Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!newshost.telstra.net!act.news.telstra.net!vic.news.telstra.net!news.mira.net.au!harbinger.cc.monash.edu.au!news.rmit.EDU.AU!news.unimelb.EDU.AU!munnari.OZ.AU!news.ecn.uoknor.edu!news.ysu.edu!usenet.ins.cwru.edu!magnus.acs.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!blackbush.xlink.net!isar.de!augsburg.isar.net!194.45.233.6!roell From: roell@xinside.com (Thomas Roell) Newsgroups: comp.os.linux.development.apps,comp.os.linux.development.system,comp.os.linux.x,comp.os.linux.hardware,comp.os.linux.setup,comp.unix.bsd.386bsd.misc,comp.unix.bsd.bsdi.misc,comp.unix.bsd.netbsd.misc,comp.unix.bsd.freebsd.misc Subject: Re: Sometimes you need X server source (Was: Why to not buy Matrox Date: 03 Apr 1996 16:45:35 GMT Organization: X Inside Inc. Lines: 97 Message-ID: <ROELL.96Apr3184535@blah.xinside.com> References: <4j21ph$crr@slappy.cs.utexas.edu> <ROELL.96Mar30135144@blah.xinside.com> <vwjk9zzy8w1.fsf@osfb.aber.ac.uk> NNTP-Posting-Host: blah.a.isar.de In-reply-to: pcg@aber.ac.uk's message of 01 Apr 1996 21:41:34 +0100 Xref: euryale.cc.adfa.oz.au comp.os.linux.development.apps:14140 comp.os.linux.development.system:20669 comp.os.linux.x:28552 comp.os.linux.hardware:35423 comp.os.linux.setup:48996 comp.unix.bsd.386bsd.misc:447 comp.unix.bsd.bsdi.misc:2987 comp.unix.bsd.netbsd.misc:2750 comp.unix.bsd.freebsd.misc:16638 >roell> X Inside starts with X Consortium code, and NOTHING that has to >roell> do with XF86. > >But isn't a large part of the X Consortium code relating to x86 servers >a previous release of XF86? Probably you ignore it, for I can imagine >yours has a better internal architecture, but still... Nope. Only in xc/programs/Xserver/hw/xfree*. And this is really the HW dependent part. While it might be unecessary big in size, it has really nothing to do with the rest of the X Consortiums code. First thing I did with X11R6.1 was a 'rm -rf xc/programs/Xserver/hw/xfree*', and it still worked on all of my boxes here. >roell> And this is one of the worst offenses you can make to my person >roell> (who happens to write drivers for Accelerated-X), that we would >roell> haven sunken to deeply that we had to use this broken code. > >Well, now you are being a bit excessive: XF86-3 is not polished as I am >confident your work is, but it is not quite "broken". This is a discussion I will not have on a newsgroup. I do consider the architecture of the XF86 server as seriously broken. In order to explain this I have to go very deep into details, which would boar people on the one side and would just invite too many people without a clue to start flaming. I do not consider the hw/xfree* stuff at a level which I would useable for any serious work with our X-Server product. That is biased and a rather personal oppinion. Let's understand it this way and keep it that way. >roell> IMHO, Xfree86 is STOCK X11R6 plus a few sample sample clients >roell> precompiled. > >Well, this impression is not entirely correct: > >* x86 platform support in X11R6 was largely contributed by the XF86 > project; they regularly contribute their current improvements back to > the X Consortium, so that it _becomes_ the stock X11R6. Wrong. Much of the original x86 support was added to X11R5 (and went over to X11R6) by SGCS, which was done by Mark Snitilty and myself. >* The current XF86 release, 3.1.2, does support quite a bit more newer > cards/chips than the XF86 release that got folded back into X11R6, so > it is not quite "stock" X11R6 either. Remember that I restricted 'STOCK X11R6' to the libraries and clients. That's why I really hate the name Xfree86. People use it to refer to both the X-Servers and the rest of the package, which obviously is meant to cause confusion. I refer with Xfree86 the complete package and use XF86 when I want to talk about the individual X-Servers. I don't want to put their work down, but the calculation is really: Xfree86 = X11R.XX + XF86 It's just anyoing for somebody who spend a great deal of working in that area to see people who firmly believe that their window system is Xfree86, while in reality they use X11R6 with the XF86 X-Servers. Lot of uncessesary confusion. >* While essentially all XF86 work is on the servers, the XF86 > precompiled bundles I know include all the standard X clients and > libraries, quite a few contributed ones, and even some extras, > including a few developed for XF86 specifically, and they do amount to > a fairly complete X environment. That's exactly what I'm saying. Although I do obviousely disagree with the impression of 'even some extras'. >roell> is that they have added an X-Server for Intel UNIXes. > >Actually quite a few servers for rather different chipset families, and >a few extra clients related to them, and a lot of integration work, and >they have contributed them back to the X Consortium. Except for the >latter bit, this is what commercial X/Intel package companies have been >doing too. Much off the integration work predates their activities. Actually looking at normal X11R6.1 you have to patch quite a few things to get it compiled directly under FreeBSD. Needless to say that a couple of those so called 'integrations' do cause a lot of problems for other people which then from their positions do consider them as breaking things. >XF86 is the Beetle of x86 X11 packages, and sold at scrap prices (by >weight of CDROM, basically :->); cheap and a bit rough. Your company is >doing the BMW range of x86 X11 servers, and selling it at Polo >prices. *smile* I kind of like that ;-) - Thomas -- Denver Office THOMAS ROELL /\ Das Reh springt hoch, +1(303)298-7478 X INSIDE INC / \/\ das Reh springt weit, 1801 Broadway, Suite 1710 / \ \/\ was soll es tun, Denver, CO 80202 roell@xinside.com / Oelch! \ \ es hat ja Zeit.