1. PROCEDURAL PROGRAMMING
Contents
1. OR XOR and INVERT HEX 3A OF AND bitwise logical AND giving 0A HEX 3A OF OR bitwise logical OR giving 3F HEX 3A OF XOR bitwise exclusive OR giving 35 HEX 3A INVERT bitwise inversion giving FFCS Notice that INVERT takes only one parameter Also all of the logical operations act on 16 bit values If you try to print the result with the operator you may be surprised HEX 3A INVERT prints 3B This is because FFCS is a negative number in two s complement notation and prints signed numbers To print unsigned numbers use the U unsigned print operator HEX 3A INVERT U prints FFCS The logical operations are is especially useful when you re working with I O when you need to act on specific bits For example suppose you need to read the low 4 bits of port A as a hex number from 0 to F You could read the four bits individually and write some code to merge them into a 4 bit value But it s much easier to say PORTA GETBYTE HEX OF AND which reads all 8 bits of the port and then masks off the unwanted bits What if you need the high 4 bits of the port It s only an 8 bit port so the mask should be FO instead of OF You then need to shift the bits down four places PORTA GETBYTE HEX FO AND 2 2 2 2 The 2 operator two divide gives a one bit right shift For integers this is equivalent to dividing by two hence the name Applying it four times gives a total of four shifts to2. else But there is one situation when you might want to read or write a known memory location The input and output of the IsoPod s DSP56F805 CPU is memory mapped This means that instead of accessing the peripherals with IN and OUT instructions you use normal memory fetch and store instructions So and give you access to the complete I O capability of the IsoPod processor For example the Port A Data Register is located at address 0FB1 The Port A Data Direction Register DDR is located at address 0FB2 and the Peripheral Enable Register PER is at 0FB3 Writing zero to both the DDR and PER will make Port A an input port and you can then read the Data Register to read the eight input pins HEX 0O OFB2 0O OFB3 OFB1 This is not for the inexperienced user There are no restrictions on and so it s quite possible for you to lock up the IsoPod completely by writing the wrong value to the wrong location You should refer to Motorola s DSP56F801 803 805 807 16 Bit Digital Signal Processor User s Manual for a complete description of the on chip I O of the DSP56F805 processor and its memory addresses Besides isn t this easier PORTA GETBYTE ServoPod owners note The ServoPod uses the DSP56F807 processor which has different I O addresses from the IsoPod s DSP56F805 1 7 Logical Operations IsoMax also lets you perform logical operations on 16 bit values The four basic arithmetic operators are AND
3. the right The corresponding left shift operator is 2 two times 1 8 Adding New Definitions Much of the time you will be using the IsoMax operations interactively as we have seen above or in the CONDITION or CAUSES phrases of an IsoMax state machine What you have learned so far is sufficient for these uses although you ll probably want to look at the IsoMax Glossary in Section 18 to see the full range of operations which are available to you But there may come a time when you want to create a procedural subroutine This might be because e There s a complex function you perform frequently and you re tired of typing it all the time or e You want to write a computer program in the traditional sense and commit it to the IsoPod s memory In either case you do this by adding a new word to the IsoMax dictionary This new word will contain your complex function or your application program Just like with an English dictionary you add a new word by first giving the name of the new word and then defining that word using only words which are already known IsoMax marks the start and end of a new definition with and as follows name of new word definition The spaces after and before are required The name of the new word can be any combination of up to 31 non blank characters just like VARIABLE names Let s go back to our recent example and assume that a 4 bit DIP switch is c
4. use your new word U gt There is no distinction between built in operators and user functions you can add new comparison arithmetic and logical operators as you please For this reason IsoMax is called an extensible language you can add new language elements at any time 1 12 Word list The complete word list is found in the IsoMax Glossary Section 18 at the end of this manual
5. 1 PROCEDURAL PROGRAMMING The Finite State Machine portions of IsoMax are now covered What remains to be discussed is the procedural portions of the conditions and actions The IsoMax procedural language is very similar to the programming language Forth There are some significant changes because of the Finite State Machine functions the Object Oriented functions and the architecture of the IsoPod processor which has separate data and program memories and can t address memory as bytes But if you are familiar with Forth and particularly Max Forth you can skip most of this section 1 1 The Dictionary All of the commands and operations known to IsoMax are kept in a dictionary You will frequently hear these commands and operations called words because they are the words in the IsoMax language In fact you can print out a list of all the words known to IsoMax with the command WORDS This just prints a list of the words Their definitions are much longer You can find short definitions for all these words in section 18 of this manual IsoMax Glossary Possible point of confusion Don t confuse a word in the language with a word of memory A memory word on the IsoPod is 16 bits of storage A word of the language can be any symbol made of non blank characters When there is risk of confusion we will generally refer to 16 bits of storage as a cell of memory 1 2 The Stacks Numbers a
