diff options
| author | Sven Eisenhauer <sven@sven-eisenhauer.net> | 2023-11-10 15:11:48 +0100 |
|---|---|---|
| committer | Sven Eisenhauer <sven@sven-eisenhauer.net> | 2023-11-10 15:11:48 +0100 |
| commit | 33613a85afc4b1481367fbe92a17ee59c240250b (patch) | |
| tree | 670b842326116b376b505ec2263878912fca97e2 /Bachelor/Digitaltechnik 2/SS07/P6/abel.primer.html | |
| download | Studium-33613a85afc4b1481367fbe92a17ee59c240250b.tar.gz Studium-33613a85afc4b1481367fbe92a17ee59c240250b.tar.bz2 | |
Diffstat (limited to 'Bachelor/Digitaltechnik 2/SS07/P6/abel.primer.html')
| -rw-r--r-- | Bachelor/Digitaltechnik 2/SS07/P6/abel.primer.html | 1613 |
1 files changed, 1613 insertions, 0 deletions
diff --git a/Bachelor/Digitaltechnik 2/SS07/P6/abel.primer.html b/Bachelor/Digitaltechnik 2/SS07/P6/abel.primer.html new file mode 100644 index 0000000..14667a2 --- /dev/null +++ b/Bachelor/Digitaltechnik 2/SS07/P6/abel.primer.html @@ -0,0 +1,1613 @@ +<!DOCTYPE html PUBLIC "-//w3c//dtd html 4.0 transitional//en"> +<html><head> + + + <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> + <meta name="GENERATOR" content="Mozilla/4.5 [en]C-NECCK (Win98; U) [Netscape]"> + <meta name="Author" content="J. Van der Spiegel"> + <meta name="Description" content="Overview of ABEL Hardware Description Language"> + <meta name="KeyWords" content="ABEL, HDL, Digial Design"><title>HDL-ABEL Primer</title></head><body> + +<center> +<h1> +<i>University of Pennsylvania<br> +</i><b><font size="+1"><a href="http://www.ee.upenn.edu/">Department of Electrical +Engineering</a></font></b></h1></center> + +<center> +<h1> +ABEL-HDL Primer</h1></center> + +<h2> +<font size="+1">ABEL Primer</font> <a name="Contents"></a><font size="+1">Contents</font></h2> + +<ul> +<li> +1. <a href="#Introduction">Introduction</a></li> + +<li> +2. <a href="#structure">Basic structure of an ABEL source file</a></li> + +<li> +3. <a href="#Declarations">Declarations</a> (module, pin, node, constants)</li> + +<li> +4. <a href="#Numbers">Numbers</a></li> + +<li> +5. <a href="#Directives">Directives</a></li> + +<li> +6. <a href="#Sets">Sets</a></li> + +<ul> +<li> +a. <a href="#Indexing">Indexing or accessing a set</a></li> + +<li> +b. <a href="#Set%20Operations">Set operations</a></li> +</ul> + +<li> +6. <a href="#Operators">Operators</a></li> + +<ul> +<li> +a. <a href="#Logical">Logical Operators</a></li> + +<li> +b. <a href="#Arithmetic">Arithmetic operators</a></li> + +<li> +c. <a href="#Relational">Relational operators</a></li> + +<li> +d. <a href="#Assignment">Assignment operators</a></li> + +<li> +e. <a href="#priority">Operator priority</a></li> +</ul> + +<li> +8. <a href="#Logic">Logic description</a></li> + +<ul> +<li> +a. <a href="#Equations">Equations</a></li> + +<ul> +<li> +<a href="#%22When-Then-Else%22">When-Then-Else statement</a></li> +</ul> + +<li> +b. <a href="#Truth">Truth Table</a></li> + +<li> +c. <a href="#State">State Description</a></li> + +<ul> +<li> +<a href="#State">State_diagram</a></li> + +<li> +<a href="#If-Else-Then">If-Then-Else statement</a></li> + +<li> +<a href="#with">With statement</a></li> + +<li> +<a href="#Case">Case statement</a></li> +</ul> + +<li> +d. <a href="#extensions">Dot extensions</a></li> +</ul> + +<li> +9. <a href="#vectors">Test vectors</a></li> + +<li> +10. <a href="#Property">Device Specific (Property) Statements</a></li> + +<li> +11. <a href="#Miscellaneous">Miscellaneous</a></li> + +<ul> +<li> +a. <a href="#Active-low">Active-low declarations</a></li> +</ul> + +<li> +12. Examples</li> + +<ul> +<li> +<a href="http://www.ese.upenn.edu/rca/software/abel/abel.ex2.html">Moore Finite State Machine</a></li> + +<li> +<a href="http://www.ese.upenn.edu/rca/software/abel/abel.ex1.html">Mealy Finite State Machine</a></li> + +<li> +<a href="http://www.ese.upenn.edu/rca/software/abel/abel.ex3.html">Synchronous Mealy Finite State Machine</a></li> +</ul> + +<li> +<a href="http://www.ee.upenn.edu/rca/software/xilinx/foundation/commistakes.html">Common +Mistakes</a></li> + +<li> +<a href="#References">References</a></li> + +<li> +<a href="#Acknowledgement">Acknowledgement</a></li> +</ul> + +<h2> +1 <a name="Introduction"></a>Introduction</h2> +ABEL (Advanced Boolean Equation Language) allows you to enter behavior-like +descriptions of a logic circuit. ABEL is an industry-standard hardware +description language (HDL) that was developed by Data I/O Corporation for +programmable logic devices (PLD). There are other hardware description +languages such as VHDL and Verilog. ABEL is a simpler language than VHDL +which is capable of describing systems of larger complexity. +<p>ABEL can be used to describe the behavior of a system in a variety of +forms, including logic equations, truth tables, and state diagrams using +C-like statements. The ABEL compiler allows designs to be simulated and +implemented into PLDs such as PALs, CPLDs and FPGAs. +</p><p>Following is a brief overview of some of the features and syntax of +ABEL. It is not intended to be a complete discussion of all its features. +This ABEL primer will get you started with writing ABEL code. In case you +are familiar with ABEL, this write-up can serve as a quick reference of +the most often used commands. For more advanced features, please consult +an ABEL manual or the Xilinx on-line documentation. +</p><h2> +2. Basic <a name="structure"></a>structure of an ABEL source file</h2> +An ABEL source file consists of the following elements. +<ul> +<li> +Header:<b> </b>including Module, Options and Title</li> + +<li> +Declarations: Pin, Constant, Node, Sets, States. Library.</li> + +<li> +Logic Descriptions: Equations, Truth-table, State_diagram</li> + +<li> +Test Vectors: Test_vectors</li> + +<li> +End</li> +</ul> +Keywords (words recognized by ABEL such as commands, e.g. goto, if, then, +module, etc.) are not case sensitive. User-supplied names and labels (identifier) +can be uppercase, lowercase or mixed-case, but <u>are case-sensitive</u> +(input1 is different from Input1). +<p>A typical <a name="template"></a>template is given below. +</p><ul><b><font face="Courier">module</font></b> <i>module name</i> +<p>[<b><font face="Courier">title</font></b> <i>string</i>] +</p><p>[deviceID <b><font face="Courier">device</font></b> deviceType;] +</p><p><i>pin declarations</i> +</p><p><i>other declarations</i> +</p><p><b><font face="Courier">equations</font></b> +</p><p><i>equations</i> +</p><p>[<b><font face="Courier">Test_Vectors</font></b>] +</p><p><i>test vectors</i> +</p><p><b><font face="Courier">end</font></b> <i>module name</i></p></ul> +The following source file is an <a name="example"></a>example of a +half adder: +<ul><font