TABLE OF CONTENTS
1. INTRODUCTION 1.1 VHDL DESIGN FLOW 1.2. INITIAL DESIGN ENTRY 1.3. WHAT IS IS VHDL? 1.4. HISTORY 1.5. CAPABILITES OF VHDL 1.6. ADVANTAGES OF VHDL OVER PROCEDURAL LANGUAGES 2. THE VHDL DESIGN HIERARCHY 2.1. DESING ENETITY 2.2. ENTITY DECLARATION 2.3. ARCHITECTURE BODY 2.3.1 DATA FLOW STYLE 2.3.2 BEHAVIORAL STYLE 2.3.3 STRUCTURAL STYLE 3. LOGIC SYNTHESIS 3.1. LOGIC OPTIMIZATION TECHNIQUES a)AREA OPTIMIZATION b)SPEED OPTIMIZATION OPTIMIZATION 4. FPGA 5. ALU 6 Code For 8 bit Adder-Substractor 7 Code for 8 bit Multiplier 8 Code for Final ALU 9 Future Prospects of VHDL
1. INTRODUCTION
1.1VHDL DESIGN FLOW
Initial Design Entry
Logic Optimization
Technology Mapping
Placement
Routing
Programming Unit
Configured FPGA
Fig. Design flow of an Integrated Circuit
1.2. INITIAL DESIGN ENTRY
As integrated circuit technology has improved to allow more and more components on a chip, digital systems have continued to grow in complexity. As digital systems have become more complex, detailed design of the systems at the gate and flop-flop level has become very ver y tedious and time consuming. For this reason, use of hardware languages in the digital des ign process continues to grow importance. A hardware description language language allows a digital system to be designed and debugged at a higher level before conversion to the gate and flip flop level. Use of synthesis computer aided design tools to do t his conversion is becoming more widespread. This is analogous to writing software programs in high-level languages such as C and then using a compiler to convert the programs to machine language. The two most popular hardware description languages are VHDL and Verilog.
1.3. WHAT IS VHDL?
VHDL is an acronym for very high-speed integrated circuit hardware descr iption language. It is a general-purpose hardware description language that is specifically designed to describe the organization and function of digital hardware system, circuit boards & components at many level of abstraction ranging from simple gate to complete digital electronics systems. VHDL model is a textual description of a hardware design or a piece of design that, when simulated mimics the design behavior.
1.4. HISTORY
The requirement for the language were first generated in 1980, under the Very High Speed Integrated Circuit (VHSIC) project of US US government, to enhance the electronic design process, technology, and procurement, spawning spawning development of many advanced integrated integrated circuit process technologies. In this program, a number of US companies were involved in the design of VHSIC chip for the Department of Defense (DOD), USA. At that time, most of the companies were using different HDL to describe and developed their ICs. As a result different vender could not effectively exchange design with one another. Thus a need for a standardized HDL for the design, documentation and verification of digital system was generated. A team of three of companies, IBM, TI & INTERMETRICS, developed a version of language. In 1986, VHDL was proposed as an IEEE standard. It w went ent through a number of revisions and changes until it was adopted as the IEEE 1076 standard in December 1987.
1.5. CAPABILITES OF VHDL
The following are the major capabilities that VHDL provide along with the feature that differentiate it from other Hardware Description languages.
1. The language can be used as an exchange medium between chip vender and CAD tool users. Different chip venders can provide VHDL description of their components to system designers. CAD tool users can use it to capture the behavior of the design at a high level of abstraction for functional simulation.
2. The language can also be used as a communication medium between different CAD and CAM tools. For example a schematic capture program ma y be used to generate a VHDL description for the design, which can be used as an input to a simulation program.
3. The language supports hierarchy i.e. a digital system can be modeled as a set of interconnected sub-components.
4. The language is not technology specific, but is capable of supporting different technologies. It can support various hardware technologies: for example - new logic types and new components may be defined; technology specific attributes can be used. By being technology independent the same model can be synthesized into different vendor libraries.
