GCC-3.3.3 Pass2

Ken Moffat ken at kenmoffat.uklinux.net
Tue Jul 6 06:50:15 PDT 2004


On Tue, 6 Jul 2004, [ISO-8859-1] Gökhan Ibis wrote:

> configure:973: checking for gcc
> configure:1086: checking whether the C compiler (/usr/bin/gcc296 -g ) works
> configure:1102: /usr/bin/gcc296 -o conftest -g   conftest.c  1>&5

>
> i copied the whole because I think 973 onwards everything could be
> usefull solving the problem....
>
>
>
> there are a few things I don't understand:
>
> this is Pass2- why don't we use $LFS/tools/bin/gcc but /usr/bin/gcc296 ?
> so I googled for a solution (in the faq I couldn't find this or a
> similar problem)...
>
 Review chapter 4, Final Preparations'.  In particular, did you have
/tools/bin: at the front of your PATH when this happened ?  Also, did
you remember the `set +h' ? - I guess forgetting that would cause bash
to remember it had been used for gcc when you built binutils at the
start.

> what I did afterwards:
> *) set CC=$LFS/tools/bin/gcc...didn't change anything
> *) set PATH=$LFS/tools/bin...caused many other problems

 If you mean you set PATH to only $LFS/tools/bin then yes, at best there
are a lot of essentials missing (the rest of chapter 5).

> *) after resetting $PATH to its old value I removed gcc296 from
> /usr/bin....still getting the same config.log as mentioned above
>

 That I don't understand.  But I hope you have some sort of working gcc
in /usr/bin (your "I removed" comment scares me)  - first you need to
try to identify what you missed, or did differently, in the early
stages.  Then, you probably need to go back to the beginning because
from here it sounds as if everything you've built so far used gcc-2.96,
which is not what you wanted.

 Other things you might want to check - is the /tools symlink correct
(i.e. pointing to /mnt/lfs/tools).  Did you do all of 'adjusting the
toolchain'.  Or maybe at some point you became root to try to fix
things, and never went back to being user lfs ?

Ken
-- 
 das eine Mal als Tragödie, das andere Mal als Farce




More information about the lfs-support mailing list