• State Not Answered
  • Locked Locked
  • Replies 3 replies
  • Subscribers 22 subscribers
  • Views 1015 views
  • Users 0 members are here
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

Variables that overlap in the same memory space possible?

stephen phillips

Currently I have a set of structures represented as variables such as:

MYTYPE xdata A;
MYTYPE xdata B;
MYTYPE xdata C;
MYTYPE xdata D;
MYTYPE xdata E;
MYTYPE xdata F;
MYTYPE xdata G;
MYTYPE xdata H;


unfortunately E and G happen to not be used in this version of the software. Is it possible to force E and G to exist in the same space as F (for example)? Abusing the _at_ keyword seemed an exercise in futility (of COURSE it gives an error because you need a constant location that's an int for the compiler).

Any suggestions?
To answer WHY (which I'm sure was asked instantly) it's a matter of only having 1K of xdata space. This of course doesn't solve the issue if one needs to have ALL the variables. I'm just seeing what I can do without making a worse mess.

Thanks for listening (gritting teeth maybe?)

Stephen

  • 0

    unfortunately E and G happen to not be used in this version of the software. Is it possible to force E and G to exist in the same space as F (for example)? Abusing the _at_ keyword seemed an exercise in futility (of COURSE it gives an error because you need a constant location that's an int for the compiler).

    the simnple solution

    MYTYPE xdata A;
MYTYPE xdata B;
MYTYPE xdata C;
MYTYPE xdata D;
//MYTYPE xdata E;
MYTYPE xdata F;
//MYTYPE xdata G;
MYTYPE xdata H;

    another solution

    MYTYPE xdata A;
    
MYTYPE xdata B;
    
MYTYPE xdata C;
    
MYTYPE xdata D;
    
#define  E D
    
MYTYPE xdata F;
    
#define  G F
    
MYTYPE xdata H;

    what I would do

    MYTYPE xdata A;
MYTYPE xdata B;
MYTYPE xdata C;
MYTYPE xdata D;
#ifndef ABBREVIATE
MYTYPE xdata E;
#endif
MYTYPE xdata F;
#ifndef ABBREVIATE
MYTYPE xdata G;
#endif
MYTYPE xdata H;

    Erik

  • 0

    unfortunately E and G happen to not be used in this version of the software.

    What's "unfurtunate" about that?

    And what's keeping you from simply _removing_ those variables, then? An #if NEED_FEATURE_E around the definition should work just fine for that.

  • 0 in reply to HansBernhard Broeker

    The implicate order of things?
    Well unfortunately E and G are referenced in an array of constant pointers. It's not as though I didn't think to eliminate them.
    I was thinking

    #define E F
#define G F


    As another possibility. Truth be told I was hoping to abuse the _at_ operator and do something akin to

    MYTYPE F;
MYTYPE E _at_ &F;
MYTYPE G _at_ &F;


    unfortunately there is no way to change &F to the literal the compiler expects.

    The suggestions are legitimately thought out, I just was hoping that is what I did not HAVE to do. I guess no matter what I do it will be messy.

    Stephen