cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - You can Bookmark boards, posts or articles that you'd like to access again easily! X

User Defined Struct Crashing Creo to Fatally Crash

MH_10366957
9-Granite

User Defined Struct Crashing Creo to Fatally Crash

Hello community,

 

I am working on creo 9.0.2.0

 

I am defining my own struct as this:

typedef struct ParmNote {

struct ParmNote *Next;
struct ParmNote *Last;
char Note[120];
wchar_t wcNote[120];

}NOTE_PARAM;

 

In a later function I have:

 

{

NOTE_PARAM *Current , *Next , *Last , *First;


First = (NOTE_PARAM *)(mParm->PString);
Last = First->Last;
Current = Last;
Next = (NOTE_PARAM *)malloc(sizeof (NOTE_PARAM));
Current->Next = Next;
First->Last = Next;
Current = Next;
Current->Next = NULL;
}

 

When it reaches the underlined line of code "Current->Next = Next;", creo fatally crashes. I do not understand why this is causing Creo to crash, as this exact logic worked in Creo 5.

 

Any and all help would be appreciated.

 

Thank you,

Mark Hardebeck

1 ACCEPTED SOLUTION

Accepted Solutions
FV
17-Peridot
17-Peridot
(To:MH_10366957)

if you would rewrite code to eliminate unnecessary variable assignments and replace malloc with calloc(...) to have allocated memory initialized:

 

{
   NOTE_PARAM *first = (NOTE_PARAM *)(mParm->PString);
   //assert(NULL == first->Last);
   if( NULL != first && NULL == first->Last){
       first->Last = (NOTE_PARAM *)calloc(1, sizeof (NOTE_PARAM));
    }
    //assert(NULL != first->Last);
    //assert(NULL == first->Last->Next);
}

 

the reason for crash would be easier to detect.

HIH

View solution in original post

6 REPLIES 6

Are you sure that the data contained in mParm->PString is compatible with NOTE_PARAM?

Have you verified that the pointer First->Last is indeed valid, not NULL, and not random data created when memory was allocated? It's a good rule to always initialize pointers to NULL or the equivalent when they are created, and verification before use is imperative.

If First->Last is not a valid pointer to a properly allocated NOTE_PARAM structure, that would wreak the havoc you are seeing.

Yes, I have verified that mParm->PString is compatible and that First->Last is not NULL.

FV
17-Peridot
17-Peridot
(To:MH_10366957)

if you would rewrite code to eliminate unnecessary variable assignments and replace malloc with calloc(...) to have allocated memory initialized:

 

{
   NOTE_PARAM *first = (NOTE_PARAM *)(mParm->PString);
   //assert(NULL == first->Last);
   if( NULL != first && NULL == first->Last){
       first->Last = (NOTE_PARAM *)calloc(1, sizeof (NOTE_PARAM));
    }
    //assert(NULL != first->Last);
    //assert(NULL == first->Last->Next);
}

 

the reason for crash would be easier to detect.

HIH

MH_10366957
9-Granite
(To:FV)

Thank you for the help, I rewrote it doing what you suggested and it worked as it was supposed to.

FV
17-Peridot
17-Peridot
(To:MH_10366957)

not that it matters but why not to use the STL containers instead of K&R circa self-referential structures?

MH_10366957
9-Granite
(To:FV)

It made some logic further down the pipeline easier to design.

Top Tags