Re: [v2] ext4: fix possible non-initialized variable

[Carlos Maiolino <cmaiolino@redhat.com>]

> > OK, I've applied this patch to my tree.  However, I would really
> > appreciate if you could indeed review the callers for ext4_bread() and
> > ext3_bread() and send patches for ext3 and ext4 as you have suggested,
> > thanks!!
> > 
> On my todo list.
> > 						- Ted
Ted,

In case of ext4_add_entry() I'm supposing to make the function call ext4_error()
and return -EIO in the case where ext4_bread() returns NULL and err is 0'ed,
does that matches with your thoughts or is there a better way to handle with
this?
I'm talking about ext4_add_entry() behavious mainly as an example to better
understand how we should handle these situations. In case of ext4_add_entry(),
based on our discussions ext4_bread() should not fail once dir entries should
not have HOLES, so, a NULL return should indicate a on-disk corruption or an I/O
error.

Does that makes sense?

-- 
--Carlos