Dec 122017
 
Linux news from the net.
File INTLX020.ZIP from The Programmer’s Corner in
Category UNIX Files
Linux news from the net.
File Name File Size Zip Size Zip Type
INTLX020.TXT 352275 133891 deflated

Download File INTLX020.ZIP Here

Contents of the INTLX020.TXT file


From: [email protected] (Jim Burns)

Subject: Re: Cross compilers-a question
Message-ID: <[email protected]>
Date: 8 May 92 03:13:15 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 11

in article <[email protected]>, [email protected]
FI (Risto Kankkunen) says:

> Anyway, with gcc 2.1 and binutils 1.94, it doesn't matter, wheter the
> host is big or little endian.

How is that accomplished?
--
BURNS,JIM (returned student)
Georgia Institute of Technology, 30178 Georgia Tech Station,
Atlanta Georgia, 30332 | Internet: [email protected]
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt0178a


[next article]
From: [email protected] (Michael K. Johnson)

Subject: Re: GCC 2.1, probably FAQ, but . . .
Message-ID: <[email protected]>
Date: 8 May 92 03:35:30 GMT
References: <[email protected]> <[email protected]
itoba.ca>
Sender: [email protected]
Followup-To: comp.os.linux
Organization: St. Olaf College; Northfield, MN USA
Lines: 18
In-Reply-To: [email protected]'s message of Fri, 8 May 1992 00:00:12 GMT


[...]
Same here. Could some kind soul _post_ the solution, so that some of us
can use gcc 2.1. I guess now is the time to put gcc 1.4 back ....
I need it to compile my program.

This /is/ a FAQ.

Get the new tar and compress from
tsx-11.mit.edu:/pub/linux/binaries/usr.bin/{tar,compress}
and use them to re-install GCC 2.1. The tar and compress on the root
disk for .95a are bad.

the FAQ is available as
tsx-11.mit.edu:/pub/linux/docs/FAQ

michaelkjohnson
[email protected]


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: 33Mhz 386 and IDE hard drives
Message-ID: <[email protected]>
Date: 8 May 92 04:43:33 GMT
References: <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 57
Nntp-Posting-Host: romeo.cs.colorado.edu

In article <[email protected]> [email protected] (Crai
g Metz) writes:
>
> With all the people complaining about how their 33Mhz machine won't
>work with their hard drive unde Linux, I'd like to voice the opposite. I have
>a SCI RAM 386/33Mhz (Greenleaf motherboard) and a Seagate ST1144A, and al
>versions from 0.12 to 0.95c+ work flawlessly at the full 33Mhz. At least it
>works for somebody...
>
> -Craig


Ok, I'll throw in my success story - my 386-33 works fine with both
my cached MFM controller (yes, it has a hardware lookahead buffer),
and my ST-02 SCSI adapter. My printer also works, shared libraries
work, everything works.

Linux works for MOST people. The SCSI drivers work for MOST people.
The printer driver works for MOST people. You don't hear about us
because it works for us, we have no beef, and would rather
play with Linux than write news posts about it 8^)

Now, I don't know how many people run linux (we should set up a
mail server that counts people automatically, and have the
same done on Fido net), but there are a lot.

- In the CFV for comp.os.linux, there were over 800 yes votes.
- Most days I read comp.os.linux twice, and get ~60 messages
the first time, 30 the second.

That shows that a lot of netters use Linux. Add to that
everyone who is getting Linux through Fido net, or second hand
from a netter, and I'm sure it's a sizeable number.

Out of those, a small percentage are bound to have problems.

Most of the people working on Linux are at Universities,
and because of their academic status don't have X different machines
to test on, with full "official" technical documentation on important
goodies like disk controllers. Given this lack of information
(I think all Unix developers have this problem - the 4.3 Reno VAX UDA-50
driver code has a comment to the effect of "I'm not quite sure how
this works, but I heard from someone who had the DEC hardware manual
that.....") there are lots of openings where tolerances aren't just
right. Also, on workstations, and larger computers, things are
more propriety. You worry about one or two disk controllers,
and it works. There isn't as much variation. PC's are commodity
hardware, cloned by everyone. There is more variation, and room
for small timing problems to get in the way. While on a Vax,
you run it on one UDA-50 disk controller, it works there, it
will work on any UDA-50 disk controller, on a PC you have your
IDE, ESDI, MFM, RLL controllers, all "compatable" at the register level,
made by inumerable different manufacturers.

Finally, if you notice the version number, it is still < 1.0 8^)

I'd say given all that Linus and everyone have done an excellent
job.


[next article]
From: [email protected] (Jim Burns)

Subject: Re: elle(1) and rc(1) ports uploaded to tsx-11.mit.edu
Summary: problems compiling - which libs needed?
Message-ID: <[email protected]>
Date: 8 May 92 04:55:06 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 27

In article <[email protected]>, [email protected] (Bennett
E. Todd) writes:
> I uploaded my ports of elle and rc to tsx-11.mit.edu; they are in
> ~ftp/pub/linux/sources/usr.bin. The files are elle.src.tar.Z,
> elle.src.diffs.Z, rc.src.tar.Z, and rc.src.diffs.Z. In both cases the
> src.tar.Z file is the original sources from before I started my Linux
> port, and the src.diffs.Z file is suitable (after uncompressing) for
> feeding into patch -p1.

Looks interesting. I applied the patches and did make - a few sources had
warnings, but the libraries made. Xelle didn't - bunch of undefineds. I
added -ltermcap to the compile line, which reduced it to three undefineds.
What library do I need, and where do I get it? Gcc 2.1, 0.96-pre. Tail end
of output follows (thanx):

echo elib.a
elib.a
cc -o xelle eemain.o elib.a libsb.a -ltermcap
elib.a(eefill.o): Undefined symbol _chballoc referenced from text segment
elib.a(eefill.o): Undefined symbol _chbis referenced from text segment
elib.a(eefill.o): Undefined symbol _chbit referenced from text segment
make[1]: *** [B-elle] Error 1
make: *** [elle] Error 1
--
BURNS,JIM (returned student)
Georgia Institute of Technology, 30178 Georgia Tech Station,
Atlanta Georgia, 30332 | Internet: [email protected]
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt0178a


[next article]
From: [email protected] (M. Saggaf)

Subject: X and 9600 bps
Message-ID: <[email protected]>
Date: 8 May 92 04:56:59 GMT
Sender: [email protected] (News system)
Organization: Massachusetts Institute of Technology
Lines: 11
Nntp-Posting-Host: e40-008-5.mit.edu


I have a problem using my modem at 9600 bps under X. I'd lose so many
characters that communicating at that speed is unusable. No problem
with 2400 bps. Is anybody having any success using his/her modem
beyond 2400 baud under X? Please let me know. (Is there a fix, like
issuing some stty command or something?). Also, tcsh does not work
under X.

----------------
M. Saggaf
alsaggaf2athena.mit.edu


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: Diamond Speedstar
Message-ID: <[email protected]>
Date: 8 May 92 05:16:55 GMT
References: <[email protected]> <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 79
Nntp-Posting-Host: romeo.cs.colorado.edu

In article <[email protected]> [email protected] (Kevin C
ummings) writes:
>In article <[email protected]>, [email protected] (Rob Coleman) wri
tes:
>> A number of people on this newsgroup have spoken of the advantages of
>> getting an ET4000 based card for use with X. Is the "stealth" version
>> of the Diamond Speedstar a better (read: faster, more useful) card to
>> get than the "regular" Diamond Speedstar+? The Stealth is supposed to
>> be Diamond's version of an S3 coprocessed card - is this still register
>> compatible with the lesser non-coprocessed card? The price difference
>> between the two cards is $169 vs. about $259 for the Stealth. Is the
>> difference worth the extra expense?
>
>Being the proud owner of a brand spanking new (OK, its a month and a half
>old already!) Diamond Stealth VRAM video card, let me say this about that:
>
>Yes, the Stealth is register level compatible with any other clone VGA
^^^^
>card. Programming the Stealth at the register level, it is not as fast


X386 doesn't run on VGA cards. Only SVGA cards with a Tseng ET3000/4000
based chipset, and some paradise and genoa? chipsets.

>Stealth drivers for windows. If I then put up a LARGE window containing
^^^^^^

It won't run anything greater than standard VGA without a driver though,
and thus, without an appropriate s3 driver (the only ones
that exist right now for X are COMMERCIAL) it won't work with X period.

>
>If you won't be using your new video card under MS-windows, or X-windows
^^^^^^^^

To my knowledge, a free driver for x386 does not exist yet.
There are tens of thousands of netters in the world, and this could
be wrong - I'd been told the same thing about the Trident
driver, and it turns out there is a free driver.

>On the other hand, the Stealth is capable of producing a 1280x1024x16
^^^^^^^^^^^^
With x386, this is a non-issue, as the code only supports 256 byte
PACKED graphics. Most SVGA modes use a PLANAR 16 color mode
(compatability with EGA/VGA) which is radically different.

>interlaced screen while the SpeedStar+ is not. I beleive that both
>cards are 1024x768x256 capable using either interlaced or non-interlaced
>monitors.
>
>Make an imformed decision, and base it on your needs (or wants B^).
>If you will be a windows user, the Stealth is the way to go. If you
^^^^^^^
Microsoft Windows, or commercial Xservers that support the S3 only.

>want the 1280x1024 capability today, the Stealth is the way to go.
>
>If you need/want the 800x600x32K mode, the SPeedStar+ is the way to go.
>
>If you don't care about any of the above, the SPeedStar+ is cheaper.

There are also a number of generic ET4000 based boards available for
~$120

>Good luck with your decision.

People have expressed a desire to write S3 drivers. The time frame
for this is unclear though, and if these parties have not written
serious code using the s3 before at the register level, there
will be problems, and the amount of time will be much longer -

I speak from experience on this. Under pressure, I've designed, written and
debugged 2000 line programs / libraries / etc in three days, albeit a very long
and stressful period of three days.

In comparison :

The generic SCSI drivers (Now getting close to 2000 lines of C code and
header) took several months to finish because of other commitments, and
my lack of knowledge about SCSI and the Seagate hardware at the
register level.


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: Linux-Activists Digest #145
Message-ID: <[email protected]>
Date: 8 May 92 05:28:17 GMT
References: <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 29
Nntp-Posting-Host: romeo.cs.colorado.edu

In article <[email protected]> [email protected]
EDDIE.MIT.EDU (Scott Beckstead) writes:
>
>
> Ok I've seen a lot of discussion about wing-ding high fallutin Svga
>thinga ma bobbers but I need to know if X will run with a lowly hercules
>card. I don't really need the color and I like having an X environment
>

The free X386 runs on Tseng et4000 and et3000 cards, some paradise, and
genoa cards.

This is because Thomas Roell's x386 server is written to work with
SVGA cards that deal with stuff in 256 colors, packed not planar
format.

>to work in. Besides I haven't the cash to go out and by a $1000 +
>worth of monitor and card. Please tell me it will. It works on a sun 2
>Scott

TSENG Et4000 board : Say $120 - $170 for a REAL Dymond speedstar

Greyscale VGA monitors (good for 640 x 480) can be found for
$150. I paid ~$350 for my 14" color Samsung.

