• State Not Answered
  • Locked Locked
  • Replies 7 replies
  • Subscribers 21 subscribers
  • Views 2619 views
  • Users 0 members are here
2025 survey

We are running a survey to help us improve the experience for all of our members. If you see the survey appear, please take the time to tell us about your experience if you can.

Options
Related
How was your experience today?
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Compiler Error 143: 'Scalar' : Initialiser is not a constant

Joost a 

#define adr(x) ((word) &(x))
char tmsec[20];

struct par_struct AL[] = {

   //par.nr, lv, ln, eh, adress                         bk, pt,           ft
   {     1,  2,  8,  0, 0,                               ptUSER,       40},
   {     2,  2,  8,  0, 0,                               ptUSER,       41},
   {     3,  2,  1,  0, 0,                               ptUSER,       42},
   {     4,  0,  0,  1, adr(Tmsec),
};

It is a part of a parameter table of about 1500 elements.
When compiling this in C166, I am getting this error when the variable adr(x) is an array with more than 6 elements. In C51 this worked perfect.
I also tried to write directly ((word) &tmsec[0]) instead of adr(tmsec) with the define, but this did not help.

What is the reason that it does not work when "tmsec" is an array with over 6 elements ?

Hope anyone can help..

Joost

Parents
  • 0 in reply to HansBernhard Broeker

    "You'll have to cite me chapter and verse of the standard you base that claim on before I agree that the address of an object qualifies as a compile time constant suitable for use as an initializer for a static (non variadic-array) object."

    Surely it must be so - otherwise how would we do tables of function pointers...?

    As well as assigning addresses, the linker also fixes-up references to those addresses; therefore, the compiler can take the address of an object as a copile-time constant.

    Well, that's what I think, anyway... ;-)

Reply
  • 0 in reply to HansBernhard Broeker

    "You'll have to cite me chapter and verse of the standard you base that claim on before I agree that the address of an object qualifies as a compile time constant suitable for use as an initializer for a static (non variadic-array) object."

    Surely it must be so - otherwise how would we do tables of function pointers...?

    As well as assigning addresses, the linker also fixes-up references to those addresses; therefore, the compiler can take the address of an object as a copile-time constant.

    Well, that's what I think, anyway... ;-)

Children
No data