| View previous topic :: View next topic |
| Author |
Message |
john.ma
Joined: 19 Nov 2012
Posts: 23
|
| PIC 24F Char Comparison |
 Posted: Mon Nov 19, 2012 9:04 am |
 |
|
Hi Everyone,
Not sure this is related to my work environment, but I seem to encounter some pretty annoying bugs with compiler version 4.117 using MPLAB as the editor and simulator.
1. Why does this code think str[2] is NOT equal to ch, however str[2] is equal to 'i'? Had a similar issue with pointers but worked around with an exor comparison.
| Code: |
#include <24fj128GA010.h>
//#device PASS_STRINGS=IN_RAM
#fuses NODEBUG // Debug Mode. 1.11 NODEBUG No Debug mode for ICD
#fuses HS,PR_PLL // Crystal HS, 4xPLL
#fuses NOIESO // No 2-speed Start Up
#fuses NOJTAG // No JTAG
#fuses NOPROTECT // No Memory Protection
#fuses NOWDT // No WatchDog Timer
#fuses NOCOE // Debug Mode. 1.11 NODEBUG No Debug mode for ICD
#use delay(clock=32M, crystal=8M)
#use rs232(UART1, STREAM=serial_debug, baud=9600, bits=8)
#include "string.h"
int main ()
{
char str[] ="This is a simple string";
char ch = 'i';
int out = 0;
if(str[2] == ch){
out = 1;}
if(str[2] == 'i'){
out = 2;}
return 0;
}
|
2. Often, not always when i press 'step over' the code executes until the end of the program. I read somewhere that others had this issue but could prevent that behavior with breakpoints. This doesn't seem to work for me.
Help much appreciated.
Regards,
John |
|
 |
Ttelmah
Joined: 11 Mar 2010
Posts: 19499
|
|
| Posted: Mon Nov 19, 2012 10:22 am |
|
|
Just built this, loaded it into MPLAB (8.85), and step over happily steps to out=1, and out=2. Tried a dozen times.
There are some problems with the code. Main, can not return anything. Where is it meant to return the value 'to'?... There is no OS. Then the code effectively drops of the end, and executes a PWRSAV #0 instruction.
This is why you should always have a stopper routines at the end to prevent this from happening.
It is perhaps interesting to note that the compiler uses a different compare for the two situations, using xor for the later, and cp for the former. However the code generated is perfectly sensible. I'd suggest that possibly your version of MPLAB has a problem.
Best Wishes |
|
|
john.ma
Joined: 19 Nov 2012
Posts: 23
|
|
| Posted: Mon Nov 19, 2012 11:12 am |
|
|
Thanks for your quick reply.
I've revised my configuration and cant seem to see anything wrong. I will try a newer compiler version...
I found this http://www.ccsinfo.com/forum/viewtopic.php?t=44538 and saw in the version release notes:
4.117 A bug in complex expressions involving != or == for PCD is fixed
4.117 The #TYPE SIGNED directive now also affects char declarations
4.117 An undeserved error from SPI_XFER in PCM/PCH is fixed
I wonder if the fix solved one issue and started another... :/ |
|
|
Ttelmah
Joined: 11 Mar 2010
Posts: 19499
|
|
| Posted: Mon Nov 19, 2012 1:09 pm |
|
|
I built with 4.118, which was the closest I had to .117.
Best Wishes |
|
|
john.ma
Joined: 19 Nov 2012
Posts: 23
|
|
| Posted: Mon Nov 19, 2012 1:25 pm |
|
|
| Thanks for taking the time Ttelmah, much appreciated |
|
|
john.ma
Joined: 19 Nov 2012
Posts: 23
|
| Confirmed. Equality errors with 4.117! |
| Posted: Tue Nov 20, 2012 3:54 pm |
|
|
| So tested it on a PHD students version of CCS (4.138) and its fine. Its a bug across all ~200ish computers in my deparment... |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