>with a monochrome screen why no a hercy.

A big (1280 x 1024) monochrome monitor on a workstation is not
uncommon. Hercules on the class of PC that Unix, and
X Windows is, and a free driver wasn't written.


[next article]
From: [email protected] (H. Peter Anvin N9ITP)
Newsgroups: comp.os.msdos.programmer,comp.unix.bsd,comp.unix.msdos,comp.os.linux
,comp.sys.ibm.pc.hardware

Subject: WANTED: Partition ID table
Message-ID: <[email protected]>
Date: 8 May 92 05:56:38 GMT
Sender: [email protected] (Usenet on news.acns)
Reply-To: [email protected] (Peter Anvin)
Followup-To: comp.os.msdos.programmer
Organization: Northwestern University, Evanston, Illinois, USA
Lines: 15

Dear netters:

A few weeks ago someone circulated a list of PC/MS-DOS hard disk partition
ID bytes on the net. I have tried archie and several FTP archives trying
to find this list, but to no avail.

If you have a copy of this list, please post it or mail it to me. It would
be greatly appreciated. If you could give me a pointer to the latest
edition, that would be even more appreciated.

Thanks in advance,

/Peter Anvin

Followups are directed to comp.os.msdos.programmer.


[next article]
From: [email protected] (Toshiyasu Morita)

Subject: Need minor help!
Keywords: Linux help
Message-ID: <[email protected]>
Date: 8 May 92 05:26:04 GMT
Organization: Netcom - Online Communication Services (408 241-9760 guest)
Lines: 21

I have a few problems with Linux:

1) I have a really bad tendency to "cat" large files and then hit
"scroll-lock" to stop the screen. Unfortunatly under Linux this
seems to screw up the keyboard driver somehow - it ignores
most keypresses aftewards and occasionally prints "reverse-i-search".
The other VCs are similarly confused, so I have to wait a minute
or so for sync to kick in, then reset. How can I keep "scroll-lock"
from confusing Linux?

2) When makefiles call a text file Linux doesn't automagically invoke
sh. For example, the GCC 2.1 makefile uses a text file called
"copy-if-changed" (I think) to copy a file if it's been changed.
The system doesn't like it and the make stops. Currently I have to
insert a "sh " before every "copy-if-changed" to invoke sh manually.
What's wrong? (I have bash-1.12 installed, if it's imporant)

Thanks in advance...

Toshi Morita
[email protected]


[next article]
From: [email protected] (Linus Benedict Torvalds)

Subject: Re: fsck and inode
Keywords: filesystem,inode
Message-ID: <[email protected]>
Date: 8 May 92 05:55:03 GMT
References:
Organization: University of Helsinki
Lines: 30

In article [email protected] (Rick Wong) writes:
>
>I am new to filesystems in unix, so please bear with me.
>
>when I ran the following command
>fsck -m /dev/hda7
>I got these output
>
>inode 266 not cleared
[ etc ]

This is normal under linux: the "-m" flag stands for minix, and means
fsck will do extra checking in the inode like the minix fsck does. The
linux filesystem is almost completely the same as the minix one, but
there are a few very minor differences. The main one is that linux
doesn't clear the i_mode field of an inode that is no longer used, and
minix does.

The result is that if you use minix' fsck, it gives warnings of the
above type after you have used linux for a while. Just so I could get
the same warnings if I wanted, I added the '-m' switch - it's there but
you aren't supposed to use it.

This was mentioned in early readme's, when you actually had to use minix
to set linux up, and people probably used the two systems side by side.
Now I assume most people are running just one or the other, so I haven't
even mentioned it in the readme's any more, as the feature no longer has
any meaning.

Linus


[next article]
From: [email protected] (David W. Summers)

Subject: Shared Libs (now fixed! Thanks!)
Summary: They are now fixed. Thanks, everyone who replied!
Keywords: Shared Libs
Message-ID: <[email protected]>
Date: 8 May 92 03:08:12 GMT
Sender: [email protected] (NetNews Administrator)
Organization: University of Arkansas
Lines: 20
Nntp-Posting-Host: uafsun4.uark.edu

Thanks to several people, my shared library problem is now fixed.

I had accidentally erased the shared library '/lib/lib92.04.06' and
most of my compiled programs were crashing.

Several people suggested that I re-boot with the installation floppy and
do a 'ln -s /usr/shared/lib/lib92.04.06 /lib'. I did that and my
problem is now fixed.

I PROBABLY could have figured it out, but it was sure faster to
get "expert" opinions and to know exactly where to look!

Thanks, Net People!
- David Summers
([email protected])

--
"Never under-estimate the bandwidth of a station-wagon
David Summers full of tapes, hurtling down the highway."
[email protected] - Tanenbaum, "Computer Networks"


[next article]
From: [email protected] (Duco Fijma)

Subject: mount lock files
Message-ID: <[email protected]>
Date: 8 May 92 09:17:47 GMT
Sender: [email protected]
Organization: University of Twente, Dept. of Computer Science
Lines: 12
Nntp-Posting-Host: utis143

Hello,

A little question about Linux mount. When logged in as root, mounting a floppy
is no+ problem. However, when logged in as a user, mount complains 'unable to
create lock file'. Something is wrong with permissions. But where in the file-
system does mount tries to create this lock-file? I could not find anything
that looked like a lock-file after mounting a floppy as root.

Duco Fijma
University of Twente
Enschede
the Netherlands


[next article]
From: [email protected] (Dylan Smith)

Subject: Re: Fidonet Mailers and timezone.
Keywords: fidonet,timezone
Message-ID: <[email protected]>
Date: 8 May 92 09:16:31 GMT
References: <[email protected]>
Sender: [email protected] (Usenet poster for nntp (tj - 17/12/91))
Organization: Transputer Centre, Bristol Polytechnic.
Lines: 24
Nntp-Posting-Host: hal-9000

In article <[email protected]> [email protected]
k (Chris Watts) writes:

>Does anybody have any alternative fidonet software for linux.

There is BinkleyTerm available - check out the fidonet.org newsgroup where
they publish Fidonews. At the end of Fidonews there is a list of software
available for various machines, and I think there's a UNIX (if not xenix)
section.

Failing that, get a cheapo XT and run Binkley on a PC, and use SLIP (on a
timed event on the PC, and on a cron on Linux) to transfer the mail.
Depending on how much Fidonet mail you want, you may be able to get away
with a 720K single floppy PC on the PC end (put it this way, I can get
away with it so long as I move all the mail off straight after it comes
in!)

BTW: I haven't tried putting Fidonet stuff on Linux yet (even from a PC)
but I intend to.



--
Email : JANET [email protected] | Everywhere else [email protected]
[email protected] | [email protected]


[next article]
From: [email protected] (Tommy Frandsen)

Subject: Re: vgatest works if in 80 x 25 mode
Message-ID: <[email protected]>
Date: 8 May 92 07:05:40 GMT
References: <[email protected]>
Sender: [email protected]
Organization: Department of Computer Science, U of Copenhagen
Lines: 22

[email protected] writes:

>vgatest works if in 80 x 25 mode. If set TM-4000 chipset to 132 x 28
>mode on bootup, vgatest is reduced in vertical height and doesn't sync
>horizontally. sync vertically VGA card is Compuadd Hi-Rez with NEC 2A
>(factory repaired, works okay 640x480 but fails to sync in 800 x 600
>mode ---- VGA card spec 35.5 kHz and monitor 35 kHz ---- previous VGA
>card (deceased) ran at 35.1 kHz I believe ) Nothing like standards.

Yes, I was a little to optimistic about the use of SVGA text modes.
It seems that it is not always possible to shift from a SVGA text
mode to a standard VGA graphics mode just by writing to the standard
VGA registers. You must also put something into the SVGA extension
registers. Since that depends on the particulary SVGA card, I can not
give a general solution (I'm currently trying to solve the problem for
my own ATI adapter).

Since the 80 x 50 mode dosen't require SVGA, you can use this text mode
also. If anyone has used vgatest with succes in a SVGA text mode, I would
be happy to hear about it. Please send mail to:

[email protected] (Tommy Frandsen)


[next article]
From: [email protected] (Tommy Frandsen)

Subject: Re: VGA graphics library for Linux
Message-ID: <[email protected]>
Date: 8 May 92 07:21:37 GMT
References: <[email protected]> <[email protected]
k.TU-Muenchen.DE>
Sender: [email protected]
Organization: Department of Computer Science, U of Copenhagen
Lines: 11

[email protected] (The Master of Symbolic Links) writes:

>I wouldn't hardcode everything like you did. There are too many incompatible
>SVGAs around. Why not using the VESA SVPMI standard for getting infos
>about the SVGA and how to set displaymodes ?

Yes, it would be really nice if would could use the VESA standard, but
this requires acces to the video BIOS, which is not possible in Linux.
I'm afraid we can't avoid hardcoding.

[email protected] (Tommy Frandsen)


[next article]
From: [email protected] (Henning Spruth)

Subject: Re: X386, Shared Libs, and OS/2 2.0 Boot Manager, etc.
Keywords: X386 OS/2 2.0 BOot Manager
Message-ID:
Date: 8 May 92 08:17:29 GMT
References: <[email protected]>
Sender: [email protected]
Lines: 22

[email protected] (David W. Summers) writes:


>(2) I posted last week about wanting to be able to boot Linux from
> OS/2 2.0 Boot Manager.

>I tried to install ShoeLace in a OS/2 2.0 Primary partition but all it
>does is scroll "8000:0000" or something similar and hit on the disk a lot.

>Does anyone have any ideas? I was successfull installing ShoeLace on
>an IDE drive with Linux 0.12 back in January or so, so I assume it is
>something to do with SCSI.

I've had not problems at all using linux, shoelace, and the OS/2 boot manager
together.

Henning
--
------------------------------------------------------------------------
Henning Spruth [email protected]
Institute of Electronic Design Automation
Technical University of Munich, Germany


[next article]
From: [email protected] (Stephen R. van den Berg)

Subject: Re: Password encryption bogus?
Message-ID: <[email protected]>
Date: 8 May 92 11:11:52 GMT
References: <[email protected]>
Sender: [email protected] (Newsfiles Owner)
Organization: RBI - RWTH Aachen
Lines: 23
Originator: [email protected]
Nntp-Posting-Host: ikki

In article <[email protected]> [email protected] writes:

>[email protected] (Karl R. Buck) writes:
>>I noticed something strange while loggin in today:

>>The last letter in my password is a capital letter. When I log in it
>>will accept both the capital and lowercase letters as valid and let me
>>in. I have not tested this any further to see if it does this for
>>other types of passwords.

>How *LONG* is your password, Karl? I can't say this for certain
>about Linux, but on most *N*X systems I've used, only the first
>*EIGHT* *CHARACTERS* really "count" (the rest is ignored).

>(Of course, more specifics about Linux' security would be appreciated!)

I was wondering, why not use MD5 to do the password encryption? It
supposedly guarantees uninvertability, and it is freely available; what's
more, you can easily persuade it to accept more than eight characters.
--
Sincerely, [email protected]
Stephen R. van den Berg (AKA BuGless). [email protected]

Real programmers don't just die, they produce core dumps.


