When the uname hack does not suffice (was: Re: Problems installing glibc2-2.3.4 Timezone)

zippo zippo752001 at comcast.net
Thu Aug 5 05:59:15 PDT 2004


That would be time for a mgaic --build=i386-pc-linux-gnu --host=i386-pc-linux-gnu

that should fool it, in every way, Uname -a still says athlon if i remember.


On Thu, 05 Aug 2004 14:54:48 +0200
"Nico R." <n-roeser at gmx.net> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Jeremy Utley wrote:
> [...]
> > Depends on what your host is.  For example, I regularly do builds on
> > my
> > Athlon-XP (arch i686) for my old 486 machines (arch i486).  By the use
> > of the kernel module uname hack, it basically makes the output of
> > uname
> > report the desired arch.  This way, GCC ends up thinking it's running
> > on a VERY FAST i486-class machine.
> 
> This can still be a problem, however I'm not aware of any LFS packages
> breaking due to this, only BLFS stuff.
> If you have some spare time, have a look at the config.guess script in
> the gmp package. My Athlon XP is an i486 for uname when I load the
> uname hack module, but that config.guess wrapper still detects it as
> athlon. :-/
> 
> Can anyone think of a way to get around such problems (except detecting
> all packages that have this "problem" and patch^H^H^H^H^Hhacking them)?
> 
> Could be a problem, or we must block CPUID somehow, it seems.
> - -- 
> Nico
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.3 (GNU/Linux)
> 
> iD8DBQFBEi4fxI5uhYOGv4URAqUWAKCeeL63r1n3PQI5qeufEgUpQsY6nACg5o/o
> v6BgRRMwSSEFIkF/PdoidHU=
> =gOBN
> -----END PGP SIGNATURE-----
> -- 
> http://linuxfromscratch.org/mailman/listinfo/lfs-dev
> FAQ: http://www.linuxfromscratch.org/faq/
> Unsubscribe: See the above information page



More information about the lfs-dev mailing list