face="Courier"><b>module</b> <i>my_first_circuit;</i></font> +<br><font face="Courier"><b>title</b> <i>'ee200 assignment 1'</i></font> +<br><font face="Courier">EE200XY <b>device</b> 'XC4003E';</font> +<br> +<br> +<p><font face="Courier">" input pins</font> +<br><font face="Courier">A, B pin 3, 5;</font> +</p><p><font face="Courier">" output pins</font> +<br><font face="Courier">SUM, Carry_out pin 15, 18 istype 'com';</font> +</p><p><b><font face="Courier">equations</font></b> +</p><p><font face="Courier">SUM = (A & !B) # (!A & B) ;</font> +<br><font face="Courier">Carry_out = A & B;</font> +</p><p><font face="Courier"><b>end</b> <i>my_first_circuit;</i></font> +<br> +<br> </p></ul> +A brief explanation of the statements follow. For a more detailed discussion +see the following sections or consult an ABEL-HDL manual. When using ABEL +with the Xilinx CAD software, you can use the ABEL wizard which will give +a template of the basic structure and insert some of the keywords. Also +the Language Assistant in the ABEL editor provides on-line help. Go to +the TOOLS ->LANGUAGE ASSISTANT menu. The Language templates give a description +of most ABEL commands, syntax, hierarchy, etc, while the Synthesis template +gives examples of typical circuits. +<h2> +3. <a name="Declarations"></a>Declarations</h2> +<b><font face="Courier">Module</font></b>: each source files starts with +a module statement followed by a module name (identifier). Large source +files often consist of multiple modules with their own title, equations, +end statement, etc. +<p><b><font face="Courier">Title</font></b>: is <u>optional</u> and can +be used to identify the project. The title name must be between single +quotes. The title line is ignored by the compiler but is handy for documentation. +</p><p><b><i>String</i></b>: is a series of ASCII characters enclosed by single +quotes. Strings are used for TITLE, OPTIONS statements, and in pin, node +and attribute declarations. +</p><p><b><font face="Courier">device</font></b>: this declaration is <u>optional</u> +and associates a device identifier with a specific programmable logic device. +The device statement must end with a semicolon. When you are using the +Xilinx CAD system to compile the design, it is better not to put the device +statement in the source file to keep your design independent of the device. +When you create a new project in Xilinx you will specify the device type +(can also be changed in the Project Manager window using the Project Information +button). The format is as follows: +</p><ul>device_id <font face="Courier">device</font> 'real_device'; +<p>Example: MY_DECODER <font face="Courier">device </font>'XC4003E'; +<br> +<br> </p></ul> +<b><i>comments</i></b>: comments can be inserted anywhere in the file and +begin with a double quote and end with another double quote or the end +of the line, whatever comes first. +<p><b><font face="Courier">pin</font></b>: pin declarations tell the compiler +which symbolic names are associated with the devices external pins. Format: +</p><ul><font face="Courier">[!]pin_id pin [pin#] [istype 'attributes'] ;</font></ul> +One can specify more than one pin per line: +<ul>[!]pin_id , pin_id, pin_id <font face="Courier">pin [</font>pin#, [pin#, +[pin#]]] [<font face="Courier">istype</font> 'attributes']; +<p>Example: +</p><p><font face="Courier">IN1, IN2, A1 pin 2, 3, 4;</font> +</p><p><font face="Courier">OUT1 pin 9 istype 'reg';</font> +</p><p><font face="Courier">ENABLE pin;</font> +</p><p><font face="Courier">!Chip_select pin 12 istype 'com';</font> +</p><p><font face="Courier">!S0..!S6 pin istype 'com';</font> +<br> +<br> </p></ul> +You do not need to specify the pin. Pin numbers can be specified later +by using a "user constraint file " when doing the compilation using Xilinx +CAD. This has the advantage that our design is more general and flexible. +The ! indicates an active low (the signal will be inverted). The <font face="Courier">istype +</font>is an optional attribute assignment for a pin such as 'com' to indicate +that the output is a combinational signal or 'reg' for a clocked signal +(registered with a flip flop). This attribute is only for output pins. +<p><b><font face="Courier">node</font></b>: node declarations have the +same format as the pin declaration. Nodes are internal signals which are +not connected to external pins. +</p><ul>Example: +<p>tmp1 <font face="Courier">node [istype</font> 'com']; +<br> +<br> </p></ul> +<b>other declarations</b> allows one to define constants, sets, macros +and expressions which can simplify the program. As an example a constant +declaration has the following format: +<ul>id [, id],... = expr [, expr].. ; +<p>Examples: +</p><p><font face="Courier">A = 21;</font> +</p><p><font face="Courier">C=2*7;</font> +</p><p><font face="Courier">ADDR = [1,0,11];</font> +</p><p><font face="Courier">LARGE = B & C;</font> +</p><p><font face="Courier">D = [D3, D2, D1, D0];</font> +</p><p><font face="Courier">D = [D3..D0];</font></p></ul> +The last two equations are equivalent. The use of ".." is handy to specify +a range. The last example makes use of vector notations. Any time you use +D in an equation, it will refer to the vector [D3, D2, D1. D0]. +<h2> +4. <a name="Numbers"></a>Numbers</h2> +Numbers can be entered in four different bases: binary, octal, decimal +and hexadecimal. The default base is decimal. Use one of the following +symbols (upper or lower case allowed) to specify the base. When no symbol +is specified it is assumed to be in the <u>decimal</u> base. You can change +the default base with the Directive "Radix" as explained in the next section. +<br> +<table border="1"> +<tbody><tr> +<td width="148"><b>BASE NAME</b></td> + +<td width="148"><b>BASE</b></td> + +<td width="148"><b>Symbol.