From owner-linux-activists@joker.cs.hut.fi Tue Sep 29 20:54:26 1992
Received: from joker.cs.hut.fi by hutcs.cs.hut.fi with SMTP id AA14127
  (5.65c8/HUTCS-S 1.4 for ); Tue, 29 Sep 1992 20:54:22 +0200
Received: by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA24542; Tue, 29 Sep 92 20:53:32 +0200
Received: from [159.56.1.20] by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA24536; Tue, 29 Sep 92 20:53:26 +0200
Received: from spitfire.raster.Kodak.COM ([129.126.252.216]) 
by cleo.raster.Kodak.COM with SMTP id AA16419
  (5.65c/IDA-1.4.4 for ); Tue, 29 Sep 1992 14:50:11 -0400
Received: from flash.acorn by spitfire.raster.Kodak.COM (4.1/SMI-4.1)
	id AA13179; Tue, 29 Sep 92 14:49:37 EDT
Received: by flash.acorn (4.1/SMI-4.1)
	id AA11716; Tue, 29 Sep 92 14:49:36 EDT
Date: Tue, 29 Sep 92 14:49:36 EDT
From: obz@raster.kodak.com (Orest Zborowski)
Message-Id: <9209291849.AA11716@flash.acorn>
Sender: owner-linux-activists@joker.cs.hut.fi
To: linux-activists@joker.cs.hut.fi
X-Note1: Remember to put 'X-Mn-Key: normal' to your mail body or header
Subject: XFree86-1.0Ya
X-Mn-Key: X11

i have received numerous mailings about the latest xfree86 offering i've
placed on tsx-11. here are a few notes that i hope clarify things:

 o	the package is beta. the real release will be of xfree86 1.1, due out
	about oct 3. currently people are testing out the server which will be
	this release and it looks fine so far.
	i imagine i'll switch the naming to be xfree86 1.1 just so there is
	no confusion, and since the binaries will have been created from the
	stock xfree86 1.1 release, which includes linux.

 o	the package is available on tsx-11.mit.edu under
	pub/linux/packages/X11/private/xfree86-1.0Ya, but you'll need to
	obtain fonts, etc from the x11v2.0 directory in the same location.

 o	the large bin package was broken down into three pieces. it is not
	three separate parts. i wish i had made that clearer. same goes for
	the xfree86 1.0Ya stuff, where i used letters, thinking it was more
	intuitive.

 o	there is a bug on the root disk. the console device has the major
	number of 4 and minor number of 1, which is /dev/tty1. it should be
	(4,0), which is /dev/tty0 (the "current" vt). since i haven't
	upgraded to a new rootdisk since 0.12, i didn't notice. please make
	the adjustment and notify your favorite root disk provider. symptoms
	of this are that the keyboard is dead.

 o	the static libs are in /usr/X386/lib. the "jump" libs are in
	/usr/X386/lib/jump. they are called jump although they are really
	classic format - too much trouble to untangle the data to create
	pure jump libs. i thought that using -L/usr/X386/lib was enough
	to cause gcc to search in that directory for static and shared
	libs but hlu has told me that's not the case. to use the shared
	libs you'll need to make symlinks to the regular jump directory
	from the /usr/X386/lib/jump subdirectory. i don't really like it
	this way but hlu does. changing your config to include
	-L/usr/X386/lib/jump is not a good idea because you won't be
	able to control static vs nonstatic that way - gcc can't tell
	the difference between these libs.

i hope this answers a lot of the questions. as there have not been any
earth shattering mass failures, i expect the binary distribution of
xfree86 1.1 to happen sometime over the weekend.

zorst

From owner-linux-activists@joker.cs.hut.fi Tue Sep 29 23:58:30 1992
Received: from joker.cs.hut.fi by hutcs.cs.hut.fi with SMTP id AA14583
  (5.65c8/HUTCS-S 1.4 for ); Tue, 29 Sep 1992 23:58:27 +0200
Received: by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA27548; Tue, 29 Sep 92 23:57:59 +0159
Received: from twok.ods.com by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA27538; Tue, 29 Sep 92 23:57:40 +0200
Received: by istwok.ods.com id AA27565
	(5.65c/IDA-1.3.5); Tue, 29 Sep 1992 16:57:27 -0500
From: David Engel 
Message-Id: <199209292157.AA27565@istwok.ods.com>
Subject: Re: XFree86-1.0Ya
Sender: owner-linux-activists@joker.cs.hut.fi
To: linux-activists@joker.cs.hut.fi
X-Note1: Remember to put 'X-Mn-Key: normal' to your mail body or header
Date: Tue, 29 Sep 92 16:57:25 CDT
Cc: linux-activists@joker.cs.hut.fi
In-Reply-To: <9209291849.AA11716@flash.acorn>; from "Orest Zborowski" at 
Sep 29, 92 2:49 pm
X-Mailer: ELM [version 2.3 PL11]