[next article]
From: [email protected] (Todd J Derr)

Subject: OS/2? Windows? huh?
Message-ID: <[email protected]>
Date: 8 May 92 09:17:08 GMT
Sender: [email protected]
Organization: University of Pittsburgh
Lines: 20
Originator: [email protected]

well... I was told about linux a few weeks ago, and finally tonight
got up the gumption to give it a go... after spending forever tweaking
my system the way it was, well, maybe it was just time for a change...

anyways, I'm impressed, to say the least. Linus, not bad for ~1 year
worth of work ๐Ÿ™‚ Anyways, I'm looking forward to ftping some more of
the binaries tomorrow and getting things up and running smoothly... but
thus far, no problems at all...

looking forward to this (esp. with better hardware on the way soon!)

later,

todd.


--
todd j. derr (412) 661-8617 | [email protected]
university of pittsburgh 648-1247 | [email protected] (BITnet)
computing and information services | comp.os.mswindows.announce moderator


[next article]
From: [email protected] (Gunter Windau)

Subject: Re: mount lock files
Message-ID:
Date: 8 May 92 12:03:35 GMT
References: <[email protected]>
Sender: [email protected] (NUnet News Owner)
Organization: University of Nijmegen
Lines: 28

[email protected] (Duco Fijma) writes:

>Hello,

>A little question about Linux mount. When logged in as root, mounting a floppy
>is no+ problem. However, when logged in as a user, mount complains 'unable to
>create lock file'. Something is wrong with permissions. But where in the file-

Nope. The mount(8) and umount(8) obviously use the mount(2) and umount(2)
system calls. The latter may only be called by the super-user.
I think the main reason for this is system security. It would be too
nice to let users make setuid programs on their systems at home and
have them mount their floppy on the boss's machine ๐Ÿ™‚

>system does mount tries to create this lock-file? I could not find anything
>that looked like a lock-file after mounting a floppy as root.

>Duco Fijma
>University of Twente
>Enschede
>the Netherlands


=========================================================================
| G\"unter Windau [email protected] |
| CoGro, Dept. Medical Physics & Biophysics |
| Univ. of Nijmegen, The Netherlands |
=========================================================================


[next article]
From: [email protected] (david matiskella)

Subject: Re: Diamond Speedstar
Message-ID: <[email protected]>
Date: 8 May 92 00:12:59 GMT
References: <[email protected]> <[email protected]>
Sender: [email protected] (USENET News System)
Reply-To: [email protected] (david matiskella)
Organization: Univ. of Notre Dame
Lines: 2


I was wondering where you can get the documentation on how to program th
e S3 chip set? Does S3 put out a generic book or do you have to get one from you
r manufacturer?


[next article]
From: [email protected] (Orest Zborowski COMP)

Subject: Re: Wow! X is great!
Message-ID: <[email protected]>
Date: 8 May 92 12:50:46 GMT
References: <[email protected]>
Distribution: comp
Organization: Printer Products Eastman Kodak
Lines: 18

>>is libobz.a going to be part of the 0.96 release? it's missing from
>>the pre 96 + X patches release, making building any X apps impossible.
>
>Hlu has already added them to the library: the newest version should
>have them and some other syscalls ([f]truncate etc). I also think it now
>supports multiple shared libraries: practical for X and math. I haven't
>had time to try it out myself yet.
>
> Linus

libobz.a is in the pre0.96-obz.cdif.Z package, but in source form only.
you need to compile it (depending on if you have hlu's latest compiler)
and make a symlink from /usr/lib/libobz.a to it.

zorst
--
zorst (orest zborowski)
[reply to [email protected]]


[next article]
From: [email protected] (Tommy Frandsen)

Subject: Re: GNUPLOT 3.0 ported to Linux
Keywords: gnuplot, linux
Message-ID: <[email protected]>
Date: 8 May 92 12:28:14 GMT
References: <[email protected]>
Sender: [email protected]
Organization: Department of Computer Science, U of Copenhagen
Lines: 30

[email protected] (Michael Pereckas) writes:

>[email protected] (Tommy Frandsen) writes:

>>Hi!

>>I have just uploaded a Linux port of GNUPLOT 3.0 to banjo.concert.net in
>>the file /pub/Linux/Incoming/gnuplot.tar.Z. This is a binary only version,
>>since my patches to the orignal sources still needs a little tidying up.
>>It requires the following to run in graphics mode:
>> - The pre-0.96 kernel or newer
>> - A VGA-adapter and an analog monochrome or color monitor
>>If you use an older kernel or don't have a VGA-adapter you should still be
>>able to run gnuplot and get printer or latex output.

>I tried it and as soon as it tries to enter graphics mode I get:

>d9fl fxxx not implemented
>Illegal Instruction

>and then the shel prompt. Is this just me, or is something wrong?
>The recently released vgalib and it's demo works fine.

>Michael Pereckas

Unfortunately, I have linked the executable with the numeric coprocessor
library (Libhard.a), so the uploaded executable will not work without a
coprocessor! I'll fix that and upload a new version next week.

[email protected] (Tommy Frandsen)


[next article]
From: [email protected] (Joel M. Hoffman)

Subject: Re: X and 9600 bps
Message-ID: <[email protected]>
Date: 8 May 92 13:14:27 GMT
References: <[email protected]>
Sender: [email protected] (USENET News system)
Organization: University of Maryland at College Park
Lines: 9
Nntp-Posting-Host: rac2.wam.umd.edu

In article <[email protected]> [email protected] (M. Sag
gaf) writes:
> I have a problem using my modem at 9600 bps under X. I'd lose so many
>characters that communicating at that speed is unusable. No problem
>with 2400 bps. Is anybody having any success using his/her modem

ARe you sure the problem is with X? I can't go above 4800 even without
X (direct RS-232 connections).

-Joel


[next article]
From: [email protected] (Henning Spruth)

Subject: Re: X stuff on european ftp site ?
Message-ID:
Date: 8 May 92 14:13:13 GMT
References: <[email protected]>
Sender: [email protected]
Lines: 9

X11 is available from ftp.regent.e-technik.tu-muenchen.de
under pub/linux/X11.

Henning
--
------------------------------------------------------------------------
Henning Spruth [email protected]
Institute of Electronic Design Automation
Technical University of Munich, Germany


[next article]
From: [email protected] (Kristian Ejvind)

Subject: Re: X windows video card support?
Message-ID:
Date: 6 May 92 13:48:06 GMT
References: <[email protected]>
<[email protected]>
<[email protected]>
Sender: [email protected] (Usenet)
Organization: The Royal Institute of Technology, Kista
Lines: 14
In-Reply-To: [email protected]'s message of Wed, 6 May 1992 09:21:
16 GMT
Nntp-Posting-Host: gaia.electrum.kth.se

In article <[email protected]> [email protected]
-muenchen.de (The Master of Symbolic Links) writes:
>Besides this the 8900 has only half of the speed as an Et4000 and therefor
>it makes no sense to support this thing ...

I don't think that this should be taken into consideration as there will
*always* be people with this kind of equipment, (me, for example) who
would like to be able to run X, despite the lack of speed, (despite
the lack of *any* speed ๐Ÿ™‚ on their machine. Of course, there might
be other points on running X on a 8900c, like the difficulcy to program
the card and that's a *much* larger problem (compared to the speed, I meen).
I don't know anything about SVGA-programming but I do encourage anyone who
would like to implement support of the 8900c to do so!!

/Kristian


[next article]
From: [email protected] (Risto Kankkunen)

Subject: Re: Cross compilers-a question
Message-ID: <[email protected]>
Date: 8 May 92 14:00:10 GMT
References: <[email protected]> <[email protected]>
Organization: University of Helsinki, Department of Computer Science
Lines: 22

Jim Burns writes:
>> Anyway, with gcc 2.1 and binutils 1.94, it doesn't matter, wheter the
>> host is big or little endian.
>
>How is that accomplished?

Well, they take care of all those tedious details I mentioned: they
don't write directly their internal strcutures into files, instead they
write them field by field, taking care of the byte order. Gas doesn't
use the host's native floating point format, but uses a general floating
point representation. Just prior to writing out a float, gas converts it
to the target format with appropriate precision. etc. etc.

Traditionally this has not been done with compilers and linkers because
of the additional work and the performance penalty. E.g. GNU binutils
1.9 doesn't seem to support targets with different byte order than the
host. 1.94 does: it lets you define multiple targets and specify
different byte orders for each. It does also many other things, so the
binaries are huge (for example strip is over 800k).

--
no sig today


[next article]
From: [email protected] (Scott Dunn)

Subject: Re: GCC, 0.95c+, pre-0.96 and weird inconsistencies
Message-ID: <[email protected]>
Date: 7 May 92 11:22:55 GMT
References: <[email protected]> <[email protected]
> <[email protected]>
Sender: [email protected] (News Administrator)
Organization: Dept of Computer Science, Heriot-Watt University, Scotland
Lines: 46

In article <[email protected]>, [email protected] (Jacob Martin Boh
n Lorensen) writes:
|> [email protected] (Rob Hooft) writes:
|>
|> >May be related problem: under gnu-chess 3.1 I have to press 4
|> >times before the program reacts to a single line command. Does anybody
|> >have a good solution for this?
|>
|> I also had another problem, which I think is located in the stdio library.
|> I compiled gnugo with no trouble at all.
|> Gnugo seemed to work, except after a couple of draws it got stuck
|> in a _very_ memory-consuming loop, eating my 8 Mb swap space in 2 minutes.
|> I fixed it by putting a fflush(stdout) in the main loop, which is why
|> I believe it to be an error in the library.
|>
|> (Running 0.95c+, GCC2.1)
|>

Thanks to everyone who replied. Everyone suggested using
fflush(stdout). I had tried that - no joy. What everyone
seemed to miss was the fact that the same prog. works properly
under 0.95c+ but not under pre0.96. I don't mean recompiling
under the relevant kernel. I mean the executable works
differently under the different kernels.

The question about gets being unsafe - because there is no
way to specify the length of the buffer - Should use fgets
which requires you to specify how many chars to read. (Thanks Mike).

I also had the above mentioned problem. Needing to press
a key 4 times before a prog would respond - using curses that is.
No idea why or how to solve it. I assume it's a bug in curses.

Also, curses doesn't have wattron/wattroff, box etc. Does someone
have a version that does?

Also what's the latest version of libc. I am trying to continue
my project work at home. But source which compiles and runs
perfectly on a sun with gcc, compiles and ......
crashes. This could be as a result of crap code, but I really
don't think (hope) so.

Any comments, known bugs in the libraries etc welcome.

Thanks in advance,

Scott.


