On slashdot.org, it has been reported that the final release of RedHat-5.2
(as opposed to betas already out) will not make use of Linux kernel v2.2 as
originally, but unofficially, stated. The claim is that RedHat-6.0 will be the
first to include the kernel v2.2, and will be released about 2 months after
the kernel is available and stable. The kernel v2.2 is not released yet, but
is expected Real Soon Now... my bet is 1-2 months.
Last year, several of us informally targeted RH-5.2 as the basis of the next
FNAL Linux release, based on a presentation by RedHat at FNAL. At that time,
RH-5.2 was expected near the end of this year, and would include support for
up to 16 processors, RAID, and real-time functionality, all of which come from
kernel v2.2. With these features unsupported at the level we would like, we
have lost much of the original motivation for investing in developing a FNAL
Linux 5.2.
There are still good reasons to upgrade to RH-5.2. The Gnome desktop and
"drag-n-drop" are maturing, I understand. Egcs 1.1b is well-supported, even if
the C++ compiler portion may still be inadequate relative to the C++ Standard.
Commercial products like Applixware are built against the latest C++ libraries
(libstdc++-2.8) and will not function without the appropriate shared libraries
being installed (I loaded them onto my RH-5.0 machine... one must be careful
to load the new libs in addition to, not in the place of, the old ones). But I
wonder if there is enough motivation to go through an FNAL OS upgrade when
RH-6.0 may be 3-4 months away.
Does someone know of a feature in RH-5.2 that is substantial enough to
motivate the development of a FNAL Linux 5.2, before the arrival of RH-6.0?
Hopefully those brave folks with RH-5.2 beta can tell us the answer.
Thanks,
Rob K.