X-Mn-Key: X11
X-Mn-Key: GCC

>  o	the static libs are in /usr/X386/lib. the "jump" libs are in
> 	/usr/X386/lib/jump. they are called jump although they are really
> 	classic format - too much trouble to untangle the data to create
> 	pure jump libs. i thought that using -L/usr/X386/lib was enough
> 	to cause gcc to search in that directory for static and shared
> 	libs but hlu has told me that's not the case. to use the shared
> 	libs you'll need to make symlinks to the regular jump directory
> 	from the /usr/X386/lib/jump subdirectory. i don't really like it
> 	this way but hlu does. changing your config to include
> 	-L/usr/X386/lib/jump is not a good idea because you won't be
> 	able to control static vs nonstatic that way - gcc can't tell
> 	the difference between these libs.

Perhaps it's time to modify ld to support the -static/-nojump options
directly instead of having gcc handle them.  What do you think H.J.?

David
-- 
David Engel                        Optical Data Systems, Inc.
david@ods.com                      1101 E. Arapaho Road
(214) 234-6400                     Richardson, TX  75081

From owner-linux-activists@joker.cs.hut.fi Wed Sep 30 00:49:02 1992
Received: from joker.cs.hut.fi by hutcs.cs.hut.fi with SMTP id AA14836
  (5.65c8/HUTCS-S 1.4 for ); Wed, 30 Sep 1992 00:48:54 +0159
Received: by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA28791; Wed, 30 Sep 92 00:48:35 +0200
Received: from dns1.eecs.wsu.edu by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA28785; Wed, 30 Sep 92 00:48:30 +0200
Received: from yardbird.eecs.wsu.edu by dns1.eecs.wsu.edu (16.6/5.910402)
	id AA05155; Tue, 29 Sep 92 15:48:39 -0700
Received:  by yardbird (5.57/2.3-EECS.WSU)
	id AA08318; Tue, 29 Sep 92 15:48:38 -0700
From: hlu@eecs.wsu.edu
Message-Id: <9209292248.AA08318@yardbird>
Subject: Re: XFree86-1.0Ya
Sender: owner-linux-activists@joker.cs.hut.fi
To: linux-activists@joker.cs.hut.fi
X-Note1: Remember to put 'X-Mn-Key: normal' to your mail body or header
Date: Tue, 29 Sep 92 15:48:37 PDT
Cc: obz@raster.kodak.com (Orest Zborowski),
        linux-activists@joker.cs.hut.fi (Linux activists)
In-Reply-To: <199209292157.AA27565@istwok.ods.com>; from "David Engel" at 
Sep 29, 92 4:57 pm
X-Mailer: ELM [version 2.3 PL11]

X-Mn-Key: X11
Mn-Key: GCC
> 
> >  o	the static libs are in /usr/X386/lib. the "jump" libs are in
> > 	/usr/X386/lib/jump. they are called jump although they are really
> > 	classic format - too much trouble to untangle the data to create
> > 	pure jump libs. i thought that using -L/usr/X386/lib was enough
> > 	to cause gcc to search in that directory for static and shared
> > 	libs but hlu has told me that's not the case. to use the shared
> > 	libs you'll need to make symlinks to the regular jump directory
> > 	from the /usr/X386/lib/jump subdirectory. i don't really like it
> > 	this way but hlu does. changing your config to include
> > 	-L/usr/X386/lib/jump is not a good idea because you won't be
> > 	able to control static vs nonstatic that way - gcc can't tell
> > 	the difference between these libs.
> 
> Perhaps it's time to modify ld to support the -static/-nojump options
> directly instead of having gcc handle them.  What do you think H.J.?
> 

I am not so sure. Since X386 is not the only package Linux is going to
have, I don't see a strong reason we have to modify ld for just that.
If people are really interested, I can provide a `specs' for it. It
will be good enough.

