Post by thread:[PIC] PIC18F1320 and jump tables

Hey all, I am trying to make a program that uses an LCD display. I know the LCD works fine, but I want to send a long string of text. Instead of doing each letter individually, I tryed to construct a jump table with dt. However, debugging shows that W always contains 8 after the jump table, instead of the ascii code for the correct letter. Heres a <snip> of my code: getstring addwf PCL,f ; A subroutine to fetch a letter dt "olleH" ; (hello backwards <snip> ; meanwhile, in the main program...... movlw 04h ; number of characters movwf number ; character counter loop: movf number,w ; put in in W call getstring ; put character number W into W IOExpLCDSendData ; macro which sends the data in W decfsz number,f goto loop <snip> BTW, I know there are ways of sending a string the right way around, but I find it easier to just reverse the string in the dt, and run the loop decrementing. Thanks in advance, -- Dave All us base are belong to you.

Here's some stuff I did to send strings to an LCD. MsgToLcd macro msg ; Set up table pointers and send a messag to LCD movlw upper(msg) movwf tblptru movlw high(msg) movwf tblptrh movlw low(msg) movwf tblptrl call LcdMsg endm LcdMsg ; Put up a message pointed to by tblprt (u:h:l) tblrd *+ ; Read the table character to tablat, then increment tblptr movfw tablat ; Get the character from the table addlw 0 ; Test w (above may not set condition codes correctly) skpnz return ; if we just got a zero byte, exit call LcdPutChar ; Go put it on the LCD goto LcdMsg ; Go do next Here's a string definition: SignOnMessageL1 db " IQ512",0 ; Line 1 - Message with terminator And here's a call of the whole mess: msgToLcd(signOnMessageL1) Harold -- FCC Rules Updated Daily at http://www.hallikainen.com

dave.w.turner@gmail.com wrote : > Hey all, > > I am trying to make a program that uses an LCD display. I know the > LCD works fine, but I want to send a long string of text. Instead of > doing each letter individually, I tryed to construct a jump table... Hi. Since the PIC18-line has good table-read capability through the TBLxxx instructions, jump-tables isn't used that much on them. In my copy of the 18F1220/1320 data sheet, the rellevant parts are in chap 6 on page 59. Check that. Jan-Erik.

After replacing the skpnz with btfsc STATUS,Z and changing movfw to movf X,w,, I tried to assembler the code. I got an error, Symbol not previously defined (tablat). Am I missing something obvious, or doesn't the 18F1320 support tables? Or do I need to setup anything else? On 5/31/05, Harold Hallikainen <spam_OUTharoldTakeThisOuThallikainen.com> wrote: {Quote hidden}> Here's some stuff I did to send strings to an LCD. > > > MsgToLcd macro msg > ; Set up table pointers and send a messag to LCD > movlw upper(msg) > movwf tblptru > movlw high(msg) > movwf tblptrh > movlw low(msg) > movwf tblptrl > call LcdMsg > endm > > LcdMsg > ; Put up a message pointed to by tblprt (u:h:l) > tblrd *+ ; Read the table character to tablat, then increment tblptr > movfw tablat ; Get the character from the table > addlw 0 ; Test w (above may not set condition codes correctly) > skpnz > return ; if we just got a zero byte, exit > call LcdPutChar ; Go put it on the LCD > goto LcdMsg ; Go do next > > > Here's a string definition: > SignOnMessageL1 > db " IQ512",0 ; Line 1 - Message with terminator > > > And here's a call of the whole mess: > > msgToLcd(signOnMessageL1) > > > Harold > > > -- > FCC Rules Updated Daily at http://www.hallikainen.com > -

Oh, now I also get lots more errors along the lines of: Error[113] MAIN.ASM 47 : Symbol not previously defined (tablat) Error[113] MAIN.ASM 16 : Symbol not previously defined (tblptru) Error[113] MAIN.ASM 18 : Symbol not previously defined (tblptrh) Error[113] MAIN.ASM 20 : Symbol not previously defined (tblptrl) On 6/1/05, .....dave.w.turnerKILLspam@spam@gmail.com <dave.w.turnerKILLspamgmail.com> wrote: {Quote hidden}> After replacing the skpnz with btfsc STATUS,Z and changing movfw to > movf X,w,, I tried to assembler the code. I got an error, Symbol not > previously defined (tablat). > > Am I missing something obvious, or doesn't the 18F1320 support tables? > Or do I need to setup anything else? > > > On 5/31/05, Harold Hallikainen <.....haroldKILLspam.....hallikainen.com> wrote: > > Here's some stuff I did to send strings to an LCD. > > > > > > MsgToLcd macro msg > > ; Set up table pointers and send a messag to LCD > > movlw upper(msg) > > movwf tblptru > > movlw high(msg) > > movwf tblptrh > > movlw low(msg) > > movwf tblptrl > > call LcdMsg > > endm > > > > LcdMsg > > ; Put up a message pointed to by tblprt (u:h:l) > > tblrd *+ ; Read the table character to tablat, then increment tblptr > > movfw tablat ; Get the character from the table > > addlw 0 ; Test w (above may not set condition codes correctly) > > skpnz > > return ; if we just got a zero byte, exit > > call LcdPutChar ; Go put it on the LCD > > goto LcdMsg ; Go do next > > > > > > Here's a string definition: > > SignOnMessageL1 > > db " IQ512",0 ; Line 1 - Message with terminator > > > > > > And here's a call of the whole mess: > > > > msgToLcd(signOnMessageL1) > > > > > > Harold > > > > > > -- > > FCC Rules Updated Daily at http://www.hallikainen.com > > --

