Subject: Re: HEADS UP: migration to fully dynamic linked "base" system
To: NetBSD-current Discussion List <current-users@NetBSD.ORG>
From: Greg A. Woods <>
List: current-users
Date: 08/26/2002 18:20:29
[ On Monday, August 26, 2002 at 12:25:06 (-0700), Jason R Thorpe wrote: ]
> Subject: Re: HEADS UP: migration to fully dynamic linked "base" system
> On Mon, Aug 26, 2002 at 09:08:40PM +0200, Johnny Billquist wrote:
>  > Yes, and I'd consider that the correct behaviour, by the way. It's ugly,
>  > but it's the right (actually the only) way it could be done. And
>  > "exporting" symbols from the static program, while nice, could be lived
>  > without as well.
> No, it's not correct behavior.  There's the "which global variable do
> you use, errno or errno?" problem.

The namespace issues do get a little hairy, though if I'm not mistaken
there are some concepts that might help a bit in the Darwin mulit-level

A static-linked libc could also always provide (i.e. always include in
the static binary) call-back functions which a dynamic-linked module
would have to use to modify global variables such as 'errno' in the
"parent" program's address space.  Dynamic modules have to be coded
specially for that to work of course (or a _lot_ of compiler magic might

In any case it's a can of worms that I want my kernel to be able to
prevent from ever being opened should I so desire it.

								Greg A. Woods

+1 416 218-0098;            <>;           <>
Planix, Inc. <>; VE3TCP; Secrets of the Weird <>