Fedora Core 2 XFree86 plans, and XFree86 4.4.0 plans, etc.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Since XFree86 4.4.0 is nearing release soon, I have started to
get an increase in the number of questions asked about it, and
about my XFree86 plans for Fedora Core and future development.  

Rather than answer them over and over again in IRC, mailing
lists, and direct email, I thought it would be a better idea to
post this brief summary publically on a few of our lists, in
hopes that you all can help me to get the message around to the
masses.  ;o)  Please DO NOT directly email me or ask me in IRC 
when it will be ready.  Keep this FAQ handy instead, and reread 
it every few days.  ;o)


Q) What release of XFree86 will Fedora Core 2 be shipping?

A) XFree86 4.3.0 is what will ship in Fedora Core 2.  It will
contain a number of bug fixes, Radeon driver fixes and other
improvements to further stabilize the 4.3.0 series, and give
Fedora Core a relatively mature XFree86 base.  I would also 
like to update the "via" driver to Alan Cox's new DRI 
enabled via driver, so that VIA EPIA users can enjoy 3D 
acceleration.  I'm probably going to do a number of other video 
driver updates and scan bugzilla for the most critical issues to 
spend some time on.  Any large-risk issues will likely not be 
addressed until 4.4.0 is integrated into the distro however if I 
believe the risk of regression to be too great to be worth taking 
for a given problem.  I've considered a great number of technical 
and other issues/factors in coming to this decision.


Q) When will XFree86 4.4.0 be officially released?

A) XFree86 4.4.0 is still under development, and it's official 
release date is not known.  XFree86 releases generally miss their 
originally projected release dates by anywhere from 1 month to 6 
months or greater.  XFree86 4.4.0 was planned to be released 
originally on Dec 15, 2003 or so, however it is still under 
development, and a new release candidate was announced earlier 
today or yesterday.  Any release date predictions are generally 
unreliable, as there are countless factors that can delay a 
release from occuring when developers originally intended to 
release the new release.  Generally, this means that many more 
bug fixes are being worked on, and it's better to have an XFree86 
release that is less buggy and released later on at an 
unpredictable time, than earlier and contain many bugs that might 
not get fixed for another year or so.  So the best answer to this 
question is the one people so hate to hear - "When XFree86.org 
thinks it is ready."


Q) Why isn't XFree86 4.4.0 going to be in Fedora Core 2?

A) The simple answer, is that XFree86 4.4.0 has not been released
yet, and the release date is not known or 100% predictable.  
There is a good chance it will be released prior to the release
of Fedora Core, however, in order for it to be integrated and
into the distribution properly, and some of the other changes I
plan on making for 4.4.0, it needs to be done early enough in the
cycle that there is time to work out any major stop-ship problems
that arise, and fix them.  I prefer to ship something known to be
relatively stable, with additional bug fixes, than to ship
something of unknown stability as a rush job, late in the cycle
without having enough proper beta testing occur.  Fedora Core is
intended to be more cutting edge, but I don't want bloody fingers
from rushed packages.  <grin> Besides, users who really want to
use 4.4.0 on Fedora Core 2 (and possibly Fedora Core 1 and older
releases also) will be able to do so very easily when I do make
my 4.4.0 rpms available (see next question).  This will allow
those who want the bleeding edge, to have it and use it, while
those who want the more stable and mature release to get that by
default.  I think this is the best overall solution, especially
since we have no idea when 4.4.0 might be released.  ;o)


Q) When will XFree86 4.4.0 rpm packages become available, and for 
   what Red Hat operating systems will they be available for?

A) I have been working on 4.4.0 packaging for a while now, and
will be releasing rpms sometime soon, but no specific date.  I
want the initial publically posted packages to be in a certain
state, and they're not quite there yet.  They will be available
on ftp://people.redhat.com accessible via FTP and yum, and I will
send out an announcement to xfree86-list@xxxxxxxxxx and
fedora-devel-list@xxxxxxxxxx to announce it with more detailed 
instructions.  The initial builds will be provided as 'rawhide' 
built rpms (but not put into rawhide itself).  At some point, 
if I have enough time, I might also provide Fedora Core 1 
rpms, and perhaps RHL 9 and RHL 8.0 rpms.  No promises, 
however if and when I make any releases, I will announce them 
on the two mailing lists mentioned above.  RPMS will be available 
for whatever architectures happen to compile through to 
completion.


