual transistor pairs to be interconnected to implement CMOS logic gates.

• The programming technology used for the programmable switches is similar to the Via-Link anti-fuse ,which is based on amorphous silicon.


Contd..• The structure of the transistor pair rows is

shown in the next slide.• The diagram shows the implementation

of a NOR gate and a NAND gate using the transistor lines.

• The transistor gates ,drains , sources can be programmable interconnected to other transistors and also to power and ground.


Structure of the Transistor Pair

The series connections across the lines is broken where necessary by permanently holding a transistor in its OFF state. A wide range of logic gates can be implemented by the transistor lines and the interconnection patterns.


contd..• The FPGAs currently offered by Crosspoint

Solutions has a total logic capacity of 4200 gates.

• The chip has 256 rows of transistor pairs and an additional 64-rows of multiplexer like structures are provided.

• With its rows based architecture ,anti-fuse programming technology and multiplexers ,the Crosspoint FPGAs are most similar to those of Actel FPGAs.


ALGOTRONIX CAL-1024

• This design has a two-dimensional mesh array structure which resembles the gate array “sea of gates” architecture .

• Like the Xilinx architecture, Algotronics used Static RAM programming technology to specify the function performed by each logic cell and to control the switching of connections between cells.

• The CAL1024 design contains 1024 identical logic cells arranged in a 32 X 32 matrix.


contd..• The design is considered to be a mesh-connected

architecture since each cell is directly connected to its nearest north, south, east, and west neighbors.

• In addition to these direct connects, two global interconnect signals are routed to each cell to distribute clock and other “low skew requirement” control signals.

• Figure in next slide shows the basic array architecture, indicating both nearest neighbor and global connections to the logic cells.


Basic Array Architecture


contd..• The basic building block of the Algotronix design is a

configurable cell containing multiplexers and a function unit.

• As indicated in the figure , the function unit is preceded by multiplexers which select the source for the X1 and X2 inputs.

• The function unit is capable of generating any logic function of the two inputs, or of operating as a D-type latch.

• There are four additional multiplexers which select the function output or one of the external inputs for routing to each of the four outputs (north, south, east, and west).


Commercially available FPGAs


FPGA Design Flow

• The earlier PLD and FPGA designs were performed largely by hand But to-days complex programmable logic devices requires the use of an integrated Computer-Aided Design (CAD) system.

• Both commercial CAD tool vendors and FPGA companies offer appropriate tools.

• For example, traditional Electronic Design Automation (EDA) vendors such as Cadence, Mentor Graphics, Synopsys, and View Logic etc. offer tools to support FPGA design. s


contd..• These tools are typically used for the front-end

design entry and simulation operations and provide the necessary interfaces to vendor-specific back-end tools for chip placement and routing.

• Examples of vendor specific tools are the Xilinx XACT system and the Altera MAX+PLUS II software.

• The Altera’s MAX+PLUS II software supports the entire design flow on either PC or workstation platforms.


Contd..

• The first step in the design process is the description of the logic circuit, which can be done either by schematic capture tool or with Boolean expressions.

• This is followed by a translation that converts the original circuit description into a standard format used by the suitable CAD tools (Ex: XILINX CAD tools).

• The circuit is then passed through CAD programs that partition it into appropriate logic blocks. Select a specific location in the FPGA for each logic block and form the required interconnections.( (Cadence, View Logic, OrCAD, etc.)


Initial Design Entry

• The detailed description of the logic circuit are entered using a schematic capture program. In the design entry phase, RTL or schematic entry is used to create the logic to be implemented in the device.

• Pin assignments can also be made, including pin placement information, and timing constraints that might be necessary for building a functioning design.

• In the design entry step a schematic or Block Design File (.bdf) is created that is the top-level design. The library of parameterized modules (LPM) functions are added and Verilog HDL code is used to add a logic block


Contd..

• The library may be either supplied by the vendor of the schematic capture program or any FPGA vendor(Like Xilinx or Altera etc).

• An alternate way to specify the logic circuit is to use a Boolean expression or state machine language.

• This is done without the graphical interface. Some times it is possible to use a mixture of both schematic and Boolean expressions


Translation to XNF Format

After the logic circuit is successfully designed and merged into one circuit ,it is translated into a special format that is understood by the CAD tools.For Xilinx this format is called Xilinx net list format or XNF.This translation utility is supported by the Xilinx or by the vendor of the logic entry tool.The translation process may also involve automatic optimizations of the circuit.