</b></td> +</tr> + +<tr> +<td width="148">Binary</td> + +<td width="148">2</td> + +<td width="148">^b </td> +</tr> + +<tr> +<td width="148">Octal</td> + +<td width="148">8</td> + +<td width="148">^o </td> +</tr> + +<tr> +<td width="148">Decimal</td> + +<td width="148">10</td> + +<td width="148">^d (default) </td> +</tr> + +<tr> +<td width="148">Hexadecimal</td> + +<td width="148">16</td> + +<td width="148">^h </td> +</tr> +</tbody></table> + +<p>Examples: +<br> +<table> +<tbody><tr> +<td width="140"> +<ul><b>Specified in ABEL</b></ul> +</td> + +<td width="140"><b>Decimal Value</b></td> +</tr> + +<tr> +<td width="140"> +<ul>35</ul> +</td> + +<td width="140"> +<ul>35</ul> +</td> +</tr> + +<tr> +<td width="140"> +<ul>^h35</ul> +</td> + +<td width="140"> +<ul>53</ul> +</td> +</tr> + +<tr> +<td width="140"> +<ul>^b101</ul> +</td> + +<td width="140"> +<ul>5</ul> +</td> +</tr> +</tbody></table> + +</p><h2> +5. <a name="Directives"></a>Directives</h2> +Directives allow advanced manipulation of the source file and processing, +and can be placed anywhere needed in a file. +<p><b>@ALTERNATE</b> +<br><i>Syntax</i> +<br>@alternate +<br> +<br>@ALTERNATE enables an alternate set of operators. Using the <a href="#Logical">alternate +operator</a> set precludes use of the ABEL-HDL addition (+), multiplication +(*) and division (/) operators because they represent the OR, AND and NOT +logical operators in the alternate set. The standard operator still work +when @ALTERNATE is in effect. The alternate operators remain in effect +until the @STANDARD directive is used or the end of the module is reached. +</p><p><b>@RADIX</b> +<br><i>Syntax</i> +<br>@radix <i>expr</i> ; +<br><i>Expr: </i> A valid expression that produces the number 2, 8, +10 or 16 to indicate a new default base number. +</p><p>The @Radix directive changes the default base. The default is base 10 +(decimal). The newly-specified default base stays in effect until another +@radix directive is issued or until the end of the module is reached. Note +that when a new @radix is issued, the specification of the new base must +be in the current base format +</p><p>Example +</p><blockquote>@radix 2; +“change default base to binary +<br>… +<br>@radix 1010; “change back from binary +to decimal</blockquote> + +<p><br><b>@STANDARD</b> +<br><i>Syntax</i> +<br>@standard +</p><p>The @standard option resets the operators to the ABEL-HDL standard. +The alternate set is chosen with the @alternative directive. +<br> +<br> +</p><h2> +</h2> + +<h2> +6. <a name="Sets"></a>Sets</h2> +A set is a collection of signals or constants used to reference a group +of signals by one name. A set is very handy to simplify logic expressions. +Any operation that is applied to a set is applied to each element. +<p>A set is a list of constants or signals separated by commas or the range +operator (..) put between square brackets (required). +</p><ul>Examples:</ul> + +<table> +<tbody><tr> +<td width="140"> +<ul><font face="Courier">[D0,D1,D2,D4,D5]</font></ul> +</td> + +<td width="234"></td> +</tr> + +<tr> +<td width="140"> +<ul><font face="Courier">[D0..D6]</font></ul> +</td> + +<td width="234">" incrementing range </td> +</tr> + +<tr> +<td width="140"> +<ul><font face="Courier">[b6..b0]</font></ul> +</td> + +<td width="234">" decrementing range </td> +</tr> + +<tr> +<td width="140"> +<ul><font face="Courier">[D7..D15]</font></ul> +</td> + +<td width="234"></td> +</tr> + +<tr> +<td width="140"> +<ul><font face="Courier">[b1,b2,a0..a3]</font></ul> +</td> + +<td width="234">" range within a larger set</td> +</tr> + +<tr> +<td width="140"> +<ul><font face="Courier">[!S7..!S0]</font></ul> +</td> + +<td width="234">"decrementing range of active -low signals</td> +</tr> +</tbody></table> + +<p>However, the following is not allowed: <tt>[D0, X];</tt> +</p><ul>in which X is also a set <tt>X = [X3..X0]</tt>; Instead one can write: +<p><tt>[D0, X3..X0]</tt>;</p></ul> + +<h3> +a. <a name="Indexing"></a>Indexing or accessing a set</h3> +Indexing allows you to access elements within a set. Use numerical values +to indicate the set index. The number refers to the bit position in the +set starting with 0 for the least significant bit of the set. Here are +some examples. +<ul><font face="Courier">D1 = [D15..D0]; "set declaration</font> +<p><font face="Courier">X2 = [X3..X0]; "set declaration</font> +</p><p><font face="Courier">X2 := D1[3..0]; "makes X2 equal to [D3, D2, D1, +D0]</font> +</p><p><font face="Courier">X2 := D1[7..4]; "makes X2 equal to [D7, D6, D5, +D4]</font></p></ul> +To access one element in the set, use the following syntax: +<ul><font face="Courier">OUT = (X[2] == 1);</font></ul> +Here a comparator operator (==) is used to convert the single-element (X[2]) +into a bit value equivalent to X2. The comparator (==) gives a"1" or "0" +depending if the comparison is True or False. Notice the +<i>difference +between the assignment operator (=) and the equal operator (==)</i>. The +assignment operator is used in equations rather than in expressions. Equations +assign the value of an expression to the output signals. +<h3> +b. <a name="Set Operations"></a>Set operations</h3> +Most of the operations can be applied to a set and are performed on each +element of the set according to the rules of Boolean algebra. Operations +are performed according to the <a href="#priority">operator's priority</a>; +operators with the same priority are performed from left to right (unless +one uses parentheses). Here are a couple of examples. +<br> +<br> +<p>Example 1: +</p><ul><font face="Courier">Signal = [D2,D1,D0]; "declaration of Signal set</font> +<p><font face="Courier">Signal = [1,0,1] & [0,1,1];" results in Signal +being "equal to [0,0,1]</font></p></ul> +Example 2: +<ul><font face="Courier">[A,B] = C & D;</font></ul> +this is equivalent to two statements: +<ul><font face="Courier">A = C & D;</font> +<p><font face="Courier">B = C & D;</font></p></ul> +Example 3: +<ul><font face="Courier">[A1,B1] = [D1,D2] & [C3,C2];</font> +<p><font face="Courier">is equivalent to: [A1,B1] = [D1 & C3, D2 & +C2];</font> +</p><p><font face="Courier">thus A1 = D1 & C3, and B1= D2 & C2.</font></p></ul> +Example 4: +<ul><font face="Courier">X & [A,B,C];</font></ul> +which is equivalent to +<ul>[X&A, X&B, X&C];</ul> +However consider the following expression +<ul><font face="Courier">2 & [A,B,C];</font></ul> +now the number "2" is first converted into a binary representation and +padded with zeros (0010) if necessary. Thus the above equation is equivalent +to: +<ul><font face="Courier">[0 & A, 1 & B, 0 & C];</font></ul> +Example 5: +<ul><font face="Courier">A=[A2,A1,A0]; "set declaration</font> +<p><font face="Courier">B=[B2,B1,B0]; "set declaration</font> +</p><p><font face="Courier">A # B;</font> is equivalent to <font face="Courier">[A2 +# B2, A1 # B1, A0 # B0];</font> +</p><p><font face="Courier">!A;</font> is equivalent to <font face="Courier">[!A2,!A1,!A0];</font></p></ul> +Example 6: +<ul><font face="Courier">[b3,b2,b1,b0] = 2;"is equivalent to b3=0,b2=0,b1=1,b0=0.</font></ul> +The number "2" is converted into binary and padded with zeros (0010). +<p>Example 7: Sets are also handy to specify logic equations. Suppose you +need to specify the equation: +</p><ul><font face="Courier">Chip_Sel = !A7 & A6 & A5;</font></ul> +This can be done using sets. First define a constant Addr set.: +<ul><font face="Courier">Addr = [A7,A6,A5];" declares a constant set Addr.