New variables not zero and zeroing them causes problems.

sccoupe

I am having some issues with new variable arrays. It seems like maybe they are overlapping code space but as you can see there is plenty of ram left. If a new 'long VariableID[100]' is created in the VAR section and then read, each long seems to already have a random number in it. Should these not be zero when they are declared? To further the issue, if I manually zero these values at the beginning of the object, the whole program locks up. Any ideas here?

Duane Degn

sccoupe wrote: »

Any ideas here?

You have a problem in your code.

Yes, variables declared in the VAR section will have the initial value of zero. Some part of your code is overwriting these values.

We won't be able to help much without seeing the code.

sccoupe

CON
  _clkmode = xtal1 + pll16x                                                   
  _xinfreq = 5_000_000

VAR

  long  ID2[100]

OBJ

  FDS    : "FullDuplexSerial"
  
PUB Main | a

   FDS.start(31,30,0,115_200)


repeat
  FDS.HEX(long[ID2][1],8)
  waitcnt(cnt + clkfreq)

This returns a value that is not zero.

kuroneko

As expected (more or less). Either use ID2[1] or long[@ID2][1].

Duane Degn

Having done this myself many times, I can tell you the reason zeroing the variable was causing you trouble was because instead of zeroing the variable you were overwriting the clock settings of the Propeller (around address 0).

I like using the first solution kuroneko suggested. IMO, the first notation looks cleaner than the second.

Remember the elements of your array are numbered 0 through 99. So the first element is ID2[0].

sccoupe

THANKS!!! This cost me a day but maybe never make the mistake again. Between that and using := in an if statement instead of == and then the compiler(and my eyes) not catching a few double brackets, it all working now.