5. It supports both synchronous s ynchronous and asynchronous asynchronous timing models.
6. It is an IEEE and ANSI STANDARD; therefore, models described in this language are portable.
7. The language supports three basic different description styles: structural, data flow and behavioral. A design may be described in any combination of these three descriptive styles.
8. It supports a wide range of abstraction abstracti on level ranging from behavioral description to very precise gate level descriptions. It does not however support modeling at or below the transistor level. It allows a design to be captured at a mixed level le vel using a single coherent language.
9. Arbitrary large design can be modeled using the language, and there are no limi tations imposed by the language on the size of a design.
10. The language has element that make large-scale modeling easier, for example component, functions, procedure, and packages.
11. Nominal propagation delays, min-max delay, setup and holding time and spike detection can all be very naturally done in this language.
12. A model can, not only describe the functionality of a design but also the information about the design itself in terms of user defined attributes such as total area and speed.
13. A common language can be used to describe library components from different vendors. Tools that understand VHDL models will have no difficulty in reading models from a variety of venders since the languages is a standard.
14. Models written in this language can be verified by simulation, since precise simulation semantics are defined for language construct.
15. The capability of defining new data types provides the power of describe and simulate a new design technique at a very high level of abstracti on without any concern the implementation details.
16.The language is publicly available and human readable.
1.6. ADVANTAGES OF VHDL OVER PROCEDURAL LANGUAGES
Main difference between VHDL and other programming languages like C/C++ are a) VHDL is a parallel language, while C/C++ a re sequential languages. Each statement occurring in VHDL is executed concurrently, while in C/C++ each statement is executed sequentially and at its own turn. In VHDL explicit constructs exi st for explicit sequential steps.
b) VHDL is a strongly typed language. language. It doesn’t allow any mismatching of types, though type conversion is permitted.
c) VHDL allows use of explicit time ti me delay, which isn’t applica ble in procedural languages. In procedural languages the right hand side value is assigned to the left hand side value as soon as the statement is executed. In VHDL there is an advantage that the computed value can be assigned to a signal after any time delay.
d) VHDL model cannot be implemented in real time application directl y like other procedural languages. It is simulated and synthesized using in built s ystem clock.
2. THE VHDL DESIGN HIERARCHY
Any hardware design can be described in terms of its operation at different levels of abstraction, from system through to logic gate. At each level of this hierarchy the overall inputs and outputs remain the same but the functionality of distinct sections become clearer. When the local inputs and outputs and the function of a block are sufficiently defined, the hardware can be designed.
VHDL is capable of describing a well-defined hardware block at any level of abstraction. A design entity is the VHDL representation of such a block and can be considered to be at the top of the design hierarchy. Within the design entity, the function of the hardware is often further decomposed by using external and internal blocks.
External blocks are other design entities that have been previously complied and stored in a library. These types of blocks are generally referred to as components . Each component can further extend the hierarchy by also using external and internal blocks to describe its own function. The present design entity can also be thought of as external block.
Internal blocks are self-contained functional units that have explicitly defined input and output signals. They therefore represent the next level down in the hierarchy. Again, an internal block may contain external or other internal blocks. The Block statement is used to create an internal block.
