Re: Re : Segmentation fault in NSCell

[Fred Kiefer]

On 07.02.2012 15:22, Lucas Schnorr wrote:
> On Tue, Feb 7, 2012 at 14:56, Mathieu Suen<mathieusuen@yahoo.fr>  wrote:
> > My guess is:
> > 0xdeadface = dead face?
> > The pointer is already freed.

Yes, this is definitely a double free isue. You are freeing an object 
that was already deallocated.

> > What did you do before the call to  [super dealloc] ?
>
> Before the [super dealloc], the code frees PieCell attributes.
>
> I think I may have found the problem. I have a NSMatrix (of PieCells)
> that is loaded from a gorm/nib resource in the awakeFromNib method of
> the class that manages the matrix. In the dealloc method of this
> manager class, the code does:
>
> [matrix removeFromSuperview];
>
> And I think I'm not supposed to do that, if the matrix variable points
> to an instance that was loaded from a gorm file. If I remove this
> line, it works - the segfault disappears, but I'm not sure if there
> isn't memory leaks.

This may not be the root cause, it should be OK the remove the matrix 
from the view hierarchy (Depending on how it got created). But if it 
works as a work around you might live with it. Otherwise you will have 
to provide more of your code to analyse.

If you want to go deeper, you should send the code where you create the 
matrix and the code where you create the PieCell. Is the PieCell the 
cell prototype for the matrix? And are you dealing with an array of 
cells somewhere in your application? That code might be of interest as well.

Fred