Post by thread:[PIC]: CCS compilers??? Was: C Compilers, the sequ

Hi all! My suggestion to everyone who has tight on money (and who hasn't) and has time to wait for next versions... ...WAIT!!! My experience comes from versions 3.038 to 3.053 (these versions in a little over one month). One of my problems was using getc() function in RX interrupt routine. My interrupt routine did not work until for a _long_ debugging and until I replaced getc() with something like. char_received = RCREG // (direct from receiver) I changed only one line. I also reported one bug with no response, but quess what? It was repaired, but it was not listed on the recent changes list. Without those bugs, it would be great compiler for smaller PICs. I would wait until version 3.210 or something like that. Their software policy is quite weird too. First you pay for a compiler (that should be so bug free you could earn your living with it) and then you must pay for fixes that make compiler working (in other words you do beta testing for them and you pay for it). I'dont know how bug free CCS's last versions have been, but maybe someone with more experience could tell us. "Keep your bullshit detector in good workin' condition." -Forrest Gump- -Tommi Paunonen -- http://www.piclist.com hint: To leave the PICList spam_OUTpiclist-unsubscribe-requestTakeThisOuTmitvma.mit.edu

I have heard ( and made) this complaint about CCS a time or two, but my attitude is changing. They have a different paradigm about their business, one that is more popular in software costing thousands. They charge an annual fee for maintenance. Pro-Engineer charges several thousand dollars a year for maintenance, after charging you at least $6000 for the program in the first place. So $99 a year is a steal in comparison. CCS is not a cheap compiler, it's payments are just spread out over time. Personnally, my company allows me to spend under $100 any way I see fit, whereas if I ask to spend $800 for thier competitors fine and excellent software, I get stonewalled. I have had very excellent response to bugs in the CCS compiler. I experienced a few problems after their last revision, which were promptly resolved within 24 hours after I telephoned them. Don't be afraid of them. If you find a bug in a function, get them on t he phone. If there isn't a fix, re-write it in assembler, most CCS functions are very simple. --Lawrence {Original Message removed}

Hi Tommi, I have CCS C PCM ver 2.7, and have been unable to get RX int rtn to work at all, I tried your code line: #device PIC16F628 #case #include "C:\PROJECTS\defs\16f628.h" #include "C:\PROJECTS\defs\defs_628.h" #fuses xt,nowdt,noprotect,put,nobrownout,nolvp #use delay (clock=4000000) #use rs232 (baud=9600, rcv=PB1, xmit=PB2, parity=N, bits=8,ERRORS) //#ZERO_RAM #INT_RDA void serial_isr(){ char char_received; char_received = RCREG; // (direct from receiver) // c=getc(); putc(char_received); } void flashled(); main () { enable_interrupts(GLOBAL); enable_interrupts(INT_RDA); flashled(); puts("\r\n Starting Test ...\r\n"); while(1){;} } I dropped the flashled() to save space, see any problems with this rtn it still does not work. I am using a max232 chip, and the rtn works fine without the interrupt rtn. IE getc() then putc(). So far CCS has been unresponsive. >One of my problems was using getc() function in RX interrupt routine. My >interrupt routine did not work until for a _long_ debugging and until I >replaced getc() with something like. > >char_received = RCREG // (direct from receiver) > >I changed only one line. -- http://www.piclist.com hint: To leave the PICList .....piclist-unsubscribe-requestKILLspam@spam@mitvma.mit.edu

Hi, I guess here is a misunderstanding. IMHO #int_rda makes sense ONLY if the pins assigned in the #USE statement are the dedicated h/w ones (RC6 and RC7). I hope this helps. Imre On Tue, 5 Feb 2002, Kevin A. Benedict wrote: {Quote hidden}> Hi Tommi, > I have CCS C PCM ver 2.7, and have been unable to get RX int rtn to > work at all, I tried your code line: > #device PIC16F628 > #case > #include "C:\PROJECTS\defs\16f628.h" > #include "C:\PROJECTS\defs\defs_628.h" > #fuses xt,nowdt,noprotect,put,nobrownout,nolvp > #use delay (clock=4000000) > #use rs232 (baud=9600, rcv=PB1, xmit=PB2, parity=N, bits=8,ERRORS) > //#ZERO_RAM > > #INT_RDA > void serial_isr(){ > char char_received; > char_received = RCREG; // (direct from receiver) > // c=getc(); > putc(char_received); > } > void flashled(); > main () > { > enable_interrupts(GLOBAL); > enable_interrupts(INT_RDA); > > flashled(); > puts("\r\n Starting Test ...\r\n"); > while(1){;} > } > > I dropped the flashled() to save space, see any problems with this rtn > it still does not work. I am using a max232 chip, and the rtn works fine > without the interrupt rtn. IE getc() then putc(). So far CCS has been > unresponsive. > > >One of my problems was using getc() function in RX interrupt routine. My > >interrupt routine did not work until for a _long_ debugging and until I > >replaced getc() with something like. > > > >char_received = RCREG // (direct from receiver) > > > >I changed only one line. > > -- > http://www.piclist.com hint: To leave the PICList > piclist-unsubscribe-requestKILLspammitvma.mit.edu > > -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Hi Everyone, Actually RB1 & RB2 are the hardware pins in a '628 :) Also check that the USART hasn't detected a farming or overrun error and stopped receiving.. Can't remember off hand if this shuts down the receiver, but I have a feeling that it does. Also, you set the pins to outputs/inputs as required, I can't remember which what the are supposed to be setup.. Cheers, Ash. --- Ashley Roll Digital Nemesis Pty Ltd http://www.digitalnemesis.com Mobile: +61 (0)417 705 718 > {Original Message removed}

On Wednesday, February 06, 2002 10:25 AM, Ashley Roll [SMTP:.....ashKILLspam.....DIGITALNEMESIS.COM] wrote: > Also check that the USART hasn't detected a farming or overrun error and A "farming error"! That would be an 'earth' problem then? (sorry) -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads

Hehe, either that or the MAX232-EIEIO takes a "smoko" break and all the bits escape through a hole in the shielding while it not watching :) Dumb typo of the week award goes to me I think.. :) although perhaps I could blame it on the spell checker, yes that's a good excuse, they'll believe that :P Cheers, Ash PS: for non Aussies and Kiwis, Smoko is like a lunch break, but before lunch (morning tea?) and is a good excuse not to do any work :) --- Ashley Roll Digital Nemesis Pty Ltd http://www.digitalnemesis.com Mobile: +61 (0)417 705 718 > {Original Message removed}