</font></ul> +One can then use the following equation to specify the address: +<ul><font face="Courier">Chip_Sel = Addr == [0,1,1];</font></ul> +which is equivalent to saying: +<ul><font face="Courier">Chip_Sel = !A7 & A6 & A5;</font></ul> +Indeed, if A7=0, A6=1 and A5=1, the expression Addr ==[0,1,1] is true (or +1) and thus Chip_Sel will be true (or 1). Another way to write the same +equation is: +<ul><font face="Courier">Chip_Sel = Addr == 3; " decimal 3 is equal to +011.</font></ul> +The above expressions are very helpful when working with a large number +of variables (ex. a 16 bit address). +<p>Example 8: +</p><p>For the same constants as in the example above, the expression, +</p><ul><font face="Courier">3 & Addr;</font></ul> +which is equivalent to +<ul><font face="Courier">[0,1,1] & [A7,A6,A5]</font> +<p><font face="Courier">[0 & A7, 1 & A6, 1 & A5]</font> +</p><p><font face="Courier">[0,A6,A5].</font></p></ul> +However, the following statement is different: +<ul><font face="Courier">3 & (Addr == 1);</font></ul> +which is equivalent to: +<ul><font face="Courier">3 & (!A7 & !A6 & A5).</font></ul> +However, the relational operator (==) gives <u>only one bit</u>, so that +the rest of the equation evaluates also to one bit, and the "3" is truncated +to "1":. Thus the above equation is equal to: +<ul><font face="Courier">1& !A7 & !A6 & A5.</font></ul> + +<h2> +7. <a name="Operators"></a>Operators</h2> +There are four basic types of operators: logical, arithmetic, relational +and assignment.. +<h3> +a. <a name="Logical"></a>Logical Operators</h3> +The table below gives the logical operators. They are performed bit by +bit. With the <a href="#Directives">@ALTERNATIVE directive</a>, one can +use the alternative set of operators as indicated in the table. +<br> +<table border="1"> +<tbody><tr> +<td width="148"><b>Operator (default)</b></td> + +<td width="148"><b>Description </b></td> + +<td width="148"><b>Alternate operator</b></td> +</tr> + +<tr> +<td width="148">!</td> + +<td width="148">NOT (ones complement)</td> + +<td width="148">/</td> +</tr> + +<tr> +<td width="148">&</td> + +<td width="148">AND</td> + +<td width="148">* </td> +</tr> + +<tr> +<td width="148">#</td> + +<td width="148">OR</td> + +<td width="148">+</td> +</tr> + +<tr> +<td width="148">$</td> + +<td width="148">XOR: exclusive or</td> + +<td width="148">:+: </td> +</tr> + +<tr> +<td width="148">!$</td> + +<td width="148">XNOR: exclusive nor</td> + +<td width="148">:*:</td> +</tr> +</tbody></table> + +<h3> +b. <a name="Arithmetic"></a>Arithmetic operators</h3> +The table below gives the arithmetic operators. Note that the last four +operators are not allowed with sets. The minus sign can have different +meanings: used between two operands it indicates subtraction (or adding +the twos complement), while used with one operator it indicates the twos +complement. +<br> +<table border="1"> +<tbody><tr> +<td width="118"><b>Operator</b></td> + +<td width="121"><b>Example</b></td> + +<td width="204"><b>Description</b></td> +</tr> + +<tr> +<td width="118">-</td> + +<td width="121">-D1</td> + +<td width="204">Twos complement (negation) </td> +</tr> + +<tr> +<td width="118">-</td> + +<td width="121">C1-C2</td> + +<td width="204">Subtraction </td> +</tr> + +<tr> +<td width="118">+</td> + +<td width="121">A+B</td> + +<td width="204">Addition </td> +</tr> + +<tr> +<td colspan="3" width="443"> +<center><b><i>The following operators are not used with sets: </i></b></center> +</td> +</tr> + +<tr> +<td width="118">*</td> + +<td width="121">A*B</td> + +<td width="204">Multiplication </td> +</tr> + +<tr> +<td width="118">/</td> + +<td width="121">A/B</td> + +<td width="204">Unsigned integer division </td> +</tr> + +<tr> +<td width="118">%</td> + +<td width="121">A%B</td> + +<td width="204">Modulus: remainder of A/B </td> +</tr> + +<tr> +<td width="118"><<</td> + +<td width="121">A<<B</td> + +<td width="204">Shift A left by B bits </td> +</tr> + +<tr> +<td width="118">>></td> + +<td width="121">A>>B</td> + +<td width="204">Shift B right by B bits </td> +</tr> +</tbody></table> + +<h3> +c. <a name="Relational"></a>Relational operators</h3> +These operators produce a Boolean value of <u>True (-1) or False (0).</u> +The logical true value of -1 in twos complement is represented by all ones +(i.e.<i>all bits will be ones</i>: ex. for a 16 bit word all bits are one: +-1 is represented by 1111 1111 1111 1111). +<br> +<table border="1"> +<tbody><tr> +<td width="148"><b>Operator </b></td> + +<td width="148"><b>Example</b></td> + +<td width="148"><b>Description </b></td> +</tr> + +<tr> +<td width="148">==</td> + +<td width="148">A==B or 3==5 (false)</td> + +<td width="148">Equal</td> +</tr> + +<tr> +<td width="148">!=</td> + +<td width="148">A!=B or 3 != 5 (true) </td> + +<td width="148">Not equal</td> +</tr> + +<tr> +<td width="148"><</td> + +<td width="148">A<B or 3 < 5 (true) </td> + +<td width="148">Less than</td> +</tr> + +<tr> +<td width="148"><=</td> + +<td width="148">A<=B or 3 <= 5 (true) </td> + +<td width="148">Less than or equal</td> +</tr> + +<tr> +<td width="148">></td> + +<td width="148">A>B or -1 > 5 (true) </td> + +<td width="148">Greater than</td> +</tr> + +<tr> +<td width="148">>=</td> + +<td width="148">A>=B or !0 >= 5 (true) </td> + +<td width="148">Greater than or equal</td> +</tr> +</tbody></table> + +<br> +<p><i><font color="#914a19">Relational operators are unsigned</font></i>. +Be careful: !0 is the one complement of 0 or 11111111 (8 bits data) which +is 255 in unsigned binary. Thus !0 > 9 is true. The expression -1>5 is +true for the same reason. +</p><p>A relational expression can be used whenever a number can be used. The +-1 or 0 will be substituted depending on the logical result. As an example +: +</p><ul>A = B !$ (C == D);</ul> +A will be equal to B if C is equal to D (true or 11111...; B XNOR 1 equals +B), otherwise, A will be equal to the complement of B (if C is not equal +to B (false or 0)). +<h3> +d. <a name="Assignment"></a>Assignment operators</h3> +These operators are used in equations to assign the value of an expression +to output signals. There are two types of assignment operators: combinational +and registered. In a combinational operator the assignment occurs immediately +without any delay. The registered assignment occurs at the next clock pulse +associated with the output. As an example, one can define a flip-flop with +the following statements: +<ul><font face="Courier">Q1 pin istype 'reg';</font> +<p><font face="Courier">Q1 := D;</font></p></ul> +The first statement defines the Q1 flip-flop by using the 'reg' as istype +(registered output). The second statement tells that the output of the +flip-flop will take the value of the D input at the next clock transition. +<br> +<table border="1"> +<tbody><tr> +<td width="221"><b>Operator</b></td> + +<td width="221"><b>Description</b></td> +</tr> + +<tr> +<td width="221">=</td> + +<td width="221">Combinational assignment </td> +</tr> + +<tr> +<td width="221">:=</td> + +<td width="221">Registered assignment</td> +</tr> +</tbody></table> + +<h3> +e. Operator <a name="priority"></a>priority</h3> +The priority of each operator is given in the following table, with priority +1 the highest and 4 the lowest. Operators with the same priority are performed +from left to right. +<br> +<table border="1"> +<tbody><tr> +<td width="148"><b>Priority</b></td> + +<td width="148"><b>Operator</b></td> + +<td width="148"><b>Description </b></td> +</tr> + +<tr> +<td width="148">1</td> + +<td width="148">-</td> + +<td width="148">Negation (twos complement) </td> +</tr> + +<tr> +<td width="148">1</td> + +<td width="148">!