Partition

• The XNF circuit is partitioned into logic cells (this partition is also known as Technology Mapping).

• This technology mapping converts the XNF circuit which is a net list of basic logic gates ,into a net list of Xilinx logic cells.

• The logic cell used depends on which Xilinx product the circuit is to be implemented in. XACT tools also attempt to optimize the circuit during this step.


Place and Route

• Place &Route is performed by using either CAD tools or manually by the user or mixture of the two.

• The first step is placement ,in which each logic cell generated during the partition step is assigned to a specific location in the FPGA.

• Automatic placement can be done using the simulated annealing algorithm.

• After the placement ,the required interconnections among the logic cells must be realized by selecting wire segments and routing switches within the FPGA interconnection resources


Contd..• The XACT tools provide a critical path timing

analyzer which provides delay information on the longest through shortest paths through the chip.

• In addition, the physical layout timing information can also be back-annotated to the schematics to get more accurate functional simulation results.

• The final step in the Xilinx design flow is the creation of the BIT file which contains the binary programming data needed to configure the SRAM bits of the target chip.

• This file is then downloaded to configure the chip for final functional and timing tests of the programmed chip.


Compilation• After creating the design it must be compiled.

Compilation converts the design into a bitstream that can be downloaded into the FPGA.

• The most important output of compilation is an SRAM Object File (.sof), which is used to program the device.

• The software also generates other report files that provide information about the code as it compiles


Contd..

• In the design flow process the simulation is very important to learn, and there are entire applications devoted to simulating hardware designs.

• There are two types of simulation, RTL and timing. RTL (or functional) simulation allows you to verify that your code is place-and-route) simulation verifies that the design meets timing and functions appropriately in the device


contd..• After completion of the design ,its

performance is checked either by downloading the configuration bits into FPGA or by using an interface to a timing simulation program.

• If the performance is not satisfactory ,suitable modifications are done at some point in the design flow.

• Once the timing and functionality is verified the implementation is complete.


APPLICATIONS OF FPGAs

• FPGAs have gained rapid acceptance over the past two decades.

• Users can apply them to a wide range of applications like random logic, integrating multiple SPLDs, device controllers, communication encoding and filtering, small- to medium-size systems with SRAM blocks, and many more.

• Another interesting FPGA application is prototyping designs to be implemented in gate arrays by using one or more large FPGAs.


contd..• Another application is the emulation of entire large

hardware systems via the use of many interconnected FPGAs.

• FPGAs offer particularly powerful solutions for meeting machine vision, industrial networking, motor control, and video surveillance needs.

• For example, the flexibility of FPGAs allow designers to quickly adapt to changing image sensor interfaces and image processing requirements, evolve analysis capabilities to keep pace with market requirements, and add features and functions long after deployment.


contd..

• FPGAs are also used as custom computing machines.

• This involves using the programmable parts to execute software, rather than compiling the software for execution on a regular CPU.

• FPGAs provide a unique combination of highly parallel custom computation, relatively low manufacturing/engineering costs, and low power requirements.


Contd..• FPGAs meet critical timing and performance

requirements with parallel processing and real-time industrial application performance, permitting greater system integration and lower development cost.

• In areas such as Industrial Networking and Imaging, where the protocols and standards are shifting and changing, the programmability of FPGAs versus fixed logic chips such as ASICs and ASSPs allows for both faster time-to-market and longer time-in-market.


FINALE

• The low cost ,fast manufacturing turnaround is the secret behind the market success of FPGAs.

• Though the large, slow programmable switches prevent FPGAs from providing the speed performance ,the improvements in architecture and CAD tools will overcome these disadvantages.

• Over time FPDs will become the dominant technology for implementing digital circuits.


References

• Field Programmable Gate Arrays – S.D Brown, R.J.Francis et al

• Field Programmable Gate array Technology- Trimberger

• FPGA and CPLD Architectures : A Tutorial -STEPHEN BROWN & JONATHAN ROSE.

• FPGA Architecture: Survey and Challenges --Ian Kuon1, Russell Tessier and Jonathan Rose1

CPLD & FPGA Architectures and Applications

Technology

Transcript of CPLD & FPGA Architectures and Applications