One of the most important tasks in kernel maintenance is not the addition
of new code, but removal of old code that is no longer useful. Unused code
bloats the kernel and, potentially, becomes a breeding ground for bugs and
security problems. Getting that code out of the way helps keep the kernel
cruft level down.
In recent times, the ax has fallen on two subsystems. The first is the InterMezzo filesystem, which has
been removed for 2.6.7. InterMezzo is a distributed filesystem from Peter
Braam and company with a number of interesting ideas, but, apparently, few
users. Maintenance has been lacking, and Mr. Braam finally agreed that it should be removed, noting
"In the past 4 years nobody has supported InterMezzo sufficiently for
it to become successful." The Lustre
filesystem, which is Mr. Braam's current project, appears to be headed for
greater success.
A patch has been posted which removes
support for the PC9800 architecture. There have been a few small
objections to this removal, drawing this
response from Alexander Viro:
So are you volunteering to maintain the port? Maintainers are MIA;
the damn thing doesn't compile; all patches it gets are basically
blind ones ("we have that API change, this ought to take care of
those drivers and let's hope that possible mistakes will be caught
by testers"). Considering the lack of testers (kinda hard to test
something that refuses to build), the above actually spells in one
word: "bitrot".
There has been a rather conspicuous shortage of people stepping up to
maintain the PC9800 port, so chances are that it will be going away
soon.
Ahem, yes Peter and Cluster FileSystems (CFS) were able to get good support for Lustre; but they have now taken Lustre's stable 1.2.x series closed (aka dual-licensed); Lustre 1.2.x adds support for Linux 2.6 (1.0.x is bound to 2.4). Lustre 1.0.x is all CFS is willing to make available via the GPL (well until the staggered GPL release of 1.2.x happens a year from now). CFS isn't even making the various Lustre 1.2.x in-tree kernel changes widely available (also Lustre 1.2.x modules still claim to be GPL as per modinfo.)
CFS is doing some good things with Lustre but they have pulled a bait and switch with the notion of Lustre actually being purely GPL'd. Everybody needs to make money (there were all sorts of GFS-like warning signs) but it still leaves a bitter taste in one's mouth. Hopefully CFS makes gobs of money (shouldn't be hard given CFS's per client pricing) and can then justify making the core Lustre filesystem purely open. Maybe their closed administrative tools, consulting/support and such will be the value add they would need to remain successful?
So there is hope, if this were to happen Lustre could very well be re-released with a GPL license that actually sticks (e.g. code makes it into Lustre that isn't sole-sourced from CFS).