[next article]
From: [email protected] (Scott Dunn)

Subject: X and supported VGA cards.
Message-ID: <[email protected]>
Date: 7 May 92 11:39:45 GMT
Sender: [email protected] (News Administrator)
Organization: Dept of Computer Science, Heriot-Watt University, Scotland
Lines: 27

Could someone please post a list of suported cards.
The only chipset that seems to be mentioned is the
et4000. I don't know of all the different chipsets
and I don't have a clue what's on my card - an
Orchid Pro Designer II. I`ll have a look tonight.
Assuming the Pro Designer is not supported is there
anyone working on a driver for it? I believe it's
quite a good card. At least, in it's price bracket.
Whether or not my card is supported, it would clear
up a lot of questions if a definitive list was posted.
5 megs is a lot to ftp ๐Ÿ™‚

Further, what is wrong with 800x600x16. That's the
resolution I used (past tense) MS-Windows in. The
extra colours add a considerable overhead, and unless
your wanting to display nice gifs in the background
I don't really see the need for them. (OK I've now
got my flak jacket on - hit me ๐Ÿ™‚ )

How difficult is it to write a driver for X. If
no-one is planning to write one for the Pro Designer
I **might** look into it this summer. Athough I
doubt if I'll get anywhere.

Cheers,

Scott.


[next article]
From: [email protected] (Brian Olsen)

Subject: Modem
Message-ID: <[email protected]>
Date: 8 May 92 16:21:47 GMT
Sender: [email protected]
Organization: University of Missouri - Rolla
Lines: 8
Originator: [email protected]


Hello!
Could anyone direct me to a guide or give me some info on how
to set up a modem to allow external access?

Thanks,
Brian
[email protected]


[next article]
From: [email protected] (Peter Thomson)

Subject: Trident (SVGA cards)
Message-ID: <[email protected]>
Date: 8 May 92 16:42:50 GMT
Sender: [email protected]
Reply-To: [email protected] (Peter Thomson)
Organization: Department of Computer Science, University of Edinburgh
Lines: 18


Hi Guys, What's happening in relation to the subject heading.

Is there anyone working on the idea of trying to get a driver
working for this card, or are we all doomed $180 for the ET4000

The bandwidth has certainly drop.

If it's physically possible to get this working would it not be worth having
a go. What would we need in the way of sources to try and fix the slight
problem of the blank screen after startx. ๐Ÿ™‚

I have very little experience with this kind of software but I'm a willing
to get involved in trying to make this work. Even if the system is very slow
running under this card at least I would get the chance to see for myself
what the problem is.

Peter.


[next article]
From: [email protected] (Alan B Clegg)

Subject: Re: Getty on the serial port?
Message-ID: <[email protected]>
Date: 8 May 92 17:15:56 GMT
References:
Sender: [email protected]
Organization: Concert Network -- Internet Operations Group
Lines: 16

In article [email protected] (Joakim Rosqvist) writes:
>Is there any way to make getty look at the serial port, so that
>I could log in via modem? I've tried "/bin/getty 9600 /dev/ttys0",
>but it just makes the modem hang up directly after sending the
>CONNECT-string.

Well, on a similar tac, but a bit farther along, I have been able to get
a getty running on my serial port, but would like to be able to go
bi-directional with the getty only grabbing the port on Carrier-Detect
when the port is not already in use. Has anyone ported or worked on
a UUGETTY type program yet?

-abc
--
[email protected] Alan Clegg - Network Programmer
KD4JML (just my luck!) MCNC -- Center for Communications


[next article]
From: [email protected] (Brian Olsen)

Subject: Where's the source for Freeze?
Message-ID: <[email protected]>
Date: 8 May 92 17:56:13 GMT
Sender: [email protected]
Organization: University of Missouri - Rolla
Lines: 7
Originator: [email protected]

Hello!
I've looked around a few places and haven't found the Freeze anywhere.
Could someone please direct me to the right site?

Thanks,
Brian
[email protected]


[next article]
From: [email protected] (Paul H. Hargrove)

Subject: Where to get gcc config file.
Message-ID: <[email protected]>
Date: 8 May 92 19:05:18 GMT
Sender: [email protected]
Organization: Cornell University, Ithaca New York, USA
Lines: 10
Nntp-Posting-Host: theory.tc.cornell.edu

I want to build a gcc-2.1 cross compiler for linux, but I don't know which of
the gcc config files to use. There are several for the i386, but no
i386linux.h. Does such a file exist? Should I be using one of the others,
like i386bsd.h? If so, which one?

Thanks much.
--
Paul H. Hargrove
[email protected]
"A witty saying proves nothing." --Voltaire


[next article]
From: [email protected] (Chris P. Sullivan)

Subject: remote logins
Keywords: remote,serial,login
Message-ID: <[email protected]>
Date: 8 May 92 19:02:37 GMT
Organization: University of Massachusetts Dartmouth
Lines: 11

Once I figure out how to get getty to work (I should probably check the FAQ
again) and I am able to connect a terminal via the SERIAL PORT, not the modem.
Does the root need to be logged on at the host computer, or can I just boot
up Linux and let it sit at the USERNAME: prompt?

2. how can I assign a password to root? (shadow-bin utils??)
3. can one log in as root via the terminal on the serial port?

Thanks,
Chris Sullivan
UMass Dartmouth


[next article]
From: [email protected] (Al Clark)

Subject: Re: X and 9600 bps
Message-ID:
Date: 8 May 92 19:04:52 GMT
References: <[email protected]> <[email protected]
du>
Organization: Netcom - Online Communication Services (408 241-9760 guest)
Lines: 20

In article <[email protected]> [email protected] (Joel M. Hoffman)
writes:
>In article <[email protected]> [email protected] (M. Sa
ggaf) writes:
>> I have a problem using my modem at 9600 bps under X. I'd lose so many
>>characters that communicating at that speed is unusable. No problem
>>with 2400 bps. Is anybody having any success using his/her modem
>
>ARe you sure the problem is with X? I can't go above 4800 even without
>X (direct RS-232 connections).
>
>-Joel


I have been using kermit with a Robotics Sporster 9600, using 19200. The only
possibly unusual issue is that I've got a 16550 UART. I get the normal
9600plus display update, with no observed errors. File transfers rarely show
an error.

--
Al - [email protected] - My opinions are my own.
*** Practice random kindness and senseless acts of beauty! ***


[next article]
From: [email protected] (david.a.cuka)

Subject: Re: ESDI drive compatibility
Message-ID: <[email protected]>
Date: 8 May 92 19:38:29 GMT
References: <_ssv#[email protected]> <[email protected]>
Organization: AT&T
Lines: 21

In article <[email protected]> [email protected] writes:
>In article <_ssv#[email protected]>, [email protected] (Daniel Pereicich Russell) wri
tes:
>> Does anyone have Linux running on an ESDI drive/controller?
...
>In order to get my ESDI drive working with Linux I had to:
> a) Low-level format my ESDI drive with an extra sector in each track.
> Other OS's are good about remapping on their own, but Linux
> can't do that yet, so I did it in hardware.
> b) Remap my drive so that it had fewer than 1024 cylinders. My drive
> has 1300 cylinders, but since Linux uses the standard DOS
> partition table, I was restricted to 10-bits to represent
> cylinder numbers, which limits me to 1024 cylinders.

To get mine working, I just used pfdisk! I have
Microscience HD1240? (150M)
Adaptec Something or other

The setup menu on the adaptec controller has an option for remapping
to 63 sector translation.

DAC


[next article]
From: [email protected] (Michael Cain)

Subject: Marginally useful program using vgalib
Message-ID: <[email protected]>
Date: 8 May 92 15:01:12 GMT
Sender: [email protected] (Radio Free Boulder)
Organization: U S WEST Advanced Technologies
Lines: 233
Originator: [email protected]
Nntp-Posting-Host: copper.advtech.uswest.com

Attached is the source for a small program that renders
a portable bitmap on a 640x480 VGA screen using the recently
posted vgalib code. I use a few different programs that pro-
duce portable bitmap graphics as their result. This program
makes it possible to examine the results without dumping it
to a printer. The code's not all that good, most of it is
derived from various other hacks that I've done over the
years.

Since it's a single small source file, I didn't bother with
a makefile. Usage is either

$ showpbm file

or

$ produce pbm graphics | showpbm

The program displays the image, then waits for a keystroke,
which causes it to restore the text display. If the bitmap
is bigger than 640x480, you get the middle part of it. I have
found that I have to set the permissions on the executable to 4555
and have root own it in order for it to run properly. Another
of those situations where I have the wrong permissions on some
device or another....

Hope this is useful to some people,
Mike Cain
U S WEST Advanced Technologies
[email protected]

---------- showpbm.c -----------
#include
#include
#include
#include "vgalinux.h"


long bitmapWidth, bitmapHeight;
int xoffset, yoffset;
int windowWidth, windowHeight;
FILE *pbmopen();


main(argc,argv)
int argc;
char *argv[];
{
FILE *input;
extern void process();
int i;

if (argc == 1) process(stdin);
for (i=1; i input = pbmopen(argv[i]);
if (input == 0) {
fprintf(stderr,"showpbm: unable to open file %s\n",argv[
i]);
exit(1);
}
process(input);
}
vga_getch();
vga_setmode(TEXT);
}


FILE *
pbmopen(name)
char *name;
{
FILE *file;
char scratch[128];

strcpy(scratch,name);
file = fopen(scratch,"rb");
if (file) return file;
strcat(scratch,".pbm");
file = fopen(scratch,"rb");
return file;
}


int
pattern[] = { 0x80, 0x40, 0x20, 0x10, 0x08, 0x04, 0x02, 0x01, };

void
process(file)
FILE *file;
{
int c, magic, x, y, i, exitFlag;
long sum, height, width;
extern void setpixel(), initWindow();

initWindow();

c = fgetc(file);
if (c != 'P') {
vga_setmode(TEXT);
fprintf(stderr,"showpbm: bad magic number\n");
exit(1);
}
c = fgetc(file);
if (c != '1' && c != '4') {
vga_setmode(TEXT);
fprintf(stderr,"showpbm: bad magic number\n");
exit(1);
}
magic = c;
c = fgetc(file);
while (isspace(c)) c = fgetc(file);
if (!isdigit(c)) {
vga_setmode(TEXT);
fprintf(stderr,"showpbm: looking for decimal width\n");
exit(1);
}
sum = c - '0';
c = fgetc(file);
while (isdigit(c)) {
sum = 10 * sum + (c - '0');
c = fgetc(file);
}
bitmapWidth = sum;
while (isspace(c)) c = fgetc(file);
if (!isdigit(c)) {
vga_setmode(TEXT);
fprintf(stderr,"showpbm: looking for decimal height\n");
exit(1);
}
sum = c - '0';
c = fgetc(file);
while (isdigit(c)) {
sum = 10 * sum + (c - '0');
c = fgetc(file);
}
bitmapHeight = sum;
while (isspace(c)) c = fgetc(file);
xoffset = (windowWidth - bitmapWidth) / 2;
yoffset = (windowHeight - bitmapHeight) / 2;

if (magic == '1') {
sum = bitmapWidth * bitmapHeight;
x = 0;
y = 0;
while (1) {
int readFlag;
readFlag = 1;
switch (c) {

case '0':
setpixel(x,y);

case '1':
sum = sum - 1;
x = x + 1;
if (x == bitmapWidth) {
x = 0;
y = y + 1;
}
break;

case '#':
while (c && c != '\n') c = fgetc(file);
readFlag = 0;
break;

case EOF:
vga_setmode(TEXT);
fprintf(stderr,"showpbm: unexpected EOF\
n");
exit(1);

default:
break;
}
if (sum == 0) break;
if (readFlag) c = fgetc(file);
if (c == EOF) break;
}
}

if (magic == '4') {
x = 0;
y = 0;
while (1) {
for (i=0; i<8; i++) {
if (!(c & pattern[i])) {
int tempx, tempy, flag;
tempx = x + xoffset;
tempy = y + yoffset;
flag = 1;
if (tempx < 0 || tempx >= windowWidth) f
lag = 0;
if (tempy < 0 || tempy >= windowHeight)
flag = 0;
if (flag) {
vga_drawpixel(tempx, tempy);
}
}
x = x + 1;
if (x == bitmapWidth) break;
}
if (x == bitmapWidth) {
x = 0;
y = y + 1;
if (y == bitmapHeight) break;
}
c = fgetc(file);
if (c == EOF) {
vga_setmode(TEXT);
fprintf(stderr,"showpbm: unexpected EOF\n");
exit(1);
}
}
}
}


void
initWindow()
{
vga_setmode(G640x480x16);
vga_setcolor(15);
windowWidth = 640;
windowHeight = 480;
}


void
setpixel(x,y)
int x;
int y;
{
x += xoffset;
y += yoffset;
vga_drawpixel(x, y);
}