6. 2 2 000 Also you must be in DECIMAL base to type a floating point number This is because E is a valid hexadecimal digit So DECIMAL 2E0 gives the floating point value 2 0 but HEX 2E0 gives the integer 2E0 hex 736 decimal You can use the command F S to display the contents of the floating point stack The command F will print and remove the topmost item on the floating point stack in a fixed point notation and the command E will print and remove the topmost item in an exponential notation To see this try DECIMAL 1 E1 2 E1 3 El F S EF E E Floating point arithmetic operations are similar to the integer operations but have the prefix F Here are add subtract multiply and divide 10 E0 3 E0 F adds 10 0 and 3 0 giving 13 0 10 E0 3 E0 F subtracts 10 0 3 0 giving 7 0 10 E0 3 E0 F multiplies 10 0 by 3 0 giving 30 0 10 E0 3 E0 F divides 10 0 by 3 0 giving 3 3333 Other floating point operations include trigonometric and transcendental functions The complete list can be found in the glossary Section 18 4 1 5 Variables Because IsoMax carries out computations on its stacks you very rarely need to use variables such as X or Y or VELOCITY or VOLTAGE But sometimes you do need to store a value between computations So IsoMax allows you to have named variables You must define a variable before you use it This is done with the command VARIABLE integ
7. EGIN UNTIL with the following differences BEGIN WHILE false true action REPEAT LE e The code that produces the exit loop value is placed condition before the WH WHI e A zero false value means terminate the loop a nonzero true value means do the action and keep looping e It s possible for the action to be performed zero times LE and the action to be taken is after the Another way to look at this a false value at the WHILE will cause the program to immediately jump to the code after the REPEAT thus exiting the loop A true value at the WHILE will cause the code immediately following the WHILE to be executed and then REPEAT will loop back to BEGIN end start DO some action LOOP This performs an action repeatedly for a given number of times This loops over values from start to end 1 The end and start values are given on the stack when DO is encountered with the start value on the top of stack and action the end value second on the stack These values can be determined from a computation but often will just be numeric constants For example lt end LOOP end 10 0 DO At LOOP will perform the action I ten times with the loop index going from 0 to 9 inclusive The operator I will always return the value of the current loop index Unlike other languages you don t need
8. d 13 and you ll see that the loop always stops short of the end value TEST 4 DO I 2 L0O0P O 10 TEST 4 This illustrates a negative increment You ll see that the loop will include the end value if it can but it will not go past it Try end values of 1 and 1 instead of 0 Remember The most common mistake made with DO loops is to get the order of the start and end values backwards The start value is the last thing put on the stack 1 10 Comparisons Now that you have the ability to change the flow of your program based on a condition you need some operators to create those true or false flags IsoMax has four operators which will let you compare two numbers ab returns true if a b a b returns true if a gt b signed numbers ab lt returns true if a lt b signed numbers a b U lt returns true if a lt b unsigned numbers Here a and b refer to any two numbers on the stack b is on top of the stack exactly as though you had typed the numbers a b at the keyboard Yov ll note that there isn t a U gt operator We ll see in the next section how to construct one There also isn t a U operator since works for both signed and unsigned numbers IsoMax also has three operators which let you examine a single number a 0 returns true if a is zero a 0 gt returns true if a is greater than zero a 0 lt returns true if a is less than zero negative Finally there is an operato
9. ddresses and data which is being operated upon are normally held on stacks Like the name implies when you put something on a stack it becomes the topmost item on the stack and everything that was already on the stack is effectively pushed down one deeper Likewise when you take something off the top of the stack the stuff underneath it pops up one position and what was the second item on the stack becomes the new top item on the stack For the most part the operation of stacks is invisible and automatic One visible effect which will be familiar to owners of Hewlett Packard calculators is that arithmetic operations require you to place the two operands on the stack first and then specify the operation to be performed In other words instead of saying 2 Ge in IsoMax you would say 2 304 We ll see more examples of this shortly There is one stack for fixed point data including integers characters and addresses and a second stack for floating point numbers You don t need to specify this each operation automatically uses the correct stack But you might need to be aware of the two different stacks if you re moving values to and from them 1 2 1 Stack Notation Whenever you type a number IsoMax puts it on the stack Integers on the integer stack floating point numbers on the floating point stack If you type a second number it gets put on the stack and the previous number gets pushed down So for i