A design entity contains an entity declaration and an architect ure body. Figure below shows the organization of the VHDL design hierarchy and illustrates the main elements of an architecture body. The port connection, enable information to be passed between blocks and, at the top level, allow the design entity to communicate with the external environment. A design entity may contain any or all of the elements shown, in the diagram. It will always have an entity declaration and architecture body at t he top level
2.1. DESING ENETITY
Entity declaration
Architecture body Component
Concurrent Statements
Component Internal block Internal block
EXTERNAL BLOCK Entity declaration declaration
Architecture body component component
component Concurrent Block statement
Concurrent Internal
Internal
INTERNAL BLOCK
block
block
statement component Internal block
Internal block
Fig.2. Hierarchical organization organization of a VHDL VHDL design
2.2 ENTITY DECLARATION The entity declaration is the interface between the external environment, such as a top-level schematic, and the design. It is at the top of the design three for every ever y external block. Hence, it will usually contain a description of the inputs to and outputs from the block in the form of a port statement. An entity declaration without a port statement does not have any external connections. The Generic statement can be used within the entity declaration to pass timing, control or environmental data into the design unit. A component declaration may also have a Generic statement. This information is useful when simulating a design, but when
synthesizing most of it is irrelevant. Hence, the t he Generic statement usually only supports the declaration of integer generics, which have a const ant value and can therefore be substituted into the associated design unit at compile time. The syntax for entity declaration is as follows
Entity entity – entity – name name is
[port (interface-signal-declaration);] end [entity] [entity-name];
2.3. ARCHITECTURE BODY The architecture body is used to specify the relationship between the inputs and outputs declared in the entity. It therefore describes descr ibes the actual function of the hardware. Although each entity must be unique, several architectures can be associated with one entity. This allows the function of a block to the changed without changing its external structure. The architecture body contains any number of concurrent statements components and internal blocks are also of this type. These statements can be considered to be executi ng asynchronously and completely independently of each other. It is therefore crucial to understand how the relationships between different concurrent statements a re going to affect not only the behavior but also the logical structure of the hardware that synthesis will produce. Additionally, certain concurrent statements may contain a number of sequentially executing statements. Such flexibility allows the architecture body to be constructed using three possible language styles. Choosing a particular approach does not lim it the design to that one style. In fact, many functional descriptions c an be implemented in more than one style. In general, a combination of two or all three styles is often required to obtain the most compact and clear VHDL code.
2.3.1 Dataflow style Dataflow style architecture models the hardware in terms of the movement of data Over continuous time between combinational logic components such as adders, decoders and primitive logic gates. It describes the register-transfer level behavior of a circuit. The language topics that are most relevant to the dataflow style of architecture include the following: Operators-logical, relational and mathematical; Operator overloading;
Concurrent assignment statements. This style is not appropriate for the modeling of sequential logic. Instead, and as its name suggests, it is best applied in the t he modeling of data driven elements such as an ari thmetic logic unit (ALU) 2.3.2 Behavioral style
The behavioral style architecture contains concurrent statements with sections of sequential statements that describe the outputs of the circuit at a discrete moment in time given particular inputs. While similar language constructs are often found in dataflow and behavioral style architectures, only the latter explicitly exhibit the notions of time and control. This style describes the function of the circuit at the algorithmic level. The aspects of VHDL that are most relevant to behavioral style architectures include the following -Process statements and sensitivity lists; -Sequential statements; -Variables. This architectural style is used to describe both sequential and combinational circuits. Hence, it is a valuable design approach for finite state machines (FSMs) or any control logic.
2.3.3. Structural style A structural style architecture describes the circuit primarily in terms of components. These may be drawn from libraries supplied by ASIC vendors, modules that are purposely produced for this design or general, user defined modules that are stored in a library created for a previous design. The choice of whether whether to use vendor specific or generic/user-defined components is an issue. An appropriate balance must be str uck between the efficiency of the optimization and technology mapping processes and the future flexibility of the design. The architecture body specific which components are contained in a design and how they are interconnected. The main VHDL topics associated with structural st yle architectures include the following: Component declarations and instantiations; Port mapping and signal interface lists; Signals (for interconnection).
A structural architecture is often used at the higher levels of a design to enable a clear distinction to be made between the various functional units and enable a hierarchical approach to be adopted in the design process .