[next article]
From: [email protected] (Brian Chojnowski)

Subject: dvips/dvilj2p .. need some help
Message-ID: <[email protected]>
Date: 8 May 92 21:55:54 GMT
Sender: [email protected] (Net Noise owner)
Organization: University of Illinois at Urbana
Lines: 12

Ok, I have finally gotten MF to work correctly (I think) and now I have
2 problems. With dvips, the fonts get created properly, but when makeing the
.ps file, when dvips spits out a [1 I get a general protection
error. Any1 ever gotten this? Secondly, dvilj2 cant ever seem to write to
xxxx.mft files, even though the directory path is open and the environment
is the same as for dvips.

Also as a note, has anyone managed to 'tex' the a386 (gas docs) without
getting a stream of errors?

-Brian Chojnowski
[email protected]


[next article]
From: [email protected]

Subject: Solution: mount lock files
Message-ID: <[email protected]>
Date: 8 May 92 21:54:16 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 34

Ok, knowing that the mount lockfile is /etc/mtab~ then when mount fail with
unable to open lock-file then the file exists already, or you don't have
write access to it. That means that you need write access /etc to run mount.
Ok, correct me if I'm wrong, but mount-ing a floppy(or whatever) is to be
considered a MMFS(Major Modification to the File System), and only an
authorized person(root) should be able to do this.

I ran a little experiment: I made /etc world writable(only for a few minutes)
and logged in as jim(me) and mounted a floppy successfully, then umountes it.
I then returned /etc to drwxr-xr-x, tried it again, and the lock-file problem
again.
WARNING: DO NOT SET /etc WORLD WRITABLE!!!!, someone could put their own
passwd file(or worse) in there, then you are SOL.

Ok, nuff said, now to my problem:
I want elm, does anyone know an ftp site for it? I am using ftpmail, so
I can't just ftp to a bunch of sites and look around. I would REALLY!!!!
appreciate a reply to this one. Please e-mail. Thanks. ๐Ÿ˜‰

OK, a minor bug in linux, I just hit the limit on a partition last night.
I had a program creating a 300k file, it got halfway done, then just sat
there (I suppose it was twiddling it's thumbs) ๐Ÿ˜‰ I think there should
be some way of avoiding this kind of situation. This was the root partit.
I killed the program, and finished it on another partition. I looked around
and deleted roughly 20-30MB of duplications. Oh, Well, back down to 50MB
total. BTW, a functional root partition with most utilities, gcc21, smail,
a few extra utilities, and kernel, ps, smail sources weighs in at just around
30MB WTG Linus!!!!.

Well, guess I talk too much, but with attmail, more you say, less it costs
per character. ๐Ÿ™‚

Jim Gifford
[email protected]


[next article]
From: [email protected]

Subject: X-Windows Story
Message-ID: <[email protected]>
Date: 8 May 92 22:08:07 GMT
Sender: [email protected]
Reply-To: [email protected]
Organization: Cornell University School of Chemical Engineering
Lines: 39
Nntp-Posting-Host: hermes.cheme.cornell.edu

First of all, my kudos to Linus and everybody who has done such a great job
with LINUX over the past ~year. I just want to share a funny story with
you all...

My setup - 386SX-16, ET4000-0.5Mb, 2Mb memory, 3Mb swap. Not exactly the
right platform to be running X under. But, since I have always wanted to
see X on my computer, I grabbed the binaries, deleted gcc2.1 so that X
would fit on my 20Mb partition, and untarred the X386 port. When I first
ran it I got a "unable to open /dev/tty0" or something similar so I
makenod /dev/tty0 c ? ? and ran startx again. This time the screen went
black and my keyboard locked up. No sync, just a deep breath and a reset.
Minuted later I ran fsck and fixed up my fs - everything seemed to be allright
so I changed the video config to 640x480 and ran startx again. This time,
after a few seconds, the screen turned to the gray crosshatch pattern and I
actually had an X cursor on the screen. I was extremely psyched. About
30 secs. later, twm was asking me where to put the xterm and xclock.
A word about performance here - yuch! Swap in, swap out, swap in, swap out.
It took quite a while to move the stupid mouse cursor around, although the
hardware panning was great. I gave the xterm input focus and tried to
type ls -l to see what would happen. Uh oh! My keyboard was frozen! I
had used the wrong major/minor numbers for /dev/tty0. Now what? With all
the swapping going on, I was sure to nail my fs if I were to reset...
I suddenly noticed that the mouse worked fine, and that even better, I could
cut and paste. Oh, I forgot to mention, twm died on me for some reason,
probably lack of memory, it's only happened once... So I was stuck with
this xterm with "$" and no window manager to quit with...
So I pasted the $ and got an error message "No such file or command." using
this I was able to cut and paste "ls" and "cd" and make my way to the /usr/
bin/X11 where I was able to run twm and logout... It took me quite a while,
and I actually needed to cut and paste a lot before I had enough letters for
ls, I think the initial error message was "Cannot execute $" or something..
The moral of the story is, "the mouse is mightier than the keyboard."

Well, I'm off, tonight a friend's going to bring 6Mb over and I'll see if the
SX gives satisfactory performance or if I'll have to get a 486/33...

Later,
Elan Feingold
Cornell University


[next article]
From: [email protected] (Paul Douglas Page)

Subject: clock.exe libX11.a
Message-ID: <[email protected]>
Date: 8 May 92 22:41:06 GMT
Organization: University of Melbourne
Lines: 16


Hi,
I have managed to get X running but only just.

The card I am using is a Tseng MegaEva/2.
Unfortunatley my Samtron monitor is only capable of locking at 35.5khz horizonta
l frequency, and I have no idea what clock the Tseng card is capable of producin
g.
Does anyone know where I can get hold of the clock.exe program mentioned in
X-window Video timing tutorial?

Also where do I get the include files and libraries for X as they didn't seem
to be in the version I got fom banjo.concert.net ?

Thanks in advance

Ernie Elu
[email protected]


[next article]
From: [email protected] (Orest Zborowski COMP)

Subject: Re: X and supported VGA cards.
Message-ID: <[email protected]>
Date: 8 May 92 22:41:03 GMT
References: <[email protected]>
Organization: Printer Products Eastman Kodak
Lines: 93

[email protected] (Scott Dunn) writes:
>Could someone please post a list of suported cards.
>The only chipset that seems to be mentioned is the
>et4000. I don't know of all the different chipsets
>and I don't have a clue what's on my card - an
>Orchid Pro Designer II. I`ll have a look tonight.
>Assuming the Pro Designer is not supported is there
>anyone working on a driver for it? I believe it's
>quite a good card. At least, in it's price bracket.
>Whether or not my card is supported, it would clear
>up a lot of questions if a definitive list was posted.
>5 megs is a lot to ftp ๐Ÿ™‚

the stock x11r5 release of x386 (on which the linux port was
based) only supports et3000, et4000, pvga1 and gvga. i've
heard of pd extensions to the trident chip and maybe others,
but haven't gotten around to trying them.

>
>Further, what is wrong with 800x600x16. That's the
>resolution I used (past tense) MS-Windows in. The
>extra colours add a considerable overhead, and unless
>your wanting to display nice gifs in the background
>I don't really see the need for them. (OK I've now
>got my flak jacket on - hit me ๐Ÿ™‚ )

the x386 server uses packed 8-bit mode by default. changing
this would probably be quite a job, and i've heard that there
is no guarantee that all x applications would honor the
restricted mode (which i consider a limitation of the app).

>
>How difficult is it to write a driver for X. If
>no-one is planning to write one for the Pro Designer
>I **might** look into it this summer. Athough I
>doubt if I'll get anywhere.

depends on how well you know your chip and how closely it
maps onto the model assumed by the default distribution.
the drivers for the chips themselves are not very large,
but they are all very similar. i don't have access to these
different chips (nor have any intention to support them),
so you're kindof on your own.

>
>Cheers,
>
>Scott.


some random ramblings...

my original intent was to provide x386 to a limited readership.
i guess it was my own fault for not restricting this more
forcefully (could it have been my intention :-)... i wanted to
make sure that it worked more often than not, that most of the
fatal bugs were fixed, and that people who wanted to obtain a
binary version of the server and clients could do so (compiling
x386 is not a one-night affair, unless you count all-night!).

since more and more people are discovering this "hidden" location
of the x386 port, i've placed my diffs on banjo.concert.net for
others to work with. these diffs are against a partial version of
the x11r5 release of x386, since i didn't have that much space and
wanted to get down to a bare minimum. with the tremendous advances
made with gcc2.1 and linux0.96 the problems associated with creating
a full-fledged distribution should be minimized.

i've been using x386 almost constantly since i finished the port and
it seems to run very well (for my needs). a friend has successfully
ported xv, xmail, spider and other clients, which has boosted my
confidence in the port.

there are problems with it, the greatest one being the screen blanking
after the server exits. i've not gotten around to trying some solutions
to the x386 server itself (which only affect the et4000 version) nor
applying the brute-force solution of writing a program (possibly using
the vgalib posted) which resets text mode. has anyone run x386 without
an et4000? and had the screen not blank out?

i'm probably going to wait until linux0.96 before working some more
on the server (it's been a long time of hacking nonstop and am getting
more pressure at work :-), since linus is running x386 and has solved
some of the kernel related problems with pty's, etc. hopefully at this
time the x shared libs will be working (i'm running a few clients
with shared libs and their sizes are stunning!), so we can begin
with a "real" distribution of x386.

zorst
[[email protected]]
--
zorst (orest zborowski)
[reply to [email protected]]


[next article]
From: [email protected] (Linus Benedict Torvalds)

Subject: Test-image of 0.96 available at banjo
Summary: just the binary
Message-ID: <[email protected]>
Date: 8 May 92 22:05:44 GMT
Organization: University of Helsinki
Lines: 37

As 0.96 has some changes in harddisk IO handling (interrupts, timings
etc), I'd like for people that have ever had problems with the harddisk
driver under linux to try out the last test-image of the 0.96 kernel
before the official release - I'd rather not have the same types of
problems that we had with 0.95.

