Thread: Segfault from calloc

Hi,

I'm getting a Segmentation fault from a calloc call. This is the first time I've ever had something of the sort, and I'm curious as to why I'm getting such a fault/what could potentially be the problem/how I can fix it.

Thanks.

That's the point I was getting at, though. I know nothing is wrong with calloc.

But you say my code has "trashed memory".

That is what I wish to know of. What does "trashed memory" mean, and where should I start looking as far as debugging goes ?

> What does "trashed memory" mean,
All of the following can cause trouble later on
- running off the end of any array (either end, whether malloc'ed or a real array)
- using an uninitialised pointer (before calling malloc)
- using a dead pointer (after calling free)
- freeing the same pointer twice
- freeing something which wasn't malloc'ed in the first place

ok, but here's the thing, I've managed to pinpoint the exact line where I've been getting the segfault :

Code:

void * buffer = calloc(1,BLOCK_SIZ);

BLOCK_SIZ is a long specified by the user. Now, based on what you've told me about trashed memory, I don't think this particular line is problematic. So I'm guessing the bug is somewhere else in the code. But then why am I getting a segmentation fault on that line !?

> So I'm guessing the bug is somewhere else in the code.
Correct

> But then why am I getting a segmentation fault on that line !?

Code:

char *p, *q;
p = malloc(10);
strcpy(p,"a string with far more than 10 characters in it");
q = malloc(10);

Imagine that the memory pool is broken up as follows
P A P A P F P A P
P is a link pointer to the next block
A is an allocated block
F is a free block.

If you write past the end of an A block (see the strcpy), you're going to trash a P
The next malloc / free routine is going to smash into that pointer, take a leap into the big unknown and promptly die with a segfault.

The segfault is the effect, not the cause.