Re: Oops... strange (to me) segfault problem

[Derek Martin <code pizzashack org>]

On Wed, Aug 04, 2004 at 02:23:27AM -0400, John Vetterli wrote:
> It looks like you're corrupting the free store.
>
> I suspect the first call to read() reads less than BLOCK_SZ bytes, in
> which case offset gets incremented but the memory block doesn't get
> resized, so the next call to read() overwrites the data after the end of
> the memory block.  

This turned out to be (part of) the problem, because I was wrong about
the size of the pipe buffer.  It's 4k, not 8k.

> Perhaps if you removed the "if ( status == BLOCK_SZ )" and resized
> the memory block unconditionally ???

This was part of the solution, but not all of it.  I also had to get
my math right.  The size of the reallocated region needed to be offset
+ status + BLOCK_SZ.  I forgot to add the number of recently read
characters to the size of the region.  This is what happens when we
code at 4am...  ;-)

Thanks for getting me on the right track.  Problem fixed.


A note about my earlier messages being dropped, for those who were
curious:

My earlier posts were PGP-signed using the PGP-MIME method.  Since my
last message wasn't, and it made it through to the list, I conclude
that when the gnome.org mail server receives such messages, it
seems to drop them on the floor without notifying ANYONE.  That seems
like broken behavior to me, but I guess now I know better.  Of course,
since I'm not the sysadmin of that system and I'm just guessing, I
could be wrong.

-- 
Derek D. Martin
http://www.pizzashack.org/
GPG Key ID: 0x81CFE75D