Are long single C source files unreliable?

Hi,
I'm needing an experienced programmers opinion about gcc and writing c programs. I've written a fairly large program 11000 + lines of code in a single program - a single file.
It compiles and runs fine on one Linux server yet on another Unix server, while it compiles fine, when executed it seems to stuble over the simplest areas of code that it should recognise - there is no error in the code - but for some reason the executable won't run - very peculair behavior.
Does gcc have trouble dealing with very large files or something? Or is it the different system the problem here? I just don't get it ! For example in the main function if I omit the following line, which initialises a global variable (struct variable)

input.is[0]='\x00';

the script will work - now what the hell is going on here? Its the same for certain other spots in the code - it just doesn't seem to like it for some reason when its perfectly correct code.
Would breaking the code down into smaller files and linking during compiling likely help? Does this improve the chances of long areas of code working more reliably?

Has anyone experienced this sort of problem before and how did you get around it,

Cheers, your help and words of wisdom will be much appreciated,

Brett

Hi,
I'm needing an experienced programmers opinion about gcc and writing c programs. I've written a fairly large program 11000 + lines of code in a single program - a single file.
It compiles and runs fine on one Linux server yet on another Unix server, while it compiles fine, when executed it seems to stuble over the simplest areas of code that it should recognise - there is no error in the code - but for some reason the executable won't run - very peculair behavior.
Does gcc have trouble dealing with very large files or something? Or is it the different system the problem here? I just don't get it ! For example in the main function if I omit the following line, which initialises a global variable (struct variable)

input.is[0]='\x00';

the script will work - now what the hell is going on here? Its the same for certain other spots in the code - it just doesn't seem to like it for some reason when its perfectly correct code.
Would breaking the code down into smaller files and linking during compiling likely help? Does this improve the chances of long areas of code working more reliably?

Has anyone experienced this sort of problem before and how did you get around it,

Cheers, your help and words of wisdom will be much appreciated,

Brett

I can't speak for GCC, but with some old compilers there is a segment size that you had to deal with. This applied also to the amount, if I remember correctly, of lines of code you could end up with in a single file. (16K or something?) I don't remember the specifics.

For purely organization's sake, you may want to split that up into smaller files, just so it's easier to find what you're looking for.

Quzah.

Sorry BJ, but I have to suspect a typo in your code. After all, gcc compiled the Linux kernel and all the supporting applications that came with your OS. Some of these are far larger than 11k lines of code.

When a compiler reports an error at (let's say) line 766, yet line 766 is a perfectly legal statement, start scanning backwards. You may find a missing (or extra) semi-colon or brace that confuses the compiler. If you still believe it is gcc and not your code, try using a different compiler.
I'm curious to know how that turns out.

Cheers,

I've had problems using GCC and large files, got an linker error. Solved this by splitting the files, it helped.

I think it is good practice anyway to split a program into multiple source files based on categorical concepts. You should try splitting it up and then running it again on multiple systems and see if it makes a difference.