use MALLOC_PERTURB_ ... or lose

[Jeff Spaleta]

On Wed, May 5, 2010 at 9:54 AM, Frank Ch. Eigler <fche at redhat.com> wrote:
> Good point.  Clearly though one can't delay the setting of the final
> release behaviors too long, or else *those* won't get tested.

I'm not arguing about what that point should be. I'm just saying that
this glib debugging stuff should sync with how we are handling kernel
debugging in rawhide as well as in pre-releases for maximum benefit.

-jef