Testing branch proposal

Matthew Burgess matthew at linuxfromscratch.org
Wed Jul 7 10:40:35 PDT 2004


On Wed, 7 Jul 2004 09:22:11 +0100
Ian Molton <spyro at f2s.com> wrote:

> On Tue, 6 Jul 2004 21:23:57 +0100
> Matthew Burgess <matthew at linuxfromscratch.org> wrote:
> 
> >  It's statically compiled
> > into bash, so there'll be a separate copy in RAM for every bash
> > process
> 
> No there wont. there will be precisely *one* copy in RAM for *all*
> bash processes. there would also be a copy in RAM for all dtynamically
> linked things that use it, and also for each other program that uses
> it statically.

Fair enough.  I'll accept your expertise on this one, but the statement
above came from an anonymous source.  I'd like to know how the
linker/loader is clever enough to discern that another process has been
statically linked to the readline stuff though.  Surely it can't be as
simple as thinking "Oh, you want bash?  Right, I have another bash
process right here. Hmmm, what stuff can we re-use that's already in
RAM?"- or is it?

Cheers,

Matt.



More information about the lfs-dev mailing list