</td> + +<td width="148">NOT </td> +</tr> + +<tr> +<td width="148">2</td> + +<td width="148">&</td> + +<td width="148">AND </td> +</tr> + +<tr> +<td width="148">2</td> + +<td width="148"><<</td> + +<td width="148">shift left </td> +</tr> + +<tr> +<td width="148">2</td> + +<td width="148">>></td> + +<td width="148">shift right </td> +</tr> + +<tr> +<td width="148">2</td> + +<td width="148">*</td> + +<td width="148">multiply </td> +</tr> + +<tr> +<td width="148">2</td> + +<td width="148">/</td> + +<td width="148">unsigned division </td> +</tr> + +<tr> +<td width="148">2</td> + +<td width="148">%</td> + +<td width="148">modulus </td> +</tr> + +<tr> +<td width="148">3</td> + +<td width="148">+</td> + +<td width="148">add </td> +</tr> + +<tr> +<td width="148">3</td> + +<td width="148">-</td> + +<td width="148">subtract </td> +</tr> + +<tr> +<td width="148">3</td> + +<td width="148">#</td> + +<td width="148">OR</td> +</tr> + +<tr> +<td width="148">3</td> + +<td width="148">$</td> + +<td width="148">XOR </td> +</tr> + +<tr> +<td width="148">3</td> + +<td width="148">!$</td> + +<td width="148">XNOR </td> +</tr> + +<tr> +<td width="148">4</td> + +<td width="148">==</td> + +<td width="148">equal </td> +</tr> + +<tr> +<td width="148">4</td> + +<td width="148">!=</td> + +<td width="148">not equal </td> +</tr> + +<tr> +<td width="148">4</td> + +<td width="148"><</td> + +<td width="148">less then </td> +</tr> + +<tr> +<td width="148">4</td> + +<td width="148"><=</td> + +<td width="148">less then or equal </td> +</tr> + +<tr> +<td width="148">4</td> + +<td width="148">></td> + +<td width="148">greater than </td> +</tr> + +<tr> +<td width="148">4</td> + +<td width="148">>=</td> + +<td width="148">greater than or equal </td> +</tr> +</tbody></table> + +<h2> +8. <a name="Logic"></a>Logic description</h2> +A logic design can be described in the following way. +<ul> +<li> +Equations</li> + +<li> +Truth Table</li> + +<li> +State Description</li> +</ul> + +<h3> +a. <a name="Equations"></a>Equations</h3> +Use the keyword <b><font face="Courier">equations</font></b> to begin the +logic descriptions. Equations specify logic expressions using the <a href="#Operators">operators</a> +described above, or "When-Then-Else" statement. +<p>The <a name=""When-Then-Else""></a>"When-Then-Else" statement +is used in equations to describe a logic function. (Note: "<a href="#If-Else-Then">If +-Then-Else</a>" is used in the State-diagram section to describe state +progression). +</p><p>The format of the "When-Then-Else" statement is as follows: +</p><ul><font face="Courier">WHEN condition THEN element=expression;</font> +<p><font face="Courier">ELSE equation;</font> +</p><p><font face="Courier">or</font> +</p><p><font face="Courier">WHEN condition THEN equation;</font></p></ul> +Examples of equations: +<ul><font face="Courier">SUM = (A & !B) # (!A & B) ;</font> +<br><font face="Courier">A0 := EN & !D1 & D3 & !D7;</font> +<p><font face="Courier">WHEN (A == B) THEN D1_out = A1;</font> +</p><ul> +<br><font face="Courier">ELSE WHEN (A == C) THEN D1_out = A0;</font></ul> +<font face="Courier">WHEN (A>B) THEN { X1 :=D1; X2 :=D2; }</font></ul> +One can use the braces { } to group sections together in blocks. The text +in a block can be on one line or span many lines. Blocks are used in equations, +state diagrams and directives. +<h3> +b. <a name="Truth"></a>Truth Tables</h3> +The keyword <b>is <font face="Courier">truth-table </font></b>and the syntax +is +<ul><font face="Courier">TRUTH_TABLE ( in_ids -> out_ids )</font> +<p><font face="Courier">inputs -> outputs ;</font></p></ul> +or +<ul><font face="Courier">TRUTH_TABLE ( in_ids :> reg_ids )</font> +<p><font face="Courier">inputs :> reg_outs ;</font></p></ul> +or +<ul><font face="Courier">TRUTH_TABLE</font> +<p><font face="Courier">( in_ids :> reg_ids -> out_ids )</font> +</p><p><font face="Courier">inputs :> reg_outs -> outputs ;</font> +<br> </p></ul> +in which "->" is for combinational output and ":>" for registered output. +The first line of a truth table (between parentheses) defines the inputs +and the output signals. The following lines gives the values of the inputs +and outputs. Each line must end with a semicolon. The inputs and outputs +can be single signals or sets. When sets are used as inputs or outputs, +use the normal set notation, i.e. signals surrounded by square brackets +and separated by commans. A don't care is represented by a ".X.". +<p>Example 1: half adder +</p><ul><tt><font face="Courier">TRUTH_TABLE ( [ A, B] -> [Sum, Carry_out] +)</font></tt> +<ul> +<ul><tt><font face="Courier">[ 0, 0 ] -> [0, 0 ] ;</font></tt> +<br><tt><font face="Courier">[ 0, 1 ] -> [1, 0 ] ;</font></tt> +<br><tt><font face="Courier">[ 1, 0 ] -> [1, 0 ] ;</font></tt> +<br><tt><font face="Courier">[ 1, 1 ] -> [1, 1 ] ;</font></tt></ul> +</ul> +</ul> +However, if one defines a set IN = [A,B]; and OUT = [Sum, Carry_out]; the +truth table becomes simpler: +<ul><tt><font face="Courier">TRUTH_TABLE (IN -> OUT )</font></tt> +<ul> +<ul><tt><font face="Courier">0 -> 0;</font></tt> +<br><tt><font face="Courier">1 -> 2;</font></tt> +<br><tt><font face="Courier">2 -> 2;</font></tt> +<br><tt><font face="Courier">3 -> 3;</font></tt></ul> +</ul> +</ul> +Example 2: An excluse OR with two intputs and one enable (EN). This example +illustrates the use of don't cares (.X.) +<ul><tt><font face="Courier">TRUTH_TABLE ([EN, A, B] -> OUT )</font></tt> +<ul> +<ul><tt><font face="Courier">[ 0, .X.,.X.] -> .X. ;</font></tt> +<br><tt><font face="Courier">[ 1, 0 , 0 ] -> 0 ;</font></tt> +<br><tt><font face="Courier">[ 1, 0 , 1 ] -> 1 ;</font></tt> +<br><tt><font face="Courier">[ 1, 1 , 0 ] -> 1 ;</font></tt> +<br><tt><font face="Courier">[ 1, 1 , 1 ] -> 0 ;</font></tt></ul> +</ul> +</ul> +Example 3: (see Example in R. Katz, section 7.2.1 and table 7.14) +<p>Truth tables