10. e S command Or you can use the command just a period character which prints the topmost stack item and removes it from the stack For example try 10 3 7 and you should see the result 3 Why 3 and not 3 33333 We re using integer math so we get 3 with a remainder of 1 not 3 33333 To see the remainder use the MOD command 10 3 MOD When the IsoPod starts running it expects decimal base 10 numbers But you can change this at any time If you type the command HEX all numbers from that point on will be entered and printed in hexadecimal base 16 To change back to base 10 type the command DECIMAL You can use this with the print command to perform simple base conversions For example HEX AO DECIMAL will print 160 the decimal equivalent of AO hex You can type negative numbers like 12 or even FFF in hexadecimal base If you want to negate the result of a computation you can use the command NEGATE For example 12 3 NEGATE will print 36 because 12 times 3 is 36 and the negative of 36 is 36 1 4 Floating Point Operations To type a floating point number into IsoMax you must include an exponent in the form Enn as the suffix of the number For example all of these represent a floating point value of 2 0 2 00 2 000E0 2E0 0 2E1 2E01 20 E 1 The E followed by a positive or negative number is required The following will not work 2 0 2 00
11. e value causes the code between 1 that code F and THEN to be performed A false value causes to be skipped false true action N 4 2 IF true action ELSE false action IF false THEN true trueaction ELSE false action THEN 3 BEGIN some action This is similar to IF the condition is true nonzero and a different action if the condition is false zero Remember that unlike some other languages THEN terminates the control structure The code following then is always executed THEN except that it performs one action if UNTIL This performs an action repeatedly until some condition is true The condition is given by a value on the stack when UNTIL is encountered A zero false value means do the BEGIN action again and will cause a loop from the UNTIL back to the BEGIN A nonzero true value means terminate the action loop and will cause execution to continue on to the code after the UNTIL Note that it is the action inside the loop that produces the true false value for UNTIL Whatever else e UNIL is done this action must include code which leaves this j exit loop value on the stack Note also that the action inside the loop will always be performed at least once 4 BEGIN condition WHILE some action REPEAT This performs an action repeatedly while some condition is true This is similar to B
12. er or FVARIABLE floating point Integer Floating point VARIABLE name FVARIABLE name In either case name is a name you choose for the variable This can be any combination of up to 31 non blank characters Even special characters and punctuation may be freely used For example the following are all valid variable names X Yl Velocity SPROFIT 4TH_SPEED Names can begin with numbers and can be entirely non alphabetic characters Two restrictions though First don t use a name that s already in use by IsoMax as you can see with WORDS This will cause confusion IsoMax will allow it but will warn you by telling you that your name is not unique Second don t use a name that s all numbers IsoMax will allow that and won t warn you and then when you type that number you ll get the variable instead of the number As you can imagine this will lead to no end of confusion Be sure that all your names have one non numeric character and remember that A through F are digits in hex You can use upper or lower case in your names but remember that IsoMax is case sensitive VELOCITY VELocity and Velocity are all different names When you have defined a variable you can store a value into that variable with the or F commands You can fetch the stored value with the and F commands Integer Floating point name name F stores a value in variable name name name F f
13. etches a value from variable name This is not like other languages which let you use just the name of a variable in place of a number in an equation To get the value of a variable you must use for integer variables or F for floating point variables So if you want to multiply floating point Principal by Interest to compute a payment you d have to type Principal F Interest F F F Of course before you could do this you would have had to define the variables FVARIABLE Principal FVARIABLE Interest and you would have had to store some values into these variables e g 10000 E0 Principal F 0 05E0 Interest F 1 6 Accessing Memory and I O When you define a VARIABLE what you re really doing is reserving a memory location and giving that memory location a name The operators and fetch from a memory location and store to a memory location respectively You can use and with any memory locations not just variables Suppose that you know a value is stored at memory address 6A2 You can get that value with HEX 6A2 Suppose you want to store a value of 1234 into that location You can use HEX 1234 6A2 When might this be useful Most of the time you ll want to use named VARIABLES because a variable will always be placed in an unused part of memory If you try to choose memory addresses yourself you might choose an address that s being used by IsoMax for something
