John R. Daily <jdaily@bbn.com> wrote:
> Michael K. Sanders wrote:
> <snip>
> > Data modified on freelist: word 0 of object 0xf867a780 size 60 previous type free (0x0 != 0xdeadbeef)
> > obio0 at obio0
> > Data modified on freelist: word 0 of object 0xf8151978 size 40 previous type ??? (0xf867z780 != 0xdeadbeef)
> <snip>
> > So, any ideas on either problem? Is 'deadbeef' supposed to be some
> > sort of sick twisted joke? :)
> 
> I'm afraid I can't contribute much to the discussion, but I can
> at least point out that "deadbeef" is Sun's standard string to
> insert into the kernel at a certain point in memory to keep an
> eye on, and determine whether that block of memory has been
> (wrongly) modified.

It's also very hard to de-reference if it's being used incorrectly
as a pointer, as all pointers must be divisible by 4 and 0xdeadbeef
is not even divisible by 2.  (I'd say "odd" in the mathmatical
sense, but that might be mis-construed 8-) And it's a bit harder
to generate bug-wise than 0x00000000 or 0xffffffff.