BTW, I am still running my own old X11 v2.0. I haven't got time to
download the real thing. I am still thinking I should wait for the
official version.


H.J.
-- 
School of EECS				Internet: hlu@eecs.wsu.edu
Washington State University		BITNET:   60935893@WSUVM1.BITNET
Pullman, WA 99164			Phone:    (509) 335-6470 (O)
USA						  (509) 334-6315 (H)
----------specs------
*asm:


*asm_final:


*cpp:


*cc1:


*cc1plus:


*endfile:


*link:


*lib:
%{g*:-lg} %{!g*:%{!p:%{!pg:-lc}}%{p:-lgmon -lc_p}%{pg:-lgmon -lc_p}}

*startfile:
%{g*:crt0.o%s} %{!g*:%{pg:gcrt0.o%s} %{!pg:%{p:gcrt0.o%s} %{!p:crt0.o%s 
%{!static:%{nojump:-L/usr/lib/shlib/nojump} %{!nojump:-L/usr/lib/shlib/
jump -L/usr/X386/lib/jump}} %{static:-L/usr/X386/lib}}}}

*switches_need_spaces:


*signed_char:
%{funsigned-char:-D__CHAR_UNSIGNED__}

*predefines:
-Dunix -Di386 -Dlinux

*cross_compile:
0

From owner-linux-activists@joker.cs.hut.fi Wed Sep 30 04:56:57 1992
Received: from joker.cs.hut.fi by hutcs.cs.hut.fi with SMTP id AA15337
  (5.65c8/HUTCS-S 1.4 for ); Wed, 30 Sep 1992 04:56:22 +0200
Received: by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA01374; Wed, 30 Sep 92 04:56:09 +0200
Received: from twok.ods.com by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA01368; Wed, 30 Sep 92 04:55:56 +0159
Received: by istwok.ods.com id AA00522
	(5.65c/IDA-1.3.5); Tue, 29 Sep 1992 21:50:42 -0500
From: David Engel 
Message-Id: <199209300250.AA00522@istwok.ods.com>
Subject: Re: XFree86-1.0Ya
Sender: owner-linux-activists@joker.cs.hut.fi
To: linux-activists@joker.cs.hut.fi
X-Note1: Remember to put 'X-Mn-Key: normal' to your mail body or header
Date: Tue, 29 Sep 92 21:50:41 CDT
Cc: obz@raster.kodak.com, linux-activists@joker.cs.hut.fi
In-Reply-To: <9209292248.AA08318@yardbird>; from "hlu@eecs.wsu.edu" at 
Sep 29, 92 3:48 pm
X-Mailer: ELM [version 2.3 PL11]

X-Mn-Key: X11
Mn-Key: GCC

> > Perhaps it's time to modify ld to support the -static/-nojump options
> > directly instead of having gcc handle them.  What do you think H.J.?
> 
> I am not so sure. Since X386 is not the only package Linux is going to
> have, I don't see a strong reason we have to modify ld for just that.
> If people are really interested, I can provide a `specs' for it. It
> will be good enough.

No, you misunderstand.  I am thinking of the way Sun's linker works.
When -lx is specified, ld looks for the library in each directory
specified with -Ldir (the same as other linkers).  However, when
-Bstatic is not specified, ld first looks for a shared library named
libx.sa.* before looking for a normal library named libx.a.  When
-Bstatic is specified, ld does not look for the shared version.

What I am proposing is that we modify the specs file so gcc does not
do anything special with -static/-nojump options but simply passes
them on to ld.  We then modify ld to first look for a jump-table lib
named libx.sa, then a classic, shared lib named libx.ca and finally a
normal lib named libx.a, in that order.  The -static/-nojump options
would be used to tell ld not to look for the shared libs.

This way, we don't have to change the library path when we switch
library styles or change the specs file when we add a new library
package.  All we have to do is specify a single library directory and
add/remove the appropriate -static/-nojump option.  For example, to
link with the shared X libs, we would specify -L/usr/X386/lib and for
the static X libs, we would specify -static -L/usr/X386/lib.

