> "Richard L. Hamilton" <[EMAIL PROTECTED]> wrote: > > > Shouldn't this use MAP_ANON and an fd arg of -1 > rather than explicitly > > opening /dev/zero? One less thing to go wrong, > > two less statements and syscalls, typically. > > The related lib is much older than MAP_ANON.
Understood, although in a perfect world, it (IMO) should have been updated when or shortly after MAP_ANON was introduced. > > > around that. And I remember that SVR2 on e.g. the > AT&T 3B15 allowed > > dereferencing at least a NULL char *, but I don't > remember whether that > > bad precedent was only intended to apply to char * > (and presumably, > > The related code was definitely broken. SVR2 sources > did e.g. expect to > see a nul character when dereferencing argv[argc]. Right; I know I remember seeing some evil code that depended on the behavior, and even something documenting it, but I don't remember the details, and it's been a long time since I had access to the old SVR2 code. This message posted from opensolaris.org _______________________________________________ opensolaris-code mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