3.LOGIC SYNTHESIS Original Network
Logic Synthesis Logic Optimization
Optimized Network Network
Technology Mapping
Optimized Circuit
Fig. 3.Logic Synthesis
A Logic synthesis software is a CAD tool that c an take a VHDL program and generate a netlist to realize exactly a function. This step consists of two separate phases called LOGIC- OPTIMIZATION and TECHNOLOGY MAPPING. 3.1.LOGIC OPTIMIZATION TECHNIQUES
The way in which VHDL program has been written influences the optimization of a circuit – some constructs synthesize into more efficient structures st ructures than others. All logic synthesis software provides the capability to optimize a circuit for area and speed. Few tools offer power optimization also. Further a design-based optimization called resource sharing is also provided. The following sections outline the optimization techniques: a.Area Optimization
Area optimization is well-studied field in digital electronics. Early techniques addressed simple matrix-based manipulation for Boolean algebra. Later developments incorporated optimizations based on don’t care conditions. The optimization may be – a a general purpose
one or targeted at particular structures s tructures (e.g. programmable logic arrays). Also underlying technology strongly affects the optimization. Where area optimization is more difficult is when particular device technology is used. For example, the Xilinx FPGA operates a number of CLBs. Each CLB can compute up to five input Boolean expressions. If the expression involves six inputs then a second CLB is ne eded. If four inputs are needed then chip area is wasted. b.Speed optimization
Some measure of speed optimization is usually achieved by area optimization, as the area optimization removes redundant circuitry and can shorten the critical path of the circuit. However, further speed optimization can be performed. A common approach is to identify the critical path and to work on reducing that. A si mple optimization might be to replace each component on the critical path with a higher rated (i.e. faster but more power- intensive equivalent) but functionally equivalent component from the library. Other optimization techniques might be to strip out strings of inverters from a design or where, for example, AND gates are slower t han OR gates to attempt to replace all ANDs with ORs.
3.2.TECHNOLOGY MAPPING After logic optimization has produced the optimized network, technology mapping transforms this network into the final circuit. This is done by selecting pieces of the network that can be implemented by one of the available ci rcuit elements, and specifying how these are to be interconnected. The circuit is optimized to reduce a cost function that typically incorporates measures of both area and delay. dela y. Conventional approaches to technology mapping have focused on using circuit elements from a limited set of simple gates, such as a Standard Cell library. The complex logic blocks used in FPGAs presents difficulties for library-based approaches because they can each implement a large number of functions. a.PLACEMENT
The next step after technology mapping is placement of logic blocks. A number of efficient techniques have been developed, which can be easily adapted to use for FPGAs. b.ROUTING
Routing refers to the process of interconnecting the various logic blocks of the FPGA. Some of the terms used in routing are: Pin – Pin – a a logic block input or output.
Connection – Connection – a a pair of logic block pins that are electrically connected.
Net – Net – a a set of logic block pins that are to be electrically electricall y connected. A
net can be divided into one or more connections.
Wire segment – segment – a a straight segment of wire that t hat is used to form a part of a connection. Routing switch – switch – a a programmable switch that is used to electrically connect two wire segments.
Track – Track – a a straight section of wire that that spans the entire width or of a routing channel. channel. A track can be composed of a number of wire segments of various lengths.
Routing channel – channel – the the rectangular area that lies l ies between two rows or two columns of logic blocks. A routing channel contains a number number of tracks.
Because of the complexity involved, the solution of large routing problems as those encountered in FPGAs, usually requires a ”divide and conquer” strategy. This approach involves three steps. 1)Partition the routing resources into routing areas that are appropriate for both the device to be routed and the routing algorithm algorithm to be employed. 2)Use a global router to assign each net to a subnet of the routing areas. The global router does not choose specific wire segments and routing switches for each connection, but rather it creates a new set of routing problems. 3)Use a detailed router to select specific wire segments and routing for each connection, within the restrictions set by the global router.
4.FPGA
Very Large Scale Integration (VLSI) Technology has opened the door to powerful digital circuits at low cost. It has become possible to build chips with more than a million transis tors. Such chips are realized using the full-custom appr oach, where all parts of VLSI circuit are ar e carefully tailored to meet a set of specific requirements. Semi-custom approaches such as standard cells and Mask-Programmed Gate Arrays Arra ys (MPGAs) have provided an easier way of designing and manufacturing Application-Specific Integrated Circuits (ASICs).
Each of these techniques, however, requires extensive manufacturing effort, taking sever al months from beginning to end. This results in a high cost until large volumes are produced. In electronics industry it is vital to reach the market in shortest possible time and also it is important that the financial risk incurred in the development of the new product be limited.
Field-Programmable Gate arrays (FPGAs) have emerged as the ultimate solution to these problems because they provide instant manufacturing and and low cost prototypes.
