Subject: Re: HEADS UP: migration to fully dynamic linked "base" system
To: Nathan J. Williams <nathanw@wasabisystems.com>
From: Johnny Billquist <bqt@update.uu.se>
List: current-users
Date: 08/27/2002 01:45:27
On 26 Aug 2002, Nathan J. Williams wrote:

> woods@weird.com (Greg A. Woods) writes:
> 
> > I also know that I don't want it to ever be possible on at least some of
> > my systems (and perhaps even all of them) for a static linked program to
> > dynamically link new code into itself.
> 
> You realize that on various architectures, including our most popular
> (x86), any program can do this by just reading new code into memory,
> since all readable memory is automatically executable? Your constraint
> is therefore impossible on those architectures.

Which is actually how dynamic linked libraries are implemented in
Unix. It's actually a hack to add something that the basic design didn't
addess. (Now I'm really picking on Unix as a whole. :-)

	Johnny

Johnny Billquist                  || "I'm on a bus
                                  ||  on a psychedelic trip
email: bqt@update.uu.se           ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol