Re: malloc attributes and realloc

[Gabriel Dos Reis <gdr at integrable-solutions dot net>]

Robert Dewar <dewar@gnat.com> writes:

[...]

| As for comparing old pointers from realloc, I am less convinced
| 
| a) it is hard to believe many programs are affected
| b) the value from a semantic point of view of this feature is marginal

I don't agree there.  I've seen many discussions about ways to have an
efficient implementation of C++'s std::vector that expands in-place.
The general impression I've got from such discussions is that people
that roll their own home-grown std::vector-like containers (for
various reasons, either because they have those containers before we
got a standard or because they don't trust the implementation from
their favorite library supplier) using malloc+realloc were actually
doing that comparaison (the vulgar form I pointed out in a previous
discussion). 
I don't believe "the value from a semantic point of view is marginal" 

-- Gaby