glibc-2.3 is out

Adam Trilling agt10 at
Fri Oct 4 23:15:32 PDT 2002

On Sat, 5 Oct 2002, Greg Schafer wrote:

> I think it makes no sense to compile a "static" glibc. But like you say, a
> Ch 5 compile of glibc may be necessary. But it would suck to have to
> compile it again in Ch 6. I'm pretty sure older LFS's used to compile glibc
> in Ch 5 and install it into the chroot prefix for use in Ch 6. But that is
> not very "clean" at all.
> But we really want to avoid compiling glibc twice if at all possible (even
> though I have advocated that very thing in the past).

I have a very simple question:

how do distro makers get around problems like the current glibc stuff?

It must be possible to upgrade from glibc-2.2.5 to glibc-2.3, because
otherwise disasterous things would happen when you tried to upgrade a
binary distro (RedHat, Mandrake, etc).  Especially since glibc is
distributed as an RPM for these systems.  If we're having this much
trouble compiling glibc and having stuff break, imagine what might happen
if you swap out the binaries.  /me turns white with fear

I'd like to go on record in favor of compiling glibc twice, if that will
in fact solve the problem.  Two reasons:

1) it's not a hack (at least not like sed-ing binaries :) )

2) we could compile ch 5 gcc and binutils against it, meaning you wouldn't
have to rebuild your entire LFS system using LFS as a host in order to
take care of the binutils-2.13 stuff we had problems with over the summer.

I'd like to mess around with this, but I don't really have a way of doing
test builds until I either figure out how to run VMWare on win2k or
acquire enough hardware to make another box.

Semi-on-topic: How does one upgrade an existing LFS system?  If I just let
make install overwrite all of my old files, am I setting myself up for big

(I want to upgrade an LFS-08022002ish box to LFS-4.0)

Adam Trilling
agt10 at

char m[9999],*n[99],*r=m,*p=m+5000,**s=n,d,c;main(){for(read(0,r,4000);c=*r;

Unsubscribe: send email to listar at
and put 'unsubscribe lfs-dev' in the subject header of the message

More information about the lfs-dev mailing list