can also be used to define sequential machines. Lets implement +a three-bit up counter which counts from 000, 001, to 111 and back to 000. +Lets call QA, QB and QC the outputs of the flip-flops. In addition, we +will generate an output OUT whenever the counter reaches the state 111. +We will also reset the counter to the state 000 when the reset signal is +high. +</p><ul><tt><font face="Courier">MODULE CNT3;</font></tt> +<p><tt><font face="Courier">CLOCK pin;</font> <font face="Courier">" input +signal</font></tt> +<br><tt><font face="Courier">RESET . pin;</font> <font face="Courier">" +input signal</font></tt> +<br><tt><font face="Courier">OUT pin istype 'com';</font> <font face="Courier">" +output signal (combinational)</font></tt> +<br><tt><font face="Courier">QC,QB,QA pin istype 'reg'; " output signal +(registered)</font></tt> +</p><p><tt><font face="Courier">[QC,QB,QA].CLK = CLOCK; "FF clocked on the +CLOCK input</font></tt> +<br><tt><font face="Courier">[QC,QB,QA].AR = RESET; "asynchronous reset +by RESET</font></tt> +</p><p><tt><font face="Courier">TRUTH_TABLE ) [QC, QB, QA] :> [QC,QB,QA] -> +OUT)</font></tt> +</p><ul> +<ul> +<ul><tt><font face="Courier">[ 0 0 0 ] :> [ 0 0 1 ] -> 0;</font></tt> +<br><tt><font face="Courier">[ 0 0 1 ] :> [ 0 1 0 ] -> 0;</font></tt> +<br><tt><font face="Courier">[ 0 1 0 ] :> [ 0 1 1 ] -> 0;</font></tt> +<br><tt><font face="Courier">[ 0 1 1 ] :> [ 1 0 0 ] -> 0;</font></tt> +<br><tt><font face="Courier">[ 1 0 0 ] :> [ 1 0 1 ] -> 0;</font></tt> +<br><tt><font face="Courier">[ 1 0 1 ] :> [ 1 1 0 ] -> 0;</font></tt> +<br><tt><font face="Courier">[ 1 1 0 ] :> [ 1 1 1 ] -> 0;</font></tt> +<br><tt><font face="Courier">[ 1 1 1 ] :> [ 0 0 0 ] -> 1;</font></tt></ul> +</ul> +</ul> +<tt><font face="Courier">END CNT3;</font></tt></ul> +For the use of .DOT extensions (.CLK and .AR) see section 7d. +<h3> +c. <a name="State"></a>State Description</h3> +The State_diagram section contains the state description for the logic +design. This section uses the State_diagram syntax and the "If-Then-Else", +"Goto", "Case" and "With" statements. Usually one declares symbolic state +names in the Declaration section, which makes reading the program often +easier. +<p><i>State declaration</i> (in the declaration section) syntax: +</p><ul><font face="Courier">state_id [, state_id ...] STATE ;</font></ul> +As an example: <tt><font size="+1">SREG = [Q1, Q2];</font></tt> associates +the state name SREG with the state defined by Q1 and Q2. +<p><u>The <b>syntax </b>for <i>State_diagram</i> is as follows:</u> +</p><ul><font face="Courier">State_diagram <i>state_reg</i></font> +<p><font face="Courier">STATE <i>state_value</i> : [equation;]</font> +</p><p><font face="Courier">[equation;]</font> +</p><p><font face="Courier">:</font> +</p><p><font face="Courier">:</font> +</p><p><font face="Courier">trans_stmt ; ...</font></p></ul> +The keyword <b><font face="Courier">state_diagram </font></b>indicates +the beginning of a state machine description. +<p>The STATE keyword and following statements describe one state of the +state diagram and includes a state value or symbolic state name, state +transition statement and an optional output equation. In the above syntax, +</p><ul> +<li> +<font face="Courier">state_reg</font>: is an identifier that defines the +signals that determine the state of the machine. This can be a symbolic +state register that has been declared earlier in the declaration section.</li> + +<li> +<font face="Courier">state_value</font>: can be an expression, a value +or a symbolic state name of the current state;</li> + +<li> +<font face="Courier">equation</font> : an equation that defines the state +machine outputs</li> + +<li> +<font face="Courier">trans_stmt</font>: the "If-Then-Else", CASE or GOTO +statements to defines the next state, followed with optional WITH transition +equations.</li> +</ul> +<a name="If-Else-Then"></a><u><b><font face="Courier">If-Then-Else</font></b> +statement:</u> +<p>This statement is used in the state_diagram section to describe the +next state and to specify mutually exclusive transition conditions. +</p><p>Syntax<font face="Courier">:</font> +</p><ul><font face="Courier">IF expression THEN state_exp</font> +<p><font face="Courier">[ELSE state_exp] ;</font></p></ul> +In which state-exp can be a logic expression or a symbolic state name. +Note that the "IF-Then-Else" statement can only be used in the state_diagram +section (use the <a href="#%22When-Then-Else%22">"When-If-Then"</a> +to describe logic functions". The ELSE clause is optional. The IF-Then-Else +statements can be nexted with <tt><font size="+1">Goto, Case</font></tt> +and <tt><font size="+1">With</font></tt> statements. +<p>Example (after R. Katz): +</p><p>in the declaration section we define first the state registers: +</p><ul><font face="Courier">SREG = [Q1, Q0];</font> "definition of state registers +<br><font face="Courier">S0 = [0, 0];</font> +<br><font face="Courier">S1 = [1, 1];</font> +<p><font face="Courier">state_diagram SREG</font> +<br><font face="Courier">state S0:</font> OUT1 = 1; +</p><ul> +<br><font face="Courier">if A then S1</font> +<br><font face="Courier">else S0;</font></ul> +<font face="Courier">state S1: OUT2 =1;</font> +<ul> +<br><font face="Courier">if A then S0</font> +<br><font face="Courier">else S1;</font></ul> +</ul> +"If-Then-Else" statements can be nested as in the following example (after +Wakerly). We assume that one has defined the registers and states in the +declaration section. +<ul><font face="Courier">state_diagram MAK</font> +<p><font face="Courier">state INIT: if RESET then INIT else LOOK;</font> +</p><p><font face="Courier">state LOOK: if REST than INIT</font> +</p><ul><font face="Courier">else if (X == LASTX) then OK</font> +<br><font face="Courier">else LOOK;</font></ul> +<font face="Courier">state OK: if RESET than INIT</font> +<ul><font face="Courier">else if Y then OK</font> +<br><font face="Courier">else if (X == LASTX) then OK</font> +<br><font face="Courier">else LOOK;</font></ul> +<font face="Courier">state OK: goto INIT;</font> +<br> </ul> +<u>"<a name="with"></a><b><font face="Courier">with</font></b>" statement:</u> +<p>Syntax: +</p><ul><font face="Courier">trans_stmt state_exp WITH equation</font> +<br><font face="Courier">[equation ] ... ;</font></ul> +in which trans_stmt can be "If-then-else", 'Goto" or a "Case" statement. +<br><tt>state_exp</tt>: is the next state, and <tt>equation</tt> is an +equation for the machine outputs. +<p>This statement can be used with the "If-Then-Else", "Goto" or "Case" +statements in place of a simple state expression. The "With" statement +allows the output equations