The image (no sources - wait till next week) is available at
banjo.concert.net: pub/Linux/Incoming/testimage.Z, and is just my
current bootimage that I'd like some feedback on.

The changes to the harddisk driver (which is the main reason I want to
make sure this version works) are just:

- interrupts enabled most of the time
- inb_p / outb_p changes

The first one is to lessen interrupt latency, the second one hopefully
helps people who had problems with the driver at high speeds. Both
changes work well for me, but then my machine seems to accept almost
anything... I hope 0.96 will work without any "a" releases.

I'd also be interested to hear if this image removes the problems with
serial lines at high speeds under X, as well as /any/ other problems. I
can still make minor bug-fixes if something turns up, but I'd want
reports by early next week or so (preferably with "testimage" in the
subject line).

The image is compiled with the us keyboard maps, and with the
floppy-device as root. It should be binary compatible with all old
versions, as well as the interim X version, so you just plug in the new
kernel and you're (hopefully) off.

Ignore this post if you don't have time to check out the image this
weekend: the image is only meant as a quick test-release to get some
fast feedback.

Linus


[next article]
From: [email protected] (Erik Fichtner)

Subject: Re: X and 9600 bps
Message-ID:
Date: 8 May 92 22:59:02 GMT
References: <[email protected]> <[email protected]
du>
Organization: University of New Mexico, Albuquerque
Lines: 23

In article <[email protected]> [email protected] (Joel M. Hoffman)
writes:
>In article <[email protected]> [email protected] (M. Sa
ggaf) writes:
>> I have a problem using my modem at 9600 bps under X. I'd lose so many
>>characters that communicating at that speed is unusable. No problem
>>with 2400 bps. Is anybody having any success using his/her modem
>
>ARe you sure the problem is with X? I can't go above 4800 even without
>X (direct RS-232 connections).
>

It's possibly a problem with both. (Nice answer,eh?) I've succesfully
downloaded stuff here at 19200 with kermit and rz. sometimes it gets really
angry and drops packets, but it eventually gets here. no problems using
it to read text from our unix systems here. but i can't make Taylor
UUCP talk at over 4800. even to a local machine.

haven't tried X yet. that comes later today.


--
Techs [email protected]
Disclaimer: any inaccuracies are caused by the phase of the moon
"...whaddya mean I ain't kind... Just not *your* kind!" - Megadeth


[next article]
From: [email protected] (Doug Mayfield)

Subject: Optiquest Monitors????
Message-ID: <[email protected]>
Date: 8 May 92 19:54:09 GMT
Sender: [email protected]
Reply-To: [email protected] (Doug Mayfield)
Organization: UC Davis, EECS Division of Computer Science
Lines: 10

Has anyone out there figured out the ModeDB
settings for and Optiquest 1000 monitor ???

Other info: to be used with Diamond SpeedStar Plus /1MB

------------------------------------------------------------
Douglas B. Mayfield
[email protected]
(Computer Science Dept.)
(University of California at Davis)


[next article]
From: [email protected] (James Henrickson)

Subject: Re: communications only work up to 4800 baud. Why?
Message-ID: <[email protected]>
Date: 7 May 92 02:41:00 GMT
References: <[email protected]>
Organization: State University of New York -- Institute of Technology
Lines: 58


>I'm using kermit on an old 286 to connect to Linux via a serial line.
>The Linux machine is a 386 with 6M and nothing else running (except
>several getty's). But as soon as I go above 4800 baud (9600, e.g.), I
>start getting transmission errors? Has anyone else had this problem,
>or, better yet, solved it?

I have an old 286 connected to Linux via a serial line, but I am running
it at 19200 baud. The thing totally locked up on me the other day, but
usually my files transfers are just slower than they should be. By locked
up, I mean that line. Everything else worked fine.

I suspect that my 286 can't keep up with the high speed. I remember seeing
that someone else was experiencing the same trouble until switching to a
UART with a FIFO buffer. I'll be trying the same thing in the near future.

>The problem does NOT lie in the comm ports or the cable. I use the
>same setup with Laplink which runs MUCH MUCH faster (112,000 ish?).

Are you talking actual speed or configured speed? At 19200, actual
transfer speeds fluctuate so much it isn't funny.

>Suggestions? Please!

After I get my new UARTs, I will run a few tests on transfer speeds and
will post the results here. I expect the biggest difference to occur when
I replace the one on the 286.

-------------------------------------------------------------------------------

Other things:

I have a dial-up line connected to ttys1. I've been able to call up, but
my processes aren't terminated if carrier is lost. Is there something I should
add to my modem's initialization string, or is the serial port driver lacking
in this respect? The way it is now, someone can call up and land in
the previous caller's accout if the previous person didn't log out before
hanging up.

I'm currently running pre-0.96 and thinking of adding two more serial ports
(for a total of four). I would like these additional ports to use IRQ2 and
IRQ5. Has anyone managed to get Linux to recognize serial ports that use
nonstandard interrupts?

I, too, will be around this summer and taking only one class. I'll be
graduating in December and will probably lose Internet access then so the
version of Linux around then is what I'll be stuck with for quite some time.
Will much development effort take place over the summer, or should I expect
things to be relatively idle? I'm willing to attempt a port of one kind or
another, if someone can point out something they would like to see. I have
no interest in duplicating someone else's efforts, though, so please keep
the ideas original. ๐Ÿ™‚

--
James L. Henrickson, Software Engineer * "machine learning for
Critical Technologies Inc. * autonomous vehicles, digital
311 Turner St Suite 303 * signal processing & parallel
Utica, NY 13501 [email protected] * computing"


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: Password encryption bogus?
Message-ID: <[email protected]>
Date: 9 May 92 00:35:31 GMT
References: <[email protected]> <1992M
[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 18
Nntp-Posting-Host: romeo.cs.colorado.edu

In article <[email protected]> [email protected]
-muenchen.de (Stephen R. van den Berg) writes:
>In article <[email protected]> [email protected] writes:
>>[email protected] (Karl R. Buck) writes:
>
>I was wondering, why not use MD5 to do the password encryption? It

If it's not the DES based crypt algorithm, it is non standard.

This breaks many things. Forexample, programs like xlock that
use getpwent(3) and crypt(3) will fail. I can't use a global
password file. Bad.


>supposedly guarantees uninvertability, and it is freely available; what's
>more, you can easily persuade it to accept more than eight characters.

UFC is also freely available, export legal, and standard. We don't
want more than 8 characters for compatability reasons.


[next article]
From: [email protected] (Darren Senn)

Subject: X, Mice, shared libraries, etc.
Message-ID: <[email protected]>
Date: 9 May 92 01:16:52 GMT
Reply-To: [email protected]
Organization: Curiosity Confederacy
Lines: 30

Well, I've got pre-0.96 running (THANK YOU LINUS for telling us about the
shared library permissions), and X11R5 is chugging along happily... well,
mostly.

I've got a Mouse Systems White Mouse (serial). This little gem has a switch
on the bottom to switch between Mouse Systems (3 button) mode and Microsoft
(2 button) mode. Needless to say, I prefer the 3 button mode. Unfortunately,
I can't seem to get Xconfig right for a mouse systems mouse. My last try
was:
MouseSystems "/dev/ttys1"
Baud 9600
I've tried it with and without the baud line, and on different serial ports,
but I can't get the thing to work. Has anyone had any success with the
Mouse Systems mice?

In the meantime, I'm just using X with the Microsoft mouse configuration.

Now about the shared libraries... I've got my old Lib92.04.06 from MCC 0.95c+,
and now I've got Lib92.04.12 that came with X. Do I need both? How does
a program choose which library to start up? Will it just use the most
recent? Are they backwards compatible?

Lastly, I've pulled xdvi off of the net, and I'm configuring that for Linux.
So the big question is: is our version of X a complete implementation of
X11R5, and are libX11.a, libXt.a, etc. complete?

--
Darren Senn Phone: (408) 479-1521
[email protected] Snail: 1785 Halterman #1
Wasurenaide -- doko e itte mo soko ni anata wa iru yo. Santa Cruz, Ca 95062


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: mount lock files
Message-ID: <[email protected]>
Date: 9 May 92 00:54:51 GMT
References: <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 21
Nntp-Posting-Host: romeo.cs.colorado.edu

In article [email protected] (Gunter Windau) writes:
>[email protected] (Duco Fijma) writes:
>
>>Hello,
>
>>A little question about Linux mount. When logged in as root, mounting a floppy
>>is no+ problem. However, when logged in as a user, mount complains 'unable to
>>create lock file'. Something is wrong with permissions. But where in the file-
>
>Nope. The mount(8) and umount(8) obviously use the mount(2) and umount(2)
>system calls. The latter may only be called by the super-user.
>I think the main reason for this is system security. It would be too
>nice to let users make setuid programs on their systems at home and
>have them mount their floppy on the boss's machine ๐Ÿ™‚
>
>>system does mount tries to create this lock-file? I could not find anything
>>that looked like a lock-file after mounting a floppy as root.
>

Make mount / umount setuid to root, executable by root and group
operator only. This is the "standard" thing to do.


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: Trident (SVGA cards)
Message-ID: <[email protected]>
Date: 9 May 92 00:59:34 GMT
References: <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 25
Nntp-Posting-Host: romeo.cs.colorado.edu

In article <[email protected]> [email protected] (Peter Thomson) writes:
>
>Hi Guys, What's happening in relation to the subject heading.
>
>Is there anyone working on the idea of trying to get a driver
>working for this card, or are we all doomed $180 for the ET4000
>
>The bandwidth has certainly drop.
>
>If it's physically possible to get this working would it not be worth having
>a go. What would we need in the way of sources to try and fix the slight
>problem of the blank screen after startx. ๐Ÿ™‚
>
>I have very little experience with this kind of software but I'm a willing
>to get involved in trying to make this work. Even if the system is very slow
>running under this card at least I would get the chance to see for myself
>what the problem is.
>
> Peter.

There is a free trident 8900 driver available. With minor
modifications to vgaBitBlt.c it will work on the Trident 8900 chips
which only support a single bank.

See poincare.geom.umn.edu:/scott/X386-Trident.tar.Z


[next article]
From: [email protected] (Michael K. Johnson)

Subject: Re: dvips/dvilj2p .. need some help
Message-ID: <[email protected]>
Date: 9 May 92 01:54:25 GMT
References: <[email protected]>
Sender: [email protected]
Organization: St. Olaf College; Northfield, MN USA
Lines: 22

In article <[email protected]> [email protected] (Brian Choj
nowski) writes:

Also as a note, has anyone managed to 'tex' the a386 (gas docs) without
getting a stream of errors?

-Brian Chojnowski
[email protected]

OK, the a386 thing is just a small piece of the full gas.texinfo file
-- just the 386-specific stuff. If you want to print out the a386
stuff, you will have to edit the file and add "\input texinfo" at the
very top, perferably on a seperate line. Also, on a seperate line at
the bottom, add the command "@bye". You may have to do more -- I just
printed out the whole gas.texinfo file, because there were things that
I wanted to know about gas in general. But that should take care of
the most heinous errors, and get you to a point that you can work
from.