Aha - I just remembered that unlike most people, I always have my assembler set to case sensitive. After changing all the registers to upper case it compiled fine. Runs fine too. Thank-you very much On 6/1/05, Jan-Erik Soderholm <EraseMEjan-erik.soderholmspam_OUTTakeThisOuTtelia.com> wrote: {Quote hidden}> dave.w.turnerspam_OUTgmail.com wrote : > > > I got an error, Symbol not previously defined (tablat). > > > You *do* include the p18f1320.inc file, right ? > > TABLAT is at least defined in *my* copy of the INC file... > > Jan-Erik. > > > > > -

dave.w.turner@gmail.com wrote: > Error[113] MAIN.ASM 47 : Symbol not previously defined (tablat) > Error[113] MAIN.ASM 16 : Symbol not previously defined (tblptru) > Error[113] MAIN.ASM 18 : Symbol not previously defined (tblptrh) > Error[113] MAIN.ASM 20 : Symbol not previously defined (tblptrl) Are you including the standard include file for your processor? Do you have case sensitivity disabled? ***************************************************************** Embed Inc, embedded system specialists in Littleton Massachusetts (978) 742-9014, http://www.embedinc.com

Yes I did include the include file. Yes, as I explained above, I do have case sensitivity enabled. I have just changed it. On 6/1/05, Olin Lathrop <@spam@olin_piclistKILLspamembedinc.com> wrote: {Quote hidden}> KILLspamdave.w.turnerKILLspamgmail.com wrote: > > Error[113] MAIN.ASM 47 : Symbol not previously defined (tablat) > > Error[113] MAIN.ASM 16 : Symbol not previously defined (tblptru) > > Error[113] MAIN.ASM 18 : Symbol not previously defined (tblptrh) > > Error[113] MAIN.ASM 20 : Symbol not previously defined (tblptrl) > > Are you including the standard include file for your processor? Do you have > case sensitivity disabled? > > > ***************************************************************** > Embed Inc, embedded system specialists in Littleton Massachusetts > (978) 742-9014, http://www.embedinc.com > -

On Tue, 2005-05-31 at 14:45 -0700, Harold Hallikainen wrote: > Here's some stuff I did to send strings to an LCD. <snip> This is pretty much what I do too. However, in addition I use the assembler to control the creation of an array of strings. m1 db (m1e-m1-1),"Message1 m1e if VERSION < 10 #define _VERSION VERSION + '0' else #define _VERSION '1',VERSION - 10 + '0' endif m2 db (m2e-m2-1),"Version: ",' ',_VERSION,"\r\n" m2e m3 db (m3e-m3-1),"Message3" m3e m4 db (m4e-m4-1),"Message4 is a long message that can\r\n" db " wrap around to the next line\r\n" m4e Messages data m1,m2,m3,m4 LastMessage M_STARTUP EQU 0 M_VERSION EQU 1 M_INFO EQU 2 M_LONG_MESSAGE EQU 3 I use Pascal's string encoding method by making the first byte be the length of the string. Notice how the assembler can figure this out for you. Next, I create an array of pointers to the strings and store that in program memory at the label Messages. In C, one would write something like this for that array: const char *Messages[] = {m1,m2,m3,m4}; (It's a little harder to get C to automatically figure out the length of a string and encode it in the Pascal style.) Finally, the constants M_* are created as symbolic indices. I could have even used the assembler to create these too by writing something like (untested): Messages Lm1 data m1 Lm2 data m2 Lm3 data m3 Lm4 data m4 LastMessage M_STARTUP EQU Lm1-Messages M_VERSION EQU Lm2-Messages etc. To use this assembler generated array, I write: MOVLW M_STARTUP ;Send the Startup message to RCALL SendMessage ;the console or lcd or whatever And for completeness, here's the SendMessage routine that is very similar to Harold's SendMessage: movwf temp0 ;Save the message number CLRF TBLPTRU ;Get a pointer to the MOVLW LOW(Messages) ;start of the messages MOVWF TBLPTRL ; MOVLW HIGH(Messages) ; MOVWF TBLPTRH ; ; RLNCF temp0,W ;Select a particular message ADDWF TBLPTRL,F ; CLRW ; ADDWFC TBLPTRH,F ; TBLRD *+ ;Get the message ptr MOVFF TABLAT,temp0 ;Low byte of message ptr TBLRD *+ ; MOVFF TABLAT,TBLPTRH ;High byte MOVFF temp0,TBLPTRL ; TBLRD *+ ;Get the message Length MOVFF TABLAT,temp1 ;(could reuse temp0...) ; smGetArray TBLRD *+ ;Get a byte of the message MOVF TABLAT,W RCALL DoSomethingWithTheNextByte DECFSZ temp1,f bra smGetArray RETURN Scott PS, I turned the sarcastic tone down a notch or two from the last reply to one of Dave T's questions...

Alan B. Pearce wrote: >>I use Pascal's string encoding method by making the first >>byte be the length of the string. > > ... > >> const char *Messages[] = {m1,m2,m3,m4}; >> >>(It's a little harder to get C to automatically figure out >>the length of a string and encode it in the Pascal style.) > > > I done something similar, but used C style strings with a null byte at the > end, then scanned the string until I reached the null. That works too as long as you know your strings don't have intentional 0's. In my case, this same routine sends arbitrary binary data. Scott