Array declaration

Newbie questions:

In the following code snippet I define an array pt with ntri*3 elements. ntri is a command line parameter and therefore not available at runtime. Surely the compiler cannot know what size pt is. Something tells me I should malloc that array - is this necessary and to what size? I'm from a FORTRAN background and used to dynamic allocation of arrays. Any direction would be appreciated.

Code:

---

typedef struct
{
        float x,y,z;
} point;

int main(int argc, char *argv[])
{
        int ntri=atoi(argv[1]);
        point pt[ntri*3];

//do something with pt ...

        return 0;
}

---

Standard C (including C99) allows only constant values in the expression in []. Some C compilers DO allow variable values in there, but that's not standards compliant, so if you want your code to be portable to many compilers, then you're better off using malloc - that works even in C89, which doesn't even allow

Code:

---

const int n = 8;
int array[n];
...

---

The other potential drawback of using the construct you describe is that if for some reason atoi returns either a negative value or a LARGE value [and you have no checks in your code to validate that the value is "small and positive"] is that the stack would overflow and the application crash. With malloc, you get back NULL if you use more space than available - and it's often a much larger space than the stack can accomodate in the first place, so you have a better chance of that working in the first place.

--
Mats

Thanks Mats

That's what I suspected. I'm using gnu c, that didn't complain, even with -Wall and I don't have lint, which would probably have picked that up. This is going to be a long learning process. I'll try googling for a freeware C99 compiler.

Eh? C99 has variable-length arrays, just not extern ones. I don't know if you're planning to use this array elsewhere, or not, but what you have so far is fine.

Quote:

---

The other potential drawback of using the construct you describe is that if for some reason atoi returns either a negative value or a LARGE value [and you have no checks in your code to validate that the value is "small and positive"] is that the stack would overflow and the application crash.

---

Just a side question here: do VLAs have to be allocated on the stack?

Some people seem to think not. http://bytes.com/groups/c/770297-c99...ay#post3070032

On the other hand, it's true that it would be quite inefficient for compilers to implement VLAs using the heap, at least in such a way that you cannot have memory leaks.

Quote:

---

Originally Posted by dwks

Just a side question here: do VLAs have to be allocated on the stack?

Some people seem to think not. http://bytes.com/groups/c/770297-c99...ay#post3070032

---

I am not sure, but I suspect that they have to be allocated on the stack. Consider:

Quote:

---

Originally Posted by C Standard, 1999 edition, section 6.2.4 paragraph 4

An object whose identifier is declared with no linkage and without the storage-class specifier static has automatic storage duration.

---

Quote:

---

Originally Posted by C Standard, 1999 edition, section 6.7.5 paragraph 3

A full declarator is a declarator that is not part of another declarator. The end of a full declarator is a sequence point. If the nested sequence of declarators in a full declarator contains a variable length array type, the type specified by the full declarator is said to be variably modified.

---

Quote:

---

Originally Posted by C Standard, 1999 edition, section 6.7.5.2 paragraph 10

All declarations of variably modified (VM) types have to be at either block scope or function prototype scope. Array objects declared with the static or extern storage-class specifier cannot have a variable length array (VLA) type. However, an object declared with the static storageclass specifier can have a VM type (that is, a pointer to a VLA type). Finally, all identifiers declared with a VM type have to be ordinary identifiers and cannot, therefore, be members of structures or unions.

---