Hope this helps people who have been having trouble with this....

michaelkjohnson
[email protected]


[next article]
From: [email protected] (H.J. Lu)

Subject: clock.exe for X386
Message-ID: <[email protected]>
Date: 9 May 92 01:58:50 GMT
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 196
Originator: [email protected]

Hi,

Here is the clock.exe for X386. You have to run it under DOS.

Have fun.

H.J.
---------
begin 644 clock.tar.Z
M'YV08]B\&;/&11D\90`H7,BPH<.'$"-*G$BQ(HB+-#*"`'"QHXR-'2]^[email protected]
M1`P:,63`N%@#[email protected];-6:XA!'[email protected],FC,X`BCHL^?0(,*?5AG#ITPMS5")3)T^G4JUJM6K6+-6;:(%"H`&`(@`F`!`!0$5I#@%``!$X;Z%'!;Z6WAA
M+8`P=M_8+62WD=U1=F/9/6;WF=UR=N?932!`883&`$Y`]@$9"60XD/U`[email protected]
M$V19D(%[email protected];9&V1ZD`4,4(AA-4?7-EP3<:W%M1C7A5P_M]0$""[email protected]!U!B^8[E299_67YF>9_EBY:'6IYJ>;/EWY:36^YN.8`""@[email protected][email protected]
M?8;U*];#6!][email protected]]976N]I_5HP`.`3`(`,59%(.8.PHX`B1A"@BS`#N!,#
M,([email protected]^%^NBBS0#LD(```L#4(<`Z]$Q8H3WK#/"/$?9HR*&'/"`@(HDN=OAA
MB!D,XH<^#_"1P!\1!.('/D`(,8<"@[email protected]"!6Z`/`.MWH:(\"=2!#C(6,"%,A
M/XE8X<\B1O"SB)>)L-.%,(T94`05,L$`@YI4T.`F#$(,P<030RRA`[email protected]#!'
M&FW`P489(,`AQQMGR!%&&R#0\08(9)1!1QERM-$4H7F\44=25AP1Q`[email protected]!!'
M'4C1H0<(;\PQ1AILL!&&HW+,P4%189Q1QIX!#;0&"%NLV28,?*PIIYM=2`"I
MI&6,,2D9(.1*T!Q["G"1)@H!>&`@[``@P#$&*"0!`+!``,4O"GEP!;4`:`%N
M`+H0$Z$NO['#``[email protected]+!+I0\VXU+_ZA"[_NP#("%[email protected]$X\\H&`"0B`4:`,!M
M7(68P:T'`)@AL0$;F9$$,08DG*4F31P`H!130`'+"[email protected]_((@-`,A2AQ+6(
M!Q0O8L%&67CQA28_M)SSSCFT/(H("GM`@L(6F```QQX+P_+()<.R`L%2K!PR
MU##+3#,`-N.L,\\^[email protected]*[email protected]+0*2+.P=,<`?'RURU&_0#`$5K?\M`#^##6#?8P,[email protected]%][2P#`X;0`$`R&E"#@0`]K`6'[email protected]`[email protected]"P#\67('+
M6E!H,@SD`/QR!P`.G-O%``!2GDX*`<`[email protected]"Z_V`N*+MC82T*^]@KP;X1,M^WT
MVR\+3'`A_3P-=]9YUWPSWU_W_+?80M-0M`U(X[!VT\H7CS(Z0R1/?,EXS^R\
MUSM+#SW00NM0-`](^["]\-U'/34Z`8AO-_E:G[^^W_\+W"C:[email protected]@/"`%[email protected]>
MV_9G,KFAXQ?\J!_SS,>UY_5-?8`3FA&*[email protected]*2%QBT,``-)!A`!D#@#^Z-SG
M`!"ZT0'H%WY`G>I8!P#7F2!VLZM=P&Z7NX#[email protected]'MJ#<+1B]L[".;]0J([email protected]/!G!^+$/"9:OB>@#8`;)YKX"PF]A\L/B$J5&
M,%UT<8E?]-\%H:@)`[email protected]`BJ1;8,%[.#"/LBM
M$"9.(:L)U[C*=2Z%[email protected]`N7Y1"!E.*'3:>`"`-.$%498ND^8BI2DQJM_`([email protected]&*$VIRA>B\EREM&4K:WG*5OZB'+&MH0FCL,"3"/`/`TB3E>9:A`^D&8US+.(>W1R`/_Y1BXV0DV+DC`[email protected]%%8;N
M#*3[!3Z"J8E0#K.8TF2F/^J`"VSBTI1EFI`F%`[email protected]*[email protected]$>[email protected]$`-<((H5`*`8
M=D``/UH!A58L`AL6189%H6%1=%#4HOCXJ".`\0$O8*$9`."$`:Y0!802PP\0
MX,M!#$*0)98$&`:Z&"$-`[email protected]%%800"CT((`3!"*QK0``!`8A1>V90`Q"*`;;7,`
[email protected]&((ZYH$(`Q#(`(`[email protected]!L`8!17`[email protected]<@&L8``K9D8A![`[email protected]"X`$`-
MFA&"`M)!`$ZU`V0%,-G*>@`/F#6`'B`[@LX6D`^AU0PN2&!:#P`BM(*`;!U:
M2XC0&D(`'G#`[email protected][email protected]\?#L"X(:`[email protected][email protected]`LI`,`#4(`N+KB!
[email protected]`!WB4,!UG;"`[email protected]:!<=/UC`,@``6&4TIKG/)6\CNHM*
M\A)B`[email protected]^UO*?B"_%@#M.I<+8-3BHKC^[email protected]_=86%W58;BC8((`%
M&,`-`J!#"[email protected]"@XVBA4(,`T*&V7=C"PPD+!8;1X3Y8Z``=3P#`A$?\`@#L
M8A<>7LLR7*"(`[email protected]`&'0P`#JT-X1US&,8L2`#[email protected]`@%@H`!V-60\2#
M`#`@&,6PH`AB#`,*7<8?`+``!:M*`QWH6`T62/``+/-`!XTEP`$4`0U:&("R
MPY!"'=Q`!T`1BE*'[email protected]'7K41%S1``0#8QR`XAH5`G.,#``C$,2P!`$#P(P=S
M&,`O_O&/*]"""I1%*""<8($`8&$+6=""[email protected],\`Q>PS,(6UG$`7/R#OMP3:2NE1M+"(:"[email protected]&'0ZP`@]`=1#[email protected](`CGCI60/@`/<[email protected],L&QIFL#9T*Y#
M`4SP#P]T:Q0B3H0#*%Q`#!/#`W5E!#`6$.T)`'84=1U%7D?Q6%S(0`#C-<`4
MEK%LJ%[[V0`81AT68(*]KH?;+NA6B`5`"R905DS<8(0V;$",/R#`!L6X`P%8
MH0I%($,&[[email protected]'*9BJ#.1`7.(8OX,"*/X'`][email protected]$4_8N`Z"H5E%!*/C'P_YR+6A
M#[email protected])X`)%%$`*I<<`">?>,49D'($*&+*Y'[email protected]&3\80"Y6,L`XK&"
M8,AA`D"[email protected]&JWH>_)F,'-#?`(H*QCK\"0Q6+0(;(22["I+/\XAD7NB(:`')6
M**T`M`5P0"0#[T",K#'"J(A!PRL
ML.ITF(`0C'"$*%CA"O,PQ1>\D([email protected]`T$)[email protected]'8#^@C+K_'.49OWC%-S[T$BP>
[email protected]!D!!GDJ1>1#[email protected]$4`5&$#11R!`*IPK_,CKO0_K#SC
M+E_$]>$N]YT;%0`M90>VM'5"SEWAI0I8J3O!9``DD+<<_GBI`U9ZW0[\XX0I
M]'Y^$']7\`]V8`"VQD+790S^<%TJ`G\K98`(<%V$\"7E4%0,@'X(M7X"H"/=
[email protected]=#T'Z=`X&>`[email protected]@@#V=UV+X`\`.((#N%(G.`!&8`#M=&OD%0<-B`ZND'\O
MZ#DV>%TN8(%%)4*!<`^55GQTH``[email protected]<'L':+X`V+<`W#,'-V4`"1<`C><`C7
MH`+!8`_8P`,H,`?H8`C!T%+O,`/JP`F+\`Z`X`R`$`C/4&F`D`\0(`<+H`C>
MI`TS(`V`D`X0L`[P,(<((`<-H`CCH`C?H`CZH`C6D'E]B`!(J`#1H`[1L`W0
MP`W^D`QU8`^[email protected]`<5(`#E,`'>,`'7,`=?$`W:4(F7Z`\[email protected]`7#8`[email protected](5:
MJ`+EL`'>L`&E.`&!4`^[email protected]_4<`_3<`]\2`Q+,@T`,`C2T&0%H`#V4`<4H`CR
MH`C7H`C[H`C<[email protected]`!V4`&96`?JL`X/8`_&``AU(`'00`[1,`W1(`W^@`RQ
M!UBZMH>P"`CF\`>!`$R`,`/5`(M(*`<[email protected]>7%0SHD`\`8`=M4`50<`56$`4(
MI0CLL`CSL`CZ0&7S*`V#T`X%P"`&L(0,8(B(J(B,F'[email protected]`[email protected]`#[email protected]`US8`"@
M*(JDB(JJB(F:R`[0$`U%L`/L4`?0P)'D:(X.``W9``W4L([M&)`"\(XR8`?)
M2`P(T`5+:`#RN`U[.`P!^0-<6`_H\`<'>0-5$`4)[email protected][email protected]%"#4`P%<)([<`UU
M8``[X`W*5PYR8`"VB(NZ6`_0``V92`?HJ([L*`US\`UQF8O7D`_KD`[#@`@;
M=))[email protected])-U0`[I2)328(=GF99KV9;U$`W6<&I:D`5=T)-U((G>()2.N0X%P(KY
M0`>5%P]/&953:8<$``AVJ)K2,`RM*0<1<`F*P`PRX`SV$`V`$`Z```C,\)MN
M>`[email protected]/7H`5Y4`!2T(Y:,`P][email protected]!QT0`'U,(7!4`<.$)`*$)EJR98"H`/E
M,`P7M0#:0`MX(QU0`"&(`^[email protected]`CJB0<(\(X#L`T3L)[XH`#[email protected]`<#8(]_,`R3$`N?H`++(`=.
M,(M\(WX,`<,`*%[email protected]`#QF8S+B`<0H`#_P)[N"9_RN8SX,`SY,`SR&)OHD`/](`<[
ML`SH\`)[email protected]/+$'<229'HT`][email protected][email protected]([email protected]`YPX`]R\`%::@'_H*4\$*>U``#H
M4`?_(`<)L`RUX&%MD*MT`#"2(R#,`Q+LHRPT`*JX`O%<`"QD``NL`.#I:)WL%`[P`%S<`"#8`P$\(WL
