> What about the posix.4 shared memory objects? Any chance we might ever
> see them? The ability to map shared memory segments to the same
> logical address in several different processes would make the use of
> pointers in shared memory possible.
>
> Johnie
>
Though I am pretty anxiously waiting for the posix.1b/posix.4
features as well, I would like you to take notice of the fact
that the feature of attaching a shared mem segment to a specified
address is already given with SysV IPC.
shm_root_p = shmat (shm_id, (char *)0xa0000000 ,0);
is a possible action. But anyway: the 'clean' solution for pointers
in shared memory would always be having addresses relative to the
segment's borders.
Volker
/*******************************************************************/
/* Volker Eberlein */
/* volker.eberlein@fr.bosch.de */
/*******************************************************************/