Thread: Does anyone??

Does anyone know what linking the floating point format means...

Following are a couple of functions which are said to "link the floating point format" ??!!

wtf!!

Code:

void link()
{
float f, *ff=&f;
return f;
}

One more...

Code:

void linkfloat()
{
float a=0, *b;
b=&a;
a=*b;
}

Pick a descriptive thread title if you want us to take you seriously.

All I see is someone making pointers to float values...

I found these functions in a couple of programs, and they don't seem to serve any purpose...

Originally Posted by juice

I found these functions in a couple of programs, and they don't seem to serve any purpose...

That's right, neither appears to do anything of value.

Turbo C <Salem, turn your eyes away>, has a "smart" linker, which should automatically detect when your program needs the floating point package, linked into it. Unfortunately, it didn't work really well, so the code you show, was designed to force it into linking it in.

If you don't need the floating point package, don't use those snippets of code in your program. If you do need it for your program (and aren't getting it), add those lines of code, just before your main function.

This is for Turbo C ONLY.

Originally Posted by Salem

> Does anyone know what linking the floating point format means...
Why are you
a) still posting threads with rubbish titles?
b) still using that ancient fossil TurboC (I thought you had gone all visual in another thread)
c) not pasting your error messages into google?

a) I thought it was a good title.
b) I am not using turbo C. I just saw the code in a book and after wasting some time cerebrating about what it really does, I gave up and posted it.
c) There was no error message.

Originally Posted by Adak

Turbo C <Salem, turn your eyes away>, has a "smart" linker, which should automatically detect when your program needs the floating point package, linked into it. Unfortunately, it didn't work really well, so the code you show, was designed to force it into linking it in.

If you don't need the floating point package, don't use those snippets of code in your program. If you do need it for your program (and aren't getting it), add those lines of code, just before your main function.

This is for Turbo C ONLY.

LOL... can't be all that smart if you have to use rubish code to make it load...

Why not just #include <math.h>

With Turbo-C just #including math.h will not force linking with the math library, you must actually use floating point numbers somewhere.

Jim

Originally Posted by juice

a) I thought it was a good title.

A good title for this thread would be ...

"Need help understanding odd functions"
or maybe
"Can you tell me what these functions do"

Most people here do not read every thread so unless there is some hint of the thread's actual content, you've cut your readership to something like 1/3 just by being too butt lazy to type a couple of extra words.