David
-- 
David Engel                        Optical Data Systems, Inc.
david@ods.com                      1101 E. Arapaho Road
(214) 234-6400                     Richardson, TX  75081

From owner-linux-activists@joker.cs.hut.fi Wed Sep 30 12:16:29 1992
Received: from joker.cs.hut.fi by hutcs.cs.hut.fi with SMTP id AA18374
  (5.65c8/HUTCS-S 1.4 for ); Wed, 30 Sep 1992 12:16:26 +0200
Received: by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA10618; Wed, 30 Sep 92 12:13:55 +0159
Received: from mailgate.ericsson.se by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA10612; Wed, 30 Sep 92 12:13:50 +0159
Received: from  (deep-thought.ericsson.se) by mailgate.ericsson.se 
(4.1/SMI-4.1-MAILGATE1.10)
	id AA25935; Wed, 30 Sep 92 11:14:01 +0100
Received: from jerry.ericsson.se by  (4.1/SMI-4.1-LME1.6)
	id AA28751; Wed, 30 Sep 92 11:15:44 BST
From: etlmike@deep-thought.ericsson.se (Mike Wilcox)
Received: by jerry.ericsson.se (4.1/client-1.3)
	id AA22392; Wed, 30 Sep 92 11:15:43 BST
Date: Wed, 30 Sep 92 11:15:43 BST
Message-Id: <9209301015.AA22392@jerry.ericsson.se>
Sender: owner-linux-activists@joker.cs.hut.fi
To: linux-activists@joker.cs.hut.fi
X-Note1: Remember to put 'X-Mn-Key: normal' to your mail body or header
Subject: XFree86-1.0Ya + Trident 8900c

X-Mn-Key: X11

Hi guys,

Just to let people know:

I have had 1.0Y working with a Trident 8900c based card, and used both
the mono and colour servers.

Equipment:  Cheap (absolutely no-name. Not even a copyright in the manual)
            8900c Trident card, 1Mb memory.
            Cheap (CA&G 1460) monitor, fixed frequency handling.
            Can cope with 1024x768 interlaced.
            Running on a not-quite so cheap 486SX, 4Mb, 8Mb Swap.
            Kernel: 0.97pl4

Performance: (all subjective) Pretty good on both servers.
             There doesn't seem to be much degradation to performance 
             in using the colour server over the mono.

I have also compiled up the olvwm window manager, which required a few changes
to the source on tsx-11.

After a lot of hassle where absolutely nothing would get displayed on
the screen (but I could still execute commands), I finally got it to
work by:

1)  Changing nothing in the "vga2" section, except the
    allowed modes.

2)  Adding the chipset specification of "tvga8900@ to the "vga256" section,
    and adding to the allowed modes. I do not have a "clocks"
    specification here.

3)  Adding the following mode definitions:

640x480     25     640  680  776  800     480  491  493  525
800x600     36     800  848  920 1024     600  601  603  625
1024x768i   45    1024 1064 1224 1264     768  777  785  817 interlace

    The 640x480 mode spec works for both servers.

Now, with the colour server, I originally had the problem that nothing
would appear on the screen. Well, maybe just my reflection....
After reading the man pages, I tried adding "+hsync" or "-hsync"
to the flags field, as appeared appropriate from my display manual.
Hey presto it all worked. Take them away again, and it still worked!
Strange things....

Note that when I tried to add both [+-]hsync and [+-]vsync, the X server
gave up with some error regarding these flags. I couldn't really follow
what it meant anyway.

Now I have one problem with the 1024 X 768 mode:

I know my monitor can only deal with this resolution when interlaced, so I
added the "interlace" flag. When I swap to this mode, I get the
desktop appearing about 1.5 time, with a block (of about 3cm height) 
of squeare-patterned grey noise between the two copies of the desktop.

So, I experimented by trying to remove the interlace flag, and using that,
it kind of works. After fiddling with the position/size controls
on the monitor, I can see that I am getting almost all of the desktop,
with slight cropping around the sides. It flickers to quite an extent,
but I expect some at this resolution.

Can anyone explain why I can't seem to get the thing to display correctly
with the interlace flag? The timing tutorials don't cover this subject:)

Another question - how can I get the clocks information back out of the
server, once it has started up?