Q) What are the major new improvements and benefits of XFree86 
   4.4.0 over previous releases?

A) The largest user-visible improvements in XFree86 4.4.0 are 
updated video drivers which support most newer video hardware to 
some degree or another, IPv6 support for the X server, 
(hopefully) saner default settings for the server/drivers.  There 
are many other things, but I can't remember it all.  Read the 
XFree86 CHANGELOG document for details.  ;o)


Q) Will there be Radeon IGP DRI support?

A) No.  Use DRI-CVS for Radeon IGP DRI support.


Q) Will there be Radeon 9500/9600/9700/9800 DRI support?

A) No.  The specifications for the 3D engine of that hardware is
not available, and nobody is working on it currently or planning
on doing so to the best of my knowledge.  If the specifications 
were to become available, it would probably be anywhere from 6 to 
12 months for stable DRI support to become available, and require 
funding to develop.  For further details, please search the DRI 
project's mailing list archives for this very frequently asked 
question, as the answer doesn't change over time.  ;o)


Q) Will you ship the Radeon merged framebuffer support from 
   DRI-CVS, so people can use pseudo-Xinerama with DRI?

A) No, it is in experimental CVS for a reason.  Once it is 
integrated into XFree86.org it will become part of a future OS 
release sometime down the road.  Users wanting to test this 
experimental support before then can build DRI-CVS or download 
prebuild drivers from the DRI project directly.


Q) What is the freedesktop.org kdrive xserver?

A) kdrive xserver, also known to many as "The freedesktop.org X
server" has some nice eye candy which Keith Packard has whipped 
up.  This X server is not intended for general purpose usage, but 
is instead an experimental X server intended for developing new X 
technology and testing it out.  It currently caters more to 
active X developers, and those interested in hacking on the 
server itself, however others do play with it as well.  It does 
not support all of the video hardware XFree86's X server 
supports, and it does not have DRI support and other fancy 
features.  It was originally designed years ago for PDAs and 
embedded systems, so it is quite small.  Search google for 
"tinyx" and "kdrive" to learn more, as well as visiting the 
freedesktop.org xserver pages.

Q) Do you plan on shipping it in Fedora Core?

A) Not as a "supported" X server intended for general purpose 
usage, no.  I do think that it has some cool geek factor though, 
and that it would be cool to include it in Fedora Core as a more 
or less "as-is" unsupported thing for people to play with who are 
interested in doing so.  Fedora-Extras/fedora.us might be a 
better place for it though.  If it does not end up there, I'll 
probably have unofficial rpms on people.redhat.com for people to 
play with at some point, and make an announcement here.


Q) Do you plan on using freedesktop.org X libraries?

A) I am currently exploring the possiblity of using the
freedesktop.org X libraries, or at least some of them, however I
have not yet reviewed them all, nor investigated the
integrational issues widely enough to draw any conclusions or
make any commitments at this point.  I do however like the idea
of a less monolithic, more modular X11 build, and I believe
modularized libraries have a number of good benefits both for
development, maintenance, and for end users.  I'm enthusiastic
about the project, and would like to find the time to contribute
more directly to it in the future.  We'll have to wait and see
how things come along.


Q) What do you think about the recent news of X.org and 
XFree86.org merging?

A) I read the Newsforge article, and think that it is encouraging
to see related projects collaborating together more closely.  
Hopefully this will benefit future development of X11 and related 
technology.


I think that pretty much summarizes the majority of questions
I've been getting lately.  If there is something that I have
overlooked, PLEASE DO NOT email me or IRC me directly, but
instead, post a message to xfree86-list, so that myself or
someone else who knows the answer can reply and multiple people
can benefit from the answer.  Of course, please be sure to check
the mailing list archives first, in case the question has been
answered already.

Take care,
TTYL


-- 
Mike A. Harris     ftp://people.redhat.com/mharris
OS Systems Engineer - XFree86 maintainer - Red Hat


_______________________________________________
xfree86-list mailing list
xfree86-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/xfree86-list
IRC: #xfree86 on irc.redhat.com

[Red Hat General]     [Red Hat Watch]     [Red Hat Development]     [Kernel Development]     [Yosemite Camping]

  Powered by Linux