to be written in terms of transitions. +</p><p>Example 1: +</p><ul><tt><font size="+1">if X#Y==1 then S1 with Z=1 else S2;</font></tt></ul> +In the above example, the output Z will be asserted as soon as the expression +after the if statement evaluates to a logic 1 (or TRUE). The expression +after the "With" keyword can be an equation that will be evaluated as soon +as the if condition is true as in example 2: +<p>Example 2: +</p><ul><tt><font size="+1">if X&!Y then S3 with Z=X#Y else S2 with Z=Y;</font></tt></ul> +The "With" statement is also useful to describe output behavior of registered +outputs, since registered outputs would lag by one clock cycle. It allows +one also for instance to specify that a registered output should have a +specific value after a particular transition. As an example [1], +<p>Example 3[1]: +</p><p><font face="Courier">state S1:</font> +</p><ul><font face="Courier">if RST then S2 with { OUT1 := 1;</font> +<ul><font face="Courier">Error-Adrs := ADDRESS; }</font></ul> +<font face="Courier">else if (ADDRESS <= ^hC101)</font> +<ul><font face="Courier">then S4</font> +<br><font face="Courier">else S1;</font></ul> +</ul> +Notice that one can use curly braces to control a group of outputs and +equations after the With keyword as in the example above. +<p>Example 3: +</p><ul><font face="Courier">state S1: if (A & B) then S2 with TK = 1</font> +<ul><font face="Courier">else S0 with TK = 0 ;</font></ul> +</ul> +You have to be aware of the timing when using the "With " statement with +combinational or asynchronous outputs (as in a Mealy machine). A Mealy +machine changes its outputs as soon as the input changes. This may cause +the output to change too quickly resulting in glitches. The outputs of +a Mealy machine will be valid at the end of a state time (i.e. just before +the clock transition). In this respect a Moore output (with synhronous +outputs) is less prone to timing errors. An example of a <a href="http://www.ese.upenn.edu/rca/software/abel/abel.ex1.html">Mealy +machine</a> and a <a href="http://www.ese.upenn.edu/rca/software/abel/abel.ex2.html">Moore machine</a> is available. +<p><a name="Case"></a><b><u><font face="Courier">Case</font> </u></b><u>statement</u> +</p><p>Syntax: +</p><ul><font face="Courier">CASE expression : state_exp;</font> +<p><font face="Courier">[ expression : state_exp; ]</font> +<br><font face="Courier">:</font> +</p><p><font face="Courier">ENDCASE ;</font></p></ul> +expression is any valid ABEL expression and state_exp is an expression +that indicates the next state (optionally followed by WITH statement). +<p>Example: +</p><ul><font face="Courier">State S0:</font> +<ul><font face="Courier">case ( A == 0) : S1;</font> +<ul><font face="Courier">( A == 1) : S0;</font></ul> +</ul> +<font face="Courier">endcase;</font></ul> +The case statement is used to list a sequence of mutually-exclusive transition +conditions and corresponding next states. The CASE statement conditions +must be mutually exclusive (no two transition conditions can be true at +the same time) or the resulting next state is unpredictable. +<h3> +d. Dot <a name="extensions"></a>extensions</h3> +One can use dot extensions to more precisely describe the behavior of the +circuit. The signal extensions are very handy and provide a means to refer +specifically to internal signals and nodes associated with a primary signal. +<p>The syntax is +</p><ul><font face="Courier">signal_name.ext</font></ul> +Some of the dot extensions are given in the following table. Extensions +are not case sensitive. Some dot extensions are general purpose (also called +architecture independent or pin-to-pin) and can used with a variety of +device architectures. Other dot extensions are used for specific classes +of device architectures and are called architecture-dependent or detailed +dot extensions. In general, you can use either dot extensions. +<br> +<center><table border="1"> +<tbody><tr> +<td width="109"><b>Dot extension</b></td> + +<td width="270"><b>Description</b></td> +</tr> + +<tr> +<td colspan="2" width="379"><b>Architecture independent or pin-to-pin extensions</b></td> +</tr> + +<tr> +<td width="109">.ACLR</td> + +<td width="270">Asynchronous register reset </td> +</tr> + +<tr> +<td width="109">.ASET</td> + +<td width="270">Asynchronous register preset </td> +</tr> + +<tr> +<td width="109">.CLK</td> + +<td width="270">Clock input to an edge-triggered flip-flop </td> +</tr> + +<tr> +<td width="109">.CLR</td> + +<td width="270">Synchronous register reset </td> +</tr> + +<tr> +<td width="109">.COM</td> + +<td width="270">Cominbational feedback from flip-flop data input </td> +</tr> + +<tr> +<td width="109">.FG</td> + +<td width="270">Register feedback</td> +</tr> + +<tr> +<td width="109">.OE</td> + +<td width="270">Output enable</td> +</tr> + +<tr> +<td width="109">.PIN</td> + +<td width="270">Pin feedback</td> +</tr> + +<tr> +<td width="109">.SET</td> + +<td width="270">Synchronous register preset </td> +</tr> + +<tr> +<td colspan="2" width="379"><b>Device Specific extensions (architecture +dependent)</b></td> +</tr> + +<tr> +<td width="109">.D</td> + +<td width="270">Data input to a D Flip flop </td> +</tr> + +<tr> +<td width="109">.J</td> + +<td width="270">J input to a JK flip-flop </td> +</tr> + +<tr> +<td width="109">.K</td> + +<td width="270">K input to a JK flip-flop </td> +</tr> + +<tr> +<td width="109">.S</td> + +<td width="270">S input to a SR flip-flop </td> +</tr> + +<tr> +<td width="109">.R</td> + +<td width="270">R input to a SR flip-flop </td> +</tr> + +<tr> +<td width="109">.T</td> + +<td width="270">T input to a T flip-flop </td> +</tr> + +<tr> +<td width="109">.Q</td> + +<td width="270">Register feedback</td> +</tr> + +<tr> +<td width="109">.PR</td> + +<td width="270">Register preset</td> +</tr> + +<tr> +<td width="109">.RE</td> + +<td width="270">Register reset</td> +</tr> + +<tr> +<td width="109">.AP</td> + +<td width="270">Asynchronous register preset </td> +</tr> + +<tr> +<td width="109">.AR</td> + +<td width="270">Asynchronous register reset </td> +</tr> + +<tr> +<td width="109">.SP</td> + +<td width="270">Synchronous register preset </td> +</tr> + +<tr> +<td width="109">.SR</td> + +<td width="270">Synchronous register reset </td> +</tr> +</tbody></table></center> + +<p>The figure below illustrates some of the extensions. +</p><p><br> +<br> +<br> +</p><center> +<p><img src="abel.primer-Dateien/abel.gif" height="226" width="621"> +</p><p>Figure 1: Illustration of DOT extensions for: (a) an architecture independent +(pin-to-pin) and (b) arhitecture dependent D-type (or T-type) Flip Flop +Architecture</p></center> + +<p>Example 1: +</p><ul><font