lpc2103 sprintf problem

Hi I do have problems with the sprintf function and the Keil compiler. The program works nicely until I increase the amount of functions, even if I don't use them. First I get the
"%02d,%d"
into the buffer if I define one extra function but the program stops working if I define more functions.

I made my own, incomplete sprintf function and then it works perfectly.

I have tried to play the size of the following parameters in the Startup.s file:
USR_Stack_Size EQU 0x00000400
Heap_Size EQU 0x00000400
but without any success.

If anyone has any clue I would be very thankful.

Regards
Gudjon

 RE: lpc2103 sprintf problem

The problem is solved but I don't know what I did (if anything) to solve it. Now it works with cstack=0x200 and heap=0x200.

Regards
Gudjon

 RE: lpc2103 sprintf problem

The problem is solved but I don't know what I did (if anything) to solve it.

Then it's not solved --- it just disappeared. If you don't know how you fixed it, you didn't.

 RE: lpc2103 sprintf problem

Hi Hans, your name sounds familiar are you a Debian-Linux user?
I can spend some time on trying to reproduce the problem if you are interested. I was running Keil under Wine on my Linux box and it may have had something to do with CRLF line endings. Today I got my JTAG connector so I moved the code to Windows and then it worked. When I compiled it again under Linux, it worked too. Other possible reason is that I made a mistake in the interrupt routine that I have fixed now. I had however commented the interrupt routine out and all related parameters but I may have missed something out. The problem has disappeared, I don't know the reason but if it doesn't appear again, I'm happy :) But do you know if it is possible to run the JTAG connector under Linux?

Regards
Gudjon

 RE: lpc2103 sprintf problem

Well, my conclusions above were incorrect. Using sprintf just gives me unpredictable results if I try to format integers. Sometimes it works sometimes not. It seems to work nicely when I use it for constant character arrays at least.

I will close the thread if I find the real problem sometime.

 RE: lpc2103 sprintf problem

your stack requirements will grow significantly if you have nested calls to functions such as sprintf.

the program stops working if I define more functions.

exception? what is the value of R14, if so?

 RE: lpc2103 sprintf problem

Hi Tamir

Thanks for the answer. I don't have any nested calls to sprintf. The main loop of the code looks like this (a bit simplified):

int
main ()
{
  char buf[30];
  char fname[30];
  unsigned char ch;
  Initialize ();
  UART0Initialize (57600);
  UART1Initialize (57600);
  CCR=0x11;
  T1TCR = 0x1;
  while (1)
    {
          ch=UART0ReadChar_nostop ();
      if (ch=='g')
            {
              sprintf(buf,"%04d.%02d.%02d %02d:%02d:%02d\r\n",(int) YEAR , (int) MONTH, (int) DOM, (int) HOUR, (int) MIN, (int) SEC);
              UART0WriteString((unsigned char *) buf);
                if (0==VNC1L_DiskPresent ())
                  {
                    sprintf(fname, "%02d%02d%02d%02d.dat",((int) YEAR%100), (int) MONTH, (int) DOM, (int) HOUR);
                    VNC1L_FileOpenWrite((unsigned char*) fname);
                    VNC1L_FileWrite(buf);
                    VNC1L_FileClose((unsigned char*) fname);
                    VNC1L_SuspendMonitor();
                    UART0WriteString((unsigned char *) fname);
                  } else {
                    sprintf(buf,"No disk found\n");
                UART0WriteString((unsigned char *) buf);
                  }
            }
    }
}

Then I added a very simple function definition without using it and the code stops working.

When you ask for the value of R14, is it during the call to sprintf you mean? I have to admit that I had some difficulties debugging that portion.

I think I will have to live with my tiny_sprintf function until I get more familiar with the microcontroller.

Regards
Gudjon

 RE: lpc2103 sprintf problem

what do you mean by "the program stops working"? does your processor enter an exception mode? if so, you can try to figure out what the last instruction executed was before the exception by looking at R14 minus 8 (when in abort or under exception modes). if no exception occurs - what does happen? again, what does "stops working" mean?