Post by thread:MPLAB bug? HELP!!!

Hi all, I've been trying to use MPLAB and have ended up very confused! What is the real result of this operation? MOVLW 01 ADDWF PC,SAME RETLW 0AH RETLW 0Eh MPLAB insists on jumping to the first RETLW 0Ah. I reckon it's wrong and plug for the second RETLW 0Eh. A zero offset ends up in an infinite loop!!?? I'm sure I knew how this all worked but after using MPLAB, I am confused and have a king size headache! Am I bugged or is it MPLAB? Also, the RETLW instruction description "example" for all the PICs I checked appears to describe the table jump operation incorrectly. TABLE ADDWF PC RETLW K1 RETLW K2 " " RETLW Kn w = 0x07 should return K8 NOT k7, right? This is just to confuse me more I guess.... Jim

Newfound Electronics <spam_OUTPICLISTTakeThisOuTMITVMA.MIT.EDU> wrote: > I've been trying to use MPLAB and have ended up very confused! > What is the real result of this operation? > > MOVLW 01 > ADDWF PC,SAME > RETLW 0AH > RETLW 0Eh > > MPLAB insists on jumping to the first RETLW 0Ah. I reckon it's wrong > and plug for the second RETLW 0Eh. > .... > Am I bugged or is it MPLAB? Jim: Check to make sure that you've defined "SAME" to "1". If so, the bug is in MPLAB. > TABLE ADDWF PC > RETLW K1 > RETLW K2 > " " > RETLW Kn > > w = 0x07 should return K8 NOT k7, right? Right. -Andy Andrew Warren - .....fastfwdKILLspam@spam@ix.netcom.com Fast Forward Engineering, Vista, California http://www.geocities.com/SiliconValley/2499

This is a known bug in MPLAB. They told us about it at the last seminar. See? The seminars are good for something. The simulator doesn't add one to the PC (for the pre-fetch) in operations which modify the PC. The work around is to add 1 yourself in code which you will run on the simulator and take it out again when you go to silicon. - Mark Sullivan -

>This is a known bug in MPLAB. They told us about it at the last seminar. See? >The seminars are good for something. The simulator doesn't add one to the PC >(for the pre-fetch) in operations which modify the PC. > >The work around is to add 1 yourself in code which you will run on the simulator >and take it out again when you go to silicon. > >- Mark Sullivan - > > Thanks to all who helped with the MPLAB bug, my sanity is restored and my headache sank into the depths of my pillow. Mark, I thought of the above too but my application was a delay and the offset was used as part of the delay like this: movlw 03 andwf baudrate xorlw ?? xorlw ?? xorlw ?? movwf dlycnt decfsz $-1 This obsure method is because I'm using a 16C57 with only 2 levels of stack and limited literal instructions. Pretty clever huh? Given 4 such tables for the RS232 TX and RX routines, it sort of gets confusing, well for me anyway! I think I'll grab the new MPLAB version. My thanks once again... Regards, Jim