face="Courier">[S6..S0].OE = ACTIVE;</font></ul> +which accesses the tri state control signal of the output buffers of the +signals S6..S0. When ACTIVE is high, the signals will be enabled, otherwise +a high Z output is generated. +<p>Example 2: +</p><ul><font face="Courier">Q.AR = reset;</font> +<p><font face="Courier">[Z.ar, Q.ar] = reset;</font></p></ul> +which resets to output of the registers (flip flops) to zero when reset +is high. +<h2> +9. Test <a name="vectors"></a>vectors</h2> +Test vectors are optional and provide a means to verify the correct operation +of a state machine. The vectors specify the expected logical operation +of a logic device by explicitly giving the outputs as a function of the +inputs. +<p>Syntax: +</p><ul> +<br><font face="Courier">Test_vectors [note]</font> +<p><font face="Courier">(input [, input ].. -> output [, output ] .. )</font> +</p><p><font face="Courier">[invalues -> outvalues ; ]</font> +<br><font face="Courier">:</font> +<br><font face="Courier">:</font></p></ul> +Example: +<ul><font face="Courier">Test_vectors</font> +<p><font face="Courier">( [A, B] -> [Sum, Carry] )</font> +</p><p><font face="Courier">[ 0, 0 ] -> [0, 0];</font> +<br><font face="Courier">[ 0, 1 ] -> [1, 0];</font> +<br><font face="Courier">[ 1, 0 ] -> [1, 0];</font> +<br><font face="Courier">[ 1, 1 ] -> [1, 1];</font></p></ul> +One can also specify the values for the set with numeric constants as shown +below. +<ul><font face="Courier">Test_vectors</font> +<p><font face="Courier">( [A, B] -> [Sum, Carry] )</font> +</p><ul><font face="Courier">0 -> 0;</font> +<br><font face="Courier">1 -> 2;</font> +<br><font face="Courier">2 -> 2;</font> +<br><font face="Courier">3 -> 3;</font></ul> +</ul> +Don't cares (.X.), clock inputs (.C.) as well as symbolic constants are +allowed, as shown in the following example. +<br> +<br> +<ul><font face="Courier">test_vectors</font> +<p><font face="Courier">( [CLK, RESET, A, B ] -> [ Y0, Y1, Y3] )</font> +</p><ul><font face="Courier">[.X., 1, .X.,.X.]->[ S0, 0, 0];</font> +<br><font face="Courier">[.C., 0, 0, 1 ] -> [ S0, 0, 0];</font> +<br><font face="Courier">[.C., 1, 1, 0 ] -> [ S0, 0, 1];</font></ul> +</ul> + +<h2> +10. <a name="Property"></a>Property Statements</h2> +ABEL allows to give device specific statements using the property statement. +This statement will be passed to the "Fitter" program during compilation. +For the CPLD devices these properties include +<ul> +<li> +Slew rates</li> + +<li> +Logic optimizations</li> + +<li> +Logic placement</li> + +<li> +Power settings</li> + +<li> +Preload values</li> +</ul> + +<h2> +11. <a name="Miscellaneous"></a>Miscellaneous</h2> + +<h3> +a. <a name="Active-low"></a>Active-low declarations</h3> +Active low signals are defined with a "!" operator, as shown below, +<ul><font face="Courier">!OUT pin istype 'com' ;</font></ul> +When this signal is used in a subsequent design description, it will be +automatically complemented. As an example consider the following description, +<ul><font face="Courier">module EXAMPLE</font> +<p><font face="Courier">A, B pin ;</font> +</p><p><font face="Courier">!OUT pin istype 'com';</font> +</p><p><font face="Courier">equations</font> +</p><p><font face="Courier">OUT = A & !B # !A & B ;</font> +</p><p><font face="Courier">end</font></p></ul> +In this example, the signal OUT is an XOR of A and B, i.e. OUT will be +"1" (High, or ON) when only one of the inputs is "1", otherwise OUT is +"0". However, the output pin is defined as !OUT , i.e. as an active-low +signal, which means that the pin will go low "0" (Active-low or ON) when +only one of the two inputs are "1". One could have obtained the same result +by inverting the signal in the equations and declaring the pin to be OUT, +as is shown in the following example. This is called <i>explicit pin-to-pin +active-low</i> (because one uses active-low signals in the equations). +<ul><font face="Courier">module EXAMPLE</font> +<p><font face="Courier">A, B pin ;</font> +</p><p><font face="Courier">OUT pin istype 'com';</font> +</p><p><font face="Courier">equations</font> +</p><p><font face="Courier">!OUT = A & !B # !A & B ;</font> +</p><p><font face="Courier">end</font></p></ul> +Active low can be specified for a set as well. As an example lets define +the sets A,B and C. +<ul><font face="Courier">A = [A2,A1,A0]; "set declaration</font> +<br><font face="Courier">B = [B2,B1.B0]; "set declaration</font> +<br><font face="Courier">X = [X2,X1.X0]; "set declaration</font> +<p><font face="Courier">!X = A & !B # !A & B;</font></p></ul> +The last equation is equivalent to writing +<ul>!X0 = <font face="Courier">A0 & !B0 # !A0 & B0;</font> +<br>!X1 = <font face="Courier">A1 & !B1 # !A1 & B1;</font> +<br>!X2 = <font face="Courier">A2 & !B2 # !A2 & B2;</font></ul> + +<h2> +<a name="References"></a>References</h2> + +<ul> +<li> +1. Xilinx-ABEL Design Software Reference Manual, Data I/O Corp., 1993.</li> + +<li> +2. The ISP Application Guide and CPLD Data Book, Application Note XAPP075, +1997, Xilinx, San Jose, CA, 1997.</li> + +<li> +3. D. Van den Bout, "Xilinx FPGA Student Manual", Prentice Hall, Englewoods +Cliff, NJ, 1997.</li> + +<li> +4. R. Katz, "Contemporary Logic Design," Benjamin/Cummings Publ. Comp., +Redwood City, CA, 1995.</li> + +<li> +5. J. Wakerly, "Digital Design," Prentice Hall, Englewoods Cliff, NJ, 1993.</li> + +<li> +6. Xilinx Foundation Series, On-line documentation, Xilinx, San Jose, CA.</li> +</ul> + +<h2> +<a name="Acknowledgement"></a>Acknowledgement</h2> +The support of Mr. Jason Feinsmith and the <b><i><font size="+1"><a href="http://www.xilinx.com/">Xilinx</a></font></i></b>Corporation +is acknowledged for providing the XILINX Foundation M1(TM) Software and +FPGA Demoboards for educational purposes. +<h2> + +<hr size="4" width="100%"></h2> +<font size="-1">Back to ABEL Primer <a href="#Contents">Contents</a> | To +to <a href="http://www.ee.upenn.edu/rca/software/xilinx/foundation/commistakes.html">Common +Mistakes</a> list | Go to the <a href="http://www.ee.upenn.edu/rca">EE +Undergraduate Lab</a> Homepage | Go to <a href="http://www.ee.upenn.edu/rca/software/xilinx.html">Xilinx +Lab Tutorial</a> Homepage | Go to the <a href="http://www.ee.upenn.edu/rca/software/xilinx/foundation/foundation.sch1.html">Foundation +Tutorial</a> page | Go to <a href="http://www.seas.upenn.edu/%7Eee200/">EE200 +</a>or +<a href="http://www.seas.upenn.edu/%7Eee200/lab/lab.html">EE200 Lab</a> Homepage +| .</font> +<hr width="100%"> +<p>Created by <a href="http://www.ee.upenn.edu/%7Ejan">J. Van der Spiegel</a>, +<jan@ee.upenn.edu> Sept. 26, 1997; Updated August 13, 1999 +</p></body></html>
\ No newline at end of file |