14. mething which takes values from the stack Maybe we need a four bit right shift frequently RIGHT SHIFT 4 2 OF 2 2 This will take a value on the stack shift it right four times and then leave the result on the stack So you see the stack is how we pass values to a function and how we get results from a function These are the input parameters and output parameters of the function Well once we ve told IsoMax what RIGHT SHIFT 4 means why can t we use that to define GET DIP SWITCH We can RIGHT SHIFT 4 2f 2 2 2 gt HEX GET DIP SWITCH PORTA GETBYTE FO AND RIGHT SHIFT 4 1 8 1 Removing definitions If you ve been typing this example in you ve probably seen the warning GET DIP SWITCH NOT UNIQUE This is IsoMax telling you that you ve defined a word twice in its dictionary This won t break IsoMax but it will cause you some confusion since you won t necessarily know what definition is being used at any given time It s better for all concerned if you tell IsoMax to forget your previous definition of the word You do this with the command FORGET GET DIP SWITCH This gets rid of the old definition and leaves you free to start a new one Strictly speaking FORGET gets rid of the most recent definition of the word If you ve defined the word twice you ll need to use FORGET twice to get rid of both definitions TAKE NOTE FORGET will not just forget the word you
15. nd gives you ab a To rearrange the values on the stack you can use SWAP ab ba ROT abc bca SWAP just swaps the top two stack items the second becomes first and the first becomes second ROT is short for rotate it rotates the top three items on the stack such that the deepest item becomes the topmost item If you do three rotates that is ROT ROT ROT you ll get right back to where you started IsoMax has many more stack operators and you re encouraged to look at the IsoMax Glossary in Section 18 to learn about more of them But these six will handle the most of the manipulations you ll need to perform 1 11 1 Example Remember that we don t have a U gt operator It ought to look like this a b U gt should return true if a gt b unsigned numbers Well if a gt b then it follows that b lt a And we do have an operator for unsigned less than U lt So all we need to do is reverse the order of a and b and then use U lt a b SWAP U lt returns true if b lt a and thus if a gt b We can take this one step further and use our ability to make definitions to add this as a new part of the IsoMax language U gt SWAP U lt This defines a new word named U gt which does the same thing as SWAP U lt Presto What s nice is that this new word becomes a part of the language just like all the words originally known to IsoMax Any place you could use the built in word U lt you can
16. nstance if you type 12 3 the 3 will be the topmost item on the stack 2 will be under it and 1 will be on the bottom You can see this with the command S print stack Try typing the command To 2 23 uS and you will see how the numbers are stacked with the 3 on the top When we are describing parameters to be put on the stack or values which are placed on the stack we will use this left to right notation The rightmost item in the description corresponds to the topmost item on the stack So when you see parameters abc you know it really means c b a with c on the top of the stack 1 3 Arithmetic Operations Most of the time you ll be using 16 bit integers These can be treated as unsigned numbers in the range 0 65535 or as signed numbers in the range 32768 32767 The four basic arithmetic operators are add subtract multiply and divide adds 10 and 2 giving 12 subtracts 10 2 giving 8 X multiplies 10 by 2 giving 20 signed numbers divides 10 by 2 giving 5 signed numbers oOo OO E NNN LN Notice the order of the operands for subtract and divide This is easy to remember because it s the same left to right order you would use if you were writing these as algebraic equations That is performs the computation 1O 2 performs the computation 1G y 2 oO Oo 2 2 If you want to see the result of these computations you can use th
17. onnected to the high 4 bits of Port A We know how to read this port mask the bits and shift them to the low 4 bits But we re going to be doing this a lot and we don t want to type that long phrase every time Also to make the code more readable and more maintainable we d like to call it something meaningful like GET DIP SWITCH Here s how you can do it HEX GET DIP SWITCH PORTA GETBYTE FO AND 2 2 2 2 Here we are telling IsoMax to add a new word GET DIP SWITCH to the dictionary The definition of this new word is PORTA GETBYTE HEX FO AND 2 2 2 2 What this means in practice is that whenever IsoMax sees GET DIP SWITCH it will perform the action PORTA GETBYTE HEX FO AND 2 2 2 2 Strictly speaking we ve created a subroutine containing those IsoMax instructions Remember that IsoMax is free format so you can split the definition across multiple lines and use spaces to indent TAKE NOTE if you re going to use numbers inside the definition you must specify the number base outside the definition In this example we put HEX before we started the definition GET DIP SWITCH will have exactly the same stack effect as its definition Since PORTA GETBYTE HEX FO AND 2 2 2 2 leaves a single value on the stack GET DIP SWITCH will leave a single value on the stack Of course we could also have written so