So, if anyone has answers, or needs help, please feel free to e-mail

Mike
-- 
Mike Wilcox                    Guildford, Surrey, England.
CME 20 Design Dept.            Tel:   +44-483-303666
Ericsson Cellular Systems.     Email: etlmike@deep-thought.ericsson.se

From owner-linux-activists@joker.cs.hut.fi Sun Oct  4 13:40:53 1992
Status: RO
X-VM-v5-Data: ([nil nil nil nil nil nil nil nil nil]
	["1773" "Sun" "4" "October" "92" "4:40:08" "PDT" "David Black" 
"dlbb0@diag.amdahl.com " nil "41" "X11v2.0 questions" "^From:" nil nil "10"])
Received: from joker.cs.hut.fi by hutcs.cs.hut.fi with SMTP id AA20139
  (5.65c8/HUTCS-S 1.4 for ); Sun, 4 Oct 1992 13:40:41 +0200
Received: by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA00627; Sun, 4 Oct 92 13:40:17 +0200
Received: from mcsun.EU.net by joker.cs.hut.fi (5.65b/6.8/S-TeKoLa)
	id AA00616; Sun, 4 Oct 92 13:40:11 +0200
Received: from charon.amdahl.com by mcsun.EU.net with SMTP
	id AA09943 (5.65b/CWI-2.177); Sun, 4 Oct 1992 12:40:22 +0100
Received: from amdahl.uts.amdahl.com (sde-hy.uts.amdahl.com) by charon.amdahl.com 
(4.0/SMI-4.1/DNS)
	id AA23156; Sun, 4 Oct 92 04:40:13 PDT
Received: by amdahl.uts.amdahl.com (/\../\ Smail3.1.14.4 #14.11)
	id ; Sun, 4 Oct 92 04:38 PDT
Received: by juts.ccc.amdahl.com (/\../\ Smail3.1.14.4 #14.6)
	id ; Sun, 4 Oct 92 04:39 PDT
Received: from brain.diag.amdahl.com by chicago.pswd.amdahl.com (4.0/SMI-4.1)
	id AA29811; Sun, 4 Oct 92 04:40:10 PDT
Received: from gyruss.pswd.amdahl.com by brain.diag.amdahl.com (4.1/SMI-4.1)
	id AA26060; Sun, 4 Oct 92 04:40:09 PDT
Message-Id: <9210041140.AA26060@brain.diag.amdahl.com>
Sender: owner-linux-activists@joker.cs.hut.fi
X-Note1: Remember to put 'X-Mn-Key: normal' to your mail body or header
X-Mailer: ELM [version 2.3 PL11]
From: dlbb0@diag.amdahl.com (David Black)
To: linux-activists@joker.cs.hut.fi
Subject: X11v2.0 questions
Date: Sun, 4 Oct 92 4:40:08 PDT

X-Mn-Key: X11

Hi, I've been using x11v2.0 for a few weeks now, and find it usable and
an improvement over previous versions. But there have been some odd
behaviors which have surfaced which I'd like to know if have been
addressed:

          	- After initially starting xdm (like after reboot), logging in
          and back out twice leaves the server in a wild video mode.
	  This cycles back to normal with two more login/logouts.
	  I've seen these strange video modes when experimenting with
	  clock indices which aren't on my card (like past number 16)...

	- X386mono won't run with my vga256 resolution settings,
	  complains about not enough memory for requested virtual size.
	  It will start up in 640x480 with some coaxing but is mostly
	  scrambled on the screen. So it's basically not usable. My
	  video card is an ET4000AX-based board which works beautifully
	  with the color server.

	- In the server kit, there is a cbrk.o file somebody wants and
	  it doesn't seem to be around. (Resolution of _cbrk reference.)
	  The path to is ddx//cbrk.o. Prevents linking of
	  new servers.

	- Actually since X386 v1.2, some games (acm and xpilot
	  to name two) don't draw the screen correctly. It appears as if
	  the correct drawing order of objects is reversed, which results
	  a lot of flickering at best, and blank screen at worst.

	- A friend and I have been working on getting xpilot to run and
	  find that X is barfing on XEventQueue or similar call. Basically
	  something that should work but doesn't. We're going to try and
	  synthesize a similar function to get around this if no other
	  solution comes our way.

I just joined the list, so I apologize if these are well-known problems.
In any case, help is requested. Thanks!

Dave