Subject: Re: COMPAT_SUNOS process memory map question
To: Christos Zoulas <firstname.lastname@example.org>
From: Perry E. Metzger <email@example.com>
Date: 10/14/2001 19:27:39
firstname.lastname@example.org (Christos Zoulas) writes:
> Old sun binaries might have mapped page 0 to 0, so that *0 does not
> core dump (remember the vax :-). This is just speculation.
If I remember right, that was never the case. Back in the early days
the thing that made compiling apps for suns a bitch was that they were
the platform that shook out all the NULL vs '\0' bugs in all the
software. Reading page zero caused a fault. I remember having this
trouble on the Sun 1's and Sun 2's at Columbia all the time -- got
very good at finding that sort of bug.
Of course, I could just be getting senile.
Perry E. Metzger email@example.com
NetBSD Development, Support & CDs. http://www.wasabisystems.com/