18. ou typed three numbers on the command line in the same left to right order Since the stack operators rearrange the values on the stack we need before and after pictures to illustrate their operation The common notation for this is abc xyz where a b c is the stack data before the operation and x y z is the stack data after the operation We ll see how this works in a moment First let s look at words that get rid of items on the stack Word Stack effect before after DROP a 2DROP a b Sas DROP simply takes whatever is on top of the stack and gets rid of it 2DROP gets rid of the top two items on the stack Anything else that may be deeper on the stack is unaffected so it s not shown in the stack notation So if you have 1 2 3 on the stack and you do a DROP yov ll wind up with 1 2 on the stack only the 3 is DROPped There are also words to duplicate items on the stack DUP a aa OVER ab aba DUP just takes whatever is on top of the stack and makes a second copy of it on the stack so then you have two of them You might want to do this if you need to test if a value is zero or nonzero and then if it s nonzero perform some computation with it OVER is trickier it makes a copy of the second item on the stack and pushes that copy onto the top of the stack pushing everything else down If that sounds confusing just remember that OVER takes a b a
19. r which will turn true to false and false to true a NOT logical inverse of a Do not confuse NOT with INVERT INVERT is a bitwise operator which individually inverts all 16 bits of a value on the stack NOT is a logical operator which returns true a nonzero value actually FFFF if the value on the stack is false zero and returns false zero if the value on the stack is true any nonzero value INVERT is for bits NOT is for true false values Of course these comparisons are also useful in IsoMax state machines The phrase between CONDITION and CAUSES must leave a true false value on the stack How this value is produced is up to you it might come from testing an input bit or it might come from comparing two numeric values 1 11 Stack Operations For most short pieces of code like IsoMax CONDITION and CAUSES phrases you ll only have one or two things on the stack and they Il be in the right place at the right time But sometimes values get put on the stack in the wrong order or you have an extra value that you don t need or maybe you need a value twice To handle these little details IsoMax provides an assortment of stack operators Another word about stack notation recall that we use the notation abe to signify that there are three values on the stack with a on the bottom b in the middle and c on the top This is how they d be on the stack if y
20. specify it will forget all words you have defined since that word In the last example if you had typed FORGET RIGHT SHIFT 4 you would also lose the definition of GET DIP SWITCH This can be useful if you ve written a few dozen words and you want to forget them all so you If you put HEX inside the definition that number base won t take effect until later when you execute GET DIP SWITCH This is sometimes useful but usually is not what you want can start over you don t need to type a few dozen FORGET commands Just forget the first word and all the following words will go too 1 9 Program Control We ve seen how to write subroutines as new word definitions how to do arithmetic and logical functions how to store data in memory variables and how to do I O There s still one thing missing before we can write any computer program how do we perform actions conditionally That is how do we change the flow of control of the program based upon an input or the result of a calculation IsoMax offers six different constructs for program control These correspond to the basic control structures from the discipline of structured programming 1 IF some action THEN This performs an action if some condition is true The condition is given by a value on the stack when IF is encountered A zero value is false and any nonzero value will be considered true A tru
21. to use a variable for this So in this example the action I will print the loop index on each pass through the loop end start DO some action n LOOP 6699 This is similar to DO LOOP except that the loop index is incremented by n instead of 1 n is the value on top of the stack when LOOP is encountered it is usually a constant but could be the result of a computation It may be positive or negative If the increment is negative end must be less than start and the loop will proceed all the way to the end value not end 1 IMPORTANT LIMITATION All of these control structures can be used inside a word definition and inside a CONDITION or CAUSES phrase in an IsoMax state machine But they can not be used interactively from the command line 1 9 1 DO Loop Example It s instructive to write some simple definitions which show how DO loops work DECIMAL TEST 1 10 0 DO I LOOP TEST 1 This will just print the value of the loop index as it goes from 0 to 9 TEST 2 DO I LOOP 10 0 TEST 2 30 20 TEST 2 This has the same action as TEST 1 but instead of hard coding the loop limits inside the definition we are passing them as parameters on the stack So we can try the loop with a number of different start and end parameters TEST 3 DO I 3 LOOP 10 0 TEST 3 This illustrates an increment greater than 1 You can try different end values say 11 12 an
Download Pdf Manuals
Related Search