[OpenAFS-devel] openafs-kernel-1.0.4-24.1.i386.rpm for Redhat 7.1 is i686, not i386

[Derek Atkins]

It works for me because I specifically setup my build-kernels to
use ARCH=386, which implies it will work on all architectures.
I didn't build the RH 7 packages, so I cannot comment on the
state of the kernels that Derrick used.

-derek

Jeremy Katz <katzj@linuxpower.org> writes:

> --FLPM4o+7JoHGki3m
> Content-Type: text/plain; charset=us-ascii
> Content-Disposition: inline
> Content-Transfer-Encoding: quoted-printable
> 
> On Wednesday, May 09 2001, Todd M. Lewis said:
> > I installed vanilla RedHat 7.1 on two boxes, then installed the
> > openafs-kernel-1.0.4-24.1.i386.rpm (and friends). After necessary
> > modifications to config files for our cell, the first machine work
> > perfectly (thank you), while the second wouldn't load the kernel module.
> >=20
> > The problem turns out to be the machine type (uname -m); the modules
> > seem to work fine on an i686, but not on an i586. RedHat's installer
> > actually installs different kernels appropriate to the specific machine,
> > and the symbols in openafs-kernel-1.0.4-24.1.i386.rpm actually match
> > their i686 kernel.
> 
> Ugh, I was afraid this was going to happen in the stock packages since
> it was happening in mine but Derek said it worked.  Okie, guess I'll
> work on extracting out a patch for the stock spec file so that you can
> build with --target=3Dix86 and get the right thing for i386, i586, and
> i686 openafs-kernel packages.  This will, unfortunately, then require
> that you pick the right openafs package for your architecture (and
> require building for each when the packages for each release is built,
> but it really is just the kernel bits that have to be rebuilt often) but
> there's really not a way around it other than just telling people on
> non-i686 that they have to rebuild. =3D\ =20
> =20
> > After grabbing the source rpm, rebuilding and installing the new rpms on
> > the i586, that machine now works fine too.
> >=20
> > Anyway, if people are having trouble getting the rpms on
> > http://openafs.org/ to load on some boxes, the problem may be a mismatch
> > between the kernel module and which processor their kernels are running
> > on/built for.
> 
> Nope, the SMP problem is a different one that persists through the
> latest ac kernel I tried before exams started.  Not working at NCSU over
> the summer, but I'm going to get back to looking at this once I have
> more than a modem to work with.  My best guess is another struct that's
> different in the ac kernel from the stock kernel that openafs has copied
> in.
> 
> Jeremy
> 
> --=20
> Jeremy Katz
> katzj@linuxpower.org	| jlkatz@eos.ncsu.edu
> http://linuxpower.org	| Developer, NCSU Realm Kit for Red Hat Linux
> GPG fingerprint: 367E 8B6B 5E57 2BDB 972A 4D73 C83C B4E8 89FE 392D
> 
> --FLPM4o+7JoHGki3m
> Content-Type: application/pgp-signature
> Content-Disposition: inline
> 
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.0.4 (GNU/Linux)
> Comment: For info see http://www.gnupg.org
> 
> iD8DBQE6/0tMyDy06In+OS0RAmnUAKDke8kSGS+vrWzF0S4jHh7RhrWcYACgugQp
> i2Mkke4Ur5yAKRj0pXhR1WA=
> =rhcr
> -----END PGP SIGNATURE-----
> 
> --FLPM4o+7JoHGki3m--
> _______________________________________________
> OpenAFS-devel mailing list
> OpenAFS-devel@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-devel

-- 
       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
       warlord@MIT.EDU                        PGP key available