Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 267
  • Last Modified:

Problem porting C program form redhat6.1 to redhat5.2

I have written a C programand compiled with the version of gcc which comes with redhat6.1 - I need to use the code on a machine which has redhat5.2 installed. It runs fine on 6.1 but gives a segmentation fault on 5.2. here's the code:

void main(void)
{
int x;
int *y;

//the problem occurs in the next line
x=*y;
}

The access to the *y pointer causes a segmentation fault on 5.2 but works fine on 6.1 Can anybody help?
0
ruadhan
Asked:
ruadhan
  • 3
  • 2
1 Solution
 
marcjbCommented:
void main(void)
{
int x;
int *y;

//the problem occurs in the next line
x=*y;
}

This probably causes the segmentation fault because the pointer y is not initialized.  I assume that the results of such actions are technically undefined, meaning that different builds of the code will behave differently, and can have adverse affects.  The pointer should be initialized before it is used.

Also, it should be noted that
void main
is incorrect according to the C Standard.  main should return an int, or some systems may experience problems (See Plauger's book on the C Standard).

Hope this helps,

Marc
0
 
makerpCommented:
thats right your *y pointer should be at least set to NULL

i.e.

int *y = NULL;

at the moment your pointer could be pointing any whare in memeory. i.e a segment of memeory that you dont own etc...
0
 
ruadhanAuthor Commented:
even with the NULL assignment the segmentation fault occurs
0
Concerto's Cloud Advisory Services

Want to avoid the missteps to gaining all the benefits of the cloud? Learn more about the different assessment options from our Cloud Advisory team.

 
makerpCommented:


void main(void)
{
int x;
int *y;

//the problem occurs in the next line
x=*y;
}


your problem is as follows.

x is an automatic ingeter that has been decalred but given no value (fine)

y is a pointer variable that at present points to an undefined area of memory. (fine)

you then make the contents of x equal what the *y points at. which is nothing. even if it pointed to null this would still cause an error bescase you would be making the value of x equal to value of NULL or address x000000000.

what you need to do is first of all allocate some memory for *y and then assign it a value. OR make y point to another pointer which is valid

i.e

int *y;

y = (int*)malloc(sizeof(int));
if(y)
{
     *y = 10;
}

x = *y;

now y is pointing to a valid owned piece of memory.

even if you did not assign *y a value an error would not occur, x would just have an unpredictable result in. this is because a malloc does not set the contents of a piece of memeory it just give you it, therefore referncing it before an assignment will yield its previous contents.
 
0
 
ruadhanAuthor Commented:
hi makerp - the new code works. it still bugs me though, why it the original code works correctly in redhat6.1 and slackware, but not in redhat5.2. any ideas? (the gcc versions in rh5.2 and slackware are the same, but differnt i think in rh6.1)
cheers.
0
 
makerpCommented:
possibly because the compilier has intervined in redhat6.1 and given x 0, cos y = NULL. NULL does equate to 0;

OR

Luck/Missfortune

OR

the memory that an unitilalized pointer points to in redhat6.1 is diiferent than the other compiler.

to be honest the redhat6.1  compiler should have bomded out on this ??



 
0

Featured Post

VIDEO: THE CONCERTO CLOUD FOR HEALTHCARE

Modern healthcare requires a modern cloud. View this brief video to understand how the Concerto Cloud for Healthcare can help your organization.

  • 3
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now