Happy testing - gcc-3.2 and other commited

Greg Schafer gschafer at zip.com.au
Sat Sep 14 19:50:39 PDT 2002


On Sat, Sep 14, 2002 at 02:40:27PM +0200, Mathias Korall wrote:
> On Thu, 12 Sep 2002, Greg Schafer wrote:
> 
> > On Wed, Sep 11, 2002 at 09:44:08PM -0400, Gerard Beekmans wrote:
> > > On August 27, 2002 06:35 am, Greg Schafer wrote:
> > > that's the case, can't we fix this even easier by running this:
> > > 
> > > 	PATH=$LFS/static/bin:$PATH ./configure \
> > > 		./gcc-3.2/configure --prefix=/static <and the rest here>
> > > 
> > > That would make GCC find the just compiled binutils from chapter 5 first, 
> > > which is a new enough version, which then in turn should result in a properly 
> > > compiled gcc and later on glibc?
> > 
> > Heh Gerard!
> > 
> > Mark H actually suggested this way back:-
> > 
> > http://archive.linuxfromscratch.org/mail-archives/lfs-dev/2002/07/0135.html
> > 
> > It doesn't seem to work though :( I looked into it at the time and concluded
> > that the gcc configure script wasn't checking the path even though comments within
> > the script indicated it was!
> 
> If it is just 'as' that is the problem there is a configure
> option to specify which 'as' to use. Then it should work with
> 
> configure --with-as=$LFS/static/bin/as ...
> 
> I have not tried it though.

Yes, buried in that thread (or somewhere nearby) I did actually suggest that
as a possibility, but ruled it out due to doubts whether it would actually
work..

Let me clarify:-

 * it is actually "ld" it is looking for
 * by passing the switch, I could get it to correctly pass the detection test
 * but what happens inside the chroot when trying to compile glibc? It might be
   using some hardcoded path (/mnt/lfs/static/blah...) to ld ??

The only way to find out is to test.. but I'm over it.. The HAVE_GAS_HIDDEN
thing is not ideal, but it works.. Feel free to test and report your findings.

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



More information about the lfs-dev mailing list