Segment Violation (Overflow of memory)

With the sample.c program I tries to convert a pseudo-code (from mycode.txt file) in an assembly code (to a future file.asm)

Code:

---

root@ferran-F5GL:/home/ferran/Documents/C# cat mycode.txt
data:
    db msg="Hello, World!\n";
    equ len(msg);


start:
    sys_write(len, msg, 1, 4);
    kernel;
    sys_exit(1);
    kernel;


root@ferran-F5GL:/home/ferran/Documents/C# ./sample mycode.txt
section        .data


msg        db        'Hello, World',0xa
len        equ $ - msg
Violació de segment (s'ha bolcat la memòria)
root@ferran-F5GL:/home/ferran/Documents/C#

---

"Violació de segment (s'ha bolcat la memòria)" in english means "Segment violation (overflow memory)

the sample.c code is

Code:

---

/*
This program will tries to convert a pseudo-code
 from a text file to assembly code
 which will -later- be saved in an .asm file.


(By the moment don't exists any real conversion.
The program currently makes a copy of a assembly file
meanwhile i fix this).
*/


#include <stdio.h>
#include <stdlib.h>
#include <string.h>


int main( int argc, char *argv[] ) {


    char myfile[20];
    char myline[100];
    char *token;
    const char sptr[10]=" ;()[]\t\n";


    token = (char*) malloc (sizeof (50));


    FILE *fp;


    if ( argc == 2 ) {
        strcpy(myfile, argv[1]);
    }
    else {
        printf("Invalid number or arguments.\n");
    }


    fp = fopen(myfile,"r");


    if ( fp == NULL ) {
        fputs("Error in file ", stderr);
        exit(1);
    }
    else {
        while ( feof(fp) == 0 )
        {
            fgets(myline,100,fp);
            token = strtok(myline, sptr);
              if ( strcmp(token, "data:") == 0 ) {
                    printf("section\t.data\n\n");
                    }
                else if ( strcmp(token, "bss:") == 0 ) {
                    printf("section\t.bss\n\n");
                    }
                else if ( strcmp(token, "start:") == 0 ) {
                            printf("section\t.text\n");
                            printf("global\t_start\n\n");
                            printf("\t_start:\n\n");
                    }
                else if ( strcmp(token, "db") == 0 ) {
                            printf("msg\tdb\t'Hello, World',0xa\n");
                    }
                else if ( strcmp(token, "equ") == 0 ) {
                            printf("len\tequ $ - msg\n");
                    }
                else {
                            printf(" \n");                   
                }
        }
    }
fclose( fp );
return 0;
}

---

P.D. I tried to attach here the sample.c program, but is not possible.

if you could see where is the problem.

Thank you in advance.

Get rid of this:

Code:

---

token = (char*) malloc (sizeof (50));

---

You don't need to malloc as token will be pointing to what strtok returns.

This:

Code:

---

if ( argc == 2 ) {
    strcpy(myfile, argv[1]);
}
else {
    printf("Invalid number or arguments.\n");
}

---

should have been:

Code:

---

if ( argc != 2 ) {
    printf("Invalid number or arguments.\n");
    exit(1);
}
strcpy(myfile, argv[1]);

---

Otherwise you risk accessing argv[1] even when the command line argument argument was not provided (in such a case, most likely argc == 1, in which case argv[1] is a null pointer, yet strcpy does not accept a null pointer argument, resulting in undefined behaviour).

Change this:

Code:

---

while ( feof(fp) == 0 )
{
    fgets(myline,100,fp);

---

To this:

Code:

---

while ( fgets(myline, 100, fp) )
{

---

It is usually easier to use the result of the read operation to control the loop instead of using feof because feof will only return true after EOF has been reached, but you usually don't know that it has been reached until you attempt a read operation.

Aside from these, you need to indent your code more consistently.

> root@
And create yourself a proper user account for your day to day activities.