The short story:
* the byteorder patch has been tested, fixed, and ported back to 2.0
* the ufs patch is untouched; requires byteorder patch to be applied, though.
* both are available separately as either patch or tgz,
for either 2.0 or 2.1 in recent releases, at my site:
http://www.eleves.ens.fr:8080/home/rideau/files/
The long story:
it happens that my byteorder+ufs patches for 2.0.x had ufs working fine,
but broke ntohl for userland compilations; hence I took the byteorder code
from my 2.1 patch into the 2.0 one, and found a (stupid) bug in the latter:
I had exchanged definitions for htons and ntohl! <ashamed> in the 2.1 patch.
Conclusion: 2.1 is fixed, and 2.0 benefits from clean 2.1 byteorder code,
with all the bells and whistles and correct POSIX namespace.
The patches should work on any recent 2.0 or 2.1 kernels (respectively),
as no one else touched the code in the same files as I did.
No new feature, only bug fixes and cleanups;
no algorithmic reorganization (no breakthrough, but no instability).
The patches are straigthforward, and I see no reason no to integrate it
in the mainstream kernels, unless someone can argue that bugs remain;
I doubt there is anymore bug in the byteorder patch: I wrote a test program
for all byteorder functions, and it works great (at least on i386);
its code is the same for 2.0 and 2.1.
ufs has been well tested under 2.0 and was untouched since. Under 2.1,
it wasn't much tested, but no significant changes were done over 2.0
(only cleanups and adaptation to new VFS interface),
so it should work; moreover, it fixes the fact that mainstream 2.1 ufs
doesn't even compile (maybe tag ufs "experimental" in 2.1?).
Feedback welcome.
Regards,
== Faré -=- (FR) François-René Rideau -=- (VN) Уng-Vû Bân -=- rideau@ens.fr ==
Join a project for a free reflective computing system! | 6 rue Augustin Thierry
TUNES is a Useful, Not Expedient System. | 75019 PARIS FRANCE
http://www.eleves.ens.fr:8080/home/rideau/Tunes/ -=- Reflection&Cybernethics ==