Today i had a strange problem with my customers program. It is very
fresh soft, and they are still strugling with it (developing i meant :-)
). But they were stucked on very strange problem.
[explanation of problem in program- ommited]
Problem it self.
this code will be an example.
struct doom{
int data1;
int data2;
void *mydata;
};
doom doomed;
void dosomething_with_data(doom *d)
{
// .... doing some fancy operations
// then..
if (d->mydata)
{
free(d->mydata);
d->mydata=NULL;
}
}
void main()
{
int i,j;
void *t;
// this program is just not doing anything important :-)
t=doomed.mydata=malloc(somenumberofbytes);
if (t==NULL){
return;
}
dosomething_with_data(&doomed);
// plenty of lines and finally ........
if (doomed.mydata)
{
free(doomed.mydata);
doomed.mydata=NULL;
}
// now i will get strange behavior in other structures
// seems like it is a memory leak.....
}
well, finding this error in somebody software was not easy.
Finaly i tested simmilar behavior on windows, and... it was not better.
But under compilation with m$ comiler in debug mode gived me assert on
first attempt to free allready freed memory.
Is this behavior system has to take ? (i mean segfault)
which options for gcc i have to use to get assertion in case of such
error (i guess malloc seats in glibc, thus debug version of this liblary
might have this). But i am not sure. Waiting for comments.
-- Grzegorz Jaskiewicz <gj@pointblue.com.pl> K4 Labs- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/