ALU Before writing the code of an 16 bit ALU first we have to understand what is ALU?. ALU( Arithmetic and Logic unit) is a digital circuit that performs logical and arithmetic operations.ALU is found inside every microprocessor or microcontroller in order to perform logical and arithmetic operations. Logical operation involves all logical functions like ANDing two or more inputs, OR,NOT,NOR etc. Arithmetic operation involves function like addition, subtraction etc, In the given code below the components of ALU are coded individually using VHDL and then combined together in the main ALU code using port map technique. Components of 8 bit ALU-
1) 8-bit adder-substractor - It performs addition and substraction for 8 bit input. A carry line which acts as a signal acts as the deciding bit about what function the circuit will perform. When the input carry signal is “0” the circuit acts as a 8 bit adder and if the input signal is “1” the circuit acts as an adder.
2) 8-bit multiplier-
This circuit is used to multiply 2 8 bit numbers and yield the
output as a 16 bit number.
3) 8-bit ALU- In the final circuit of ALU the components are combined together in a single entity using port map.
Code for 8 bit adder-substractor library ieee; use ieee.std_logic_1164.all; entity FULLADDER is port (a, b, c: in std_logic; sum, carry: out std_logic); end FULLADDER; architecture fulladder_behav of FULLADDER is begin sum <= (a xor b) xor c ; carry <= (a and b) or (c and (a xor b)); end fulladder_behav; --end of full adder
--definition of 4-bit addersub library ieee; use ieee.std_logic_1164.all; entity FOURBITADD is port (a,b: in std_logic_vector(3 downto downto 0); Cin,m : in std_logic; sum4bit: out std_logic_vector (3 downto 0); Cout: out std_logic); end FOURBITADD; architecture fouradder_structure of FOURBITADD is signal C,temp_carry: std_logic_vector (4 downto 0); component FULLADDER
port(a, b, c: in std_logic; sum, carry: out std_logic); end component; begin process(b,m) begin for i in 0 to 3 loop temp_carry(i)<=(b(i) xor m); end loop; end process; FA0: FULLADDER port map (a(0), temp_carry(0), Cin, sum4bit(0), c(1)); FA1: FULLADDER port map (a(1), temp_carry(1), C(1), sum4bit(1), c(2)); FA2: FULLADDER port map (a(2), temp_carry(2), C(2), sum4bit(2), c(3)); FA3: FULLADDER port map (a(3), temp_carry(3), C(3), sum4bit(3), c(4)); Cout <= c(4); end fouradder_structure; --end of 4-bit addersub
--definition of 8-bitaddersub library ieee; use ieee.std_logic_1164.all;
entity EIGHTBITADD is
port ( a,b : in std_logic_vector(7 downto 0); cin ,n: in std_logic; sum8bit : out std_logic_vector(7 downto 0); cout : out std_logic); end EIGHTBITADD;
architecture eightadder_structure of EIGHTBITADD is signal a1,a2,b1,b2,sum1,sum2 : std_logic_vector ( 3 downto 0); signal cin_temp : std_logic;
component FOURBITADD port (a,b: in std_logic_vector(3 downto downto 0); Cin,m : in std_logic; sum4bit: out std_logic_vector (3 downto 0); Cout: out std_logic); end component;
begin process(a,b) begin for i in 0 to 3 loop a1(i) <= a(i); a2(i) <= a(i+4); b1(i) <= b(i); b2(i) <=b(i+4); end loop; end process;
FourAdd1 : FOURBITADD port map (a1,b1,cin,n,sum1,cin_temp); (a1,b1,cin,n,sum1,cin_temp);
FourAdd2 : FOURBITADD port map (a2,b2,cin_temp,n,sum2,cout); (a2,b2,cin_temp,n,sum2,cout);
process(sum1, sum2) begin for i in 0 to 3 loop sum8bit(i) <= sum1(i); sum8bit(i+4) <= sum2(i); end loop; end process; end eightadder_structure; --end of 8-bit addersub
Schematic diagram of 8 bit Adder-Substractor
Code for 8 bit multiplier library ieee; use ieee.std_logic_1164.all; use ieee.std_logic_arith.all; use ieee.std_logic_unsigned.all; entity EIGHTBITMULT is port (a,b:in std_logic_vector (7 downto downto 0); output:out std_logic_vector (15 downto 0)); end EIGHTBITMULT;
architecture mult_behave of EIGHTBITMULT is signal po1,po2,po3,po4,po5,po6,po po1,po2,po3,po4,po5,po6,po7,po8:std_logic_vector(15 7,po8:std_logic_vector(15 downto 0); constant zero:std_logic_vector:="0000000000 zero:std_logic_vector:="0000000000000000"; 000000"; begin
po1<=("00000000" & a) when b(0)='1' else zero; po2<=("0000000" & a & '0') when b(1)='1' else zero; po3<=("000000" & a & "00") "00") when b(2)='1' else zero; po4<=("00000" & a & "000") "000") when b(3)='1' else zero; po5<=("0000" & a & "0000") "0000") when b(4)='1' else zero;
po6<=("000" & a & "00000") "00000") when b(5)='1' else zero; po7<=("00" & a & "000000") "000000") when b(6)='1' else zero; po8<=('0' & a & "0000000") "0000000") when b(7) ='1' else zero; output<=(po8+ po7) + (po6 + po5) + (po4+po3) + (po2+po1); end mult_behave; --end of 8 bit multiplier
Final Code for 8 bit ALU --definition of 8-bit ALU library ieee; use ieee.std_logic_1164.all; use ieee.std_logic_arith.all; use ieee.std_logic_unsigned.all; entity ALU is port(a,b: in std_logic_vector(7 downto downto 0);sel:in integer range 0 to 2; final_output2:out std_logic_vector(15 downto 0); final_output1:out std_logic_vector(7 downto 0)); end ALU;
architecture alu_structure of ALU is signal output_adder,output_sub:std_logic_vector(7 downto 0); signal output_mult: std_logic_vector(15 downto 0); signal COUT,CIN:std_logic;
component EIGHTBITADD port(a,b: in std_logic_vector(7 downto downto 0); cin,n:in std_logic; sum8bit: out std_logic_vector(7 downto 0); cout:out std_logic); end component; component EIGHTBITMULT port (a,b:in std_logic_vector (7 downto downto 0); output:out std_logic_vector (15 downto 0)); end component; begin alu1: EIGHTBITADD port map(a,b,CIN,'0',output_adder,COUT); map(a,b,CIN,'0',output_adder,COUT);
alu2: EIGHTBITADD port map(a,b,CIN,'1',output_sub,COUT); map(a,b,CIN,'1',output_sub,COUT);
alu3:EIGHTBITMULT port map(a,b,output_mult);
process(sel) begin if(sel=0) then final_output1<=output_adder; elsif(sel=1) then final_output1<=output_sub;
else final_output2<=output_mult;
end if; end process; end alu_structure; --end of 8-bit ALU
9. FUTURE PROSPECTS
Results VHDL, One of the biggest job providing sector today, have a very bright future in the coming time,but as the time is passing some other languages are making control over designing, as for example Verilog HDL. Verilog HDL is a little bit better in processing speed than VHDL. But on the other hand VHDL is also better in some fields like, all the three st yles of modeling(Data flow, Behavioral, Structural) in VHDL can be mixed together in the same program i.e. VHDL is more user friendly than Verilog HDL.
But today all over world mainly VHDL is used for chip designing at low level. Till now there are number of drawbacks in VHDL. As main drawback of VHDL is that, there are a number of features in VHDL, that can be simulated but not synthesized. As, final shape to VHDL was given by IEEE, hence IEEE is working on it continuously and trying to make available simulation features to synthesis also.
So if talk about future prospectus of VHDL, it depends on success of IEEE. If IEEE got able to implement such features that there remain no differences between simulation & synthesis with higher processing speed and designing up to more lower level of abstraction, on that day VHDL may take hold over all other HDL,s.