M(*)]Q(HNL`(M``#A4`\%``[email protected]`EL(00/``"@8`AK`0$GH!!IIA!DL!"(IA!VP5[`L(MD`[email protected],K,[email protected]+:1GI),`Y($*\D0`=$,'0=``CKH`.`0)[email protected]`+W(`P```]U
M$`-3$`[email protected](Y2``40VXX5*PU;,`##L`43,`U;(`/_H+'SB0<2<`3BT`Q.T)E+
M*@U#.I]/&J5'D*_Z`+"`L`](,:'K``("2[!4:8<">P$%:P!S``T%JP"W0``7
M*P?4X`_/L(D*,`PF&PWKH`OIER`4<`6#D`\+X`=8MP#)F`_UT`=85P__D`A.
MX`^(8`?\L%*-X`?L8`$'@`[1``!?<`!FJP^+8`?^L`#[email protected]`<,,`!VH`]&D`<(
M8`Q&D"WKX`+WX`V#X`\,(`<#X`M>"PAZP`XU8`Q"D"URH`%[email protected]`MD`-VX`"8B[CN<`R'&P.;8P3ZT*YBXB4'[email protected]/V$":`0#F="G=.H`]5EYRP``.J
M0&OHT`4`H`I)D`=:*PUTP`2PX`*J$`1!]KNJ`'OWH`1XT`/@4[SK<`^P``*U
MRP^W*P"YBPZ(HPI"D(3-JPKH8#;F6P?Y``M%H`I2,R:\NYX$T*GWH`UP)PRR
MX(SZ4`!N,E/)NJR+<('#H`B9:P1+.`"_ZP14&2(-D`[email protected]@[email protected]_SL`HR\';#
M0*%0P`*`0`M-8`_LH!X`T&68$'=8D+794`M0`#[email protected]&-)$`[email protected]$H`+(``LU)(F4N`TZ^HWVX`L9B)[email protected]`4`PP8`*(T`+V
M^``O4[%[J<7MF`4#$`Q9,`'9$,;1P,72<`0I^PR#QS?#[email protected]`"`52$7XKI;
M,@Y7,'0%8)C^P`[email protected]@`\,*\$P`/V*@`[email protected])_;0`<\H&OC\`#[email protected])UD`"%
MG!#WD,@I,`[email protected],<\4`1T$,AE4`MT!9]4``]``S_<'6"&0,OK`A?:B&+X`5X:[email protected]_M0`+HX%PU+`?9"60*``A^X`\[email protected]`]U,*%N6,"PR,@[email protected],MC`[email protected]`[*L`]@.I%XZ\O`/";\@`Z>8*>,D`]RG`C`W"7\4,[email protected];G`_:P`=)
M,(M(``E706M&0X#`)[email protected]$*:F`_S3)[email protected])[email protected](H`_I(`CZ$*6!T!".L!!`P`&;
MH!`@P`L^@`[email protected]`:V0`"_D`P;X`##P`K%L`YXT`+0T`S.L`'",`7>D`]X
MD`,_$`#KL`*HT`K%T`'XD`Q[X`R7``0!Z0([email protected]?,27F*.6R`U9`(IA/,83
M$`U9<"0PV=<$$`!R4*WW2``:"WEWEW($P`^`M0JX``O78-F```"K4-FKL`J-
MP,D_!0R6G0P,\%/(T-G`D`L.```7S`H\\`2:!0M/X''(1PH9K`JT)T+/5W$K
M)WY"EPA%0``O=P0&8`+"S73B)WC"[0"`A=Q%P-Q%H``]!P"Z$`O7L`X(H`NR
MD-VS,PO7T-M_('2+4`0"T`/8C9<][email protected]+6W<\EAF\"0A(`7HL`;I\MR)
M4`5C)=S2K0"[``%[email protected]=-S'8-Z8?73BK<.0ULWHP+H*#G2[APZM'M<.#B>PU'9P*!4';I[0`>7G;L;0#Q+0!G$'"[L#@[email protected]#L?0##@`[Q\`^[$!MM5+M?5FTP`.`T`<]00<)T`A5``.4)0=+D`@%@`NFT`"[$`H-P.3I;0*+T`&11024
M96L-<%'V(`SF"`%4CI8#@`XE``"X8`UHKN8:WMW"30&,T`06D`[email protected]`&)[email protected]$'
[email protected]$CX`")[@[email protected]"0%E6G(S:```Y/*6+H`R+``V;#@"='@*-4`F;
MG0N;_>:T\`.4-0=NIVL[@`UTH+.P#@`:0#E%)@^+,`TK``VYP&F;[email protected]"K;N2R
M#@1XUE[(`=F[9P+%4`4,X.9-0-X=,.MA#@'0K0#/?>`B!`"TAQP9/MT*L0[N
M(.WC3>W6CNV"U^QA3FS!0#FXD!#G/N[ESC5"'G3N?NV+T`3QONW.'@*-%>[1
MK1#[[email protected]&#L`\$8`<$0`L^0%G0CN%[email protected]/[email protected]`%>X&UWO`A=$.]5
MH``F0`Q]P`#_8`[email protected]!MP/EN6V*$`!/[email protected]>8`6.X`4\.+.,[email protected]`[email protected]]=0`$[
M3CD^_E(S2_(FC_(JSP!>\`%9(/-UL`RPH`%1JC,'\`'(7?(MW^+=70<;W_$?
M'_(XKO1NSM(IO_(M?^)I$'!4'`!#(/-T$`##@/,ZS_-(O_5`?\="3_0\?O0[
MXN8E?_9"U_1/GP7#@/5:__,MGWXO)[email protected]\`]U<`'@\P\\\.L-,`0\T`!U0`_H
M<`.63R5PCPYL(/0,G_'0.`@72X`/,^[email protected][UX/A^(&1"(/E/
[email protected]#H,`;_``M/[email protected]_0,JF+^'MSN1?3P&J+_8$R&VO'R(&(/MM7_NW[U)^P`#H
M``G[5`(0!M)[email protected]!Y&AOO:SX`[email protected]?'#[email protected]!E5D!R,`/)`"3`0+0P8X2!_D`"\2=<7``
M"B`Z8`;^``*.*7]P_Y(1.P`"N,\(H`,0L$^&0+M"!T%@C_TZ#<#'[L``"';H
M8`;L,[email protected]!QC`$&A<0(`.S`-$(`[email protected]!%8!^[email protected][email protected]#=+`%_,$XR`!M2P#`[email protected]
M"_*[email protected]`Q.G!I05Q<)""8($(A0P(02^&5,[email protected]%H!'J7`[email protected]
M`&`%.`$DD+7\`=?*!>,D"2Y!>7`/7LKMD`4:@6:Q`!_``FJ`[email protected])`9`H+JE'\`R)JB.\NIS&6!I^`$$X`$<@;$1
M`/Q`%40"D4-4M,\>@C(`8!+P&V;S"S;;T.$[[>`(,+($$*]`@`[email protected]#
[email protected]@(>@`-'`"XM=!8%`[email protected]@#;[email protected]([email protected](20,!)*0CDL"2#`!MR$&<`27`),D`DM01R:
M!)PP`Q`'"[email protected])WC`?^H2BM&X`#>4`.I([email protected]`9V70S(`3D`!K"`'%`3Q$`>``%4``V\@3:P#4&`%'@#9:!5
ML``00`2:PAI(B:H"#5`*,5`&W,!*/`*[email protected]@[email protected]`A1B0ER((``*Y,0T,`?^
MQ!MP`XWB412%,K`2Q\!#S`,KL0V\`3*0!LP`5(041)$.2$0Q4`M6M08`9>`-)`2Q^1#BP*>!`JB`4:%$FRH$RX!%!0*(`B\O"%LI%[email protected]_6(98(J%
MXE#[email protected]*[email protected]%'4`#K\(GFL6W*`;>@[email protected]%$X1(DI$BKCKW,`;Z#-CX"W"`3A0
M!I`""[email protected]*("`,M(IF\1#30!D(%7#1+!*:[email protected],2`HJI,2).Q(H(`EPC;'R+
M8=$LHD7:6"F(HE%$BLXQ#<3&W#@;:V-2P(USH`[,1L#89]S`&1"-=G$,X$6]
M^"K2P%'\BM'Q,Q*:[email protected],"B(!1J$22*1)(8*DXB66R.KQ$$T$00L!3CHAL0
MC6&`#)Q&.=!G_H1Y1!5)@3"*`3:@'>7B8*041Z$I-(4SD!2QHE'8BEW1/2+%
[email protected]?T^):;(N+43)^1DUQ'!GB'*"-J\(,:,>5>`MN`@A<2261/^X&TGB2PP5KA$$+$9#\1O9(U]T`W21(8:[email protected](M3DE58
M2%1A!J1CJ)@#:I$MND6_"!ASXTXL%(7Q,/Y($)`$=MUTU(RM410!*I(R[KJ#P"2+(H2\C8&"0LI)[email protected]%`6?X8D^4312`23P!)I`$)@")O$)
M%`$FP`188A*8`G8B""2!)@`[email protected]"E!`%7(`A(`2D0!)P`%[email protected]"10!27H$DL"A-
M8A'P%%*`"(#$)P`2D<"E!`%3X`D8`2H0*J5`$6`!HC$).`$[406(@+`\`GSB
M4VY**,`$4*6L[)6DTE2B2DBI*T%`[email protected]"4F`(((%2206"@!!(`LV2"F0!/N$$
M9*418)5.(%[email protected]+$&`[email protected]"HK$(6($B8"IQY;;\E(NR43[*2#DI*Z40*`(@
MH%EZ2R8`,(W`$Y`"Y')M!*)`%:"7I])[email protected]"B/0*JTF`BS5(Y+(D`[email protected]#)5)7;[email protected])+5WE%*@"
[email protected]%D:02D0*,,F$]@"E!+(P`"[email protected]#'!)9`D65V2Y09%'WFN929*_$*(($U
M,3419KLLE7R"8B8!C6DQB6;&-)6DDF*NQ'1Y!)JER1R6`/[email protected]%B2T5Y<$\
ME4_`:&9-N.D$5J*HO)3'TF)[email protected]+=5FN_R:Z5)KM;&DPI8"C1)M=LU;>RERY*WME$>B)"U$T!H&N*!+EP)[0CT*R/Z)$E:@D1R<;
M``)-X2]6BO98$.A`'[email protected]!;:`.U,0Q(!/[email protected]`N(%(D2*)(`)%`;(X7G!`$OP%"\
M`37P`K!`#(@!+\!2T($7,!;EP()\`62`#."!%_`[email protected]`.L`$O0%*,@1>0+:?`
M"\"&+D`;[email protected]$68`=,@@[email protected]!1S/%P`#<(!)N`$ZH`;0`[email protected]`S["H0B010`/
MP`$00`)\[email protected]@K`:)Y,[email protected]'=H`[email protected]*[email protected]$LQC.P)Q9H5J04!50!N$86D`;V
M1%.8%+8"[email protected]&]H1`,(\7U(!BPSU1*,/`1]P",,`%N(`;T`6XI)*LC6S`
[email protected]"H`,[!GE(6(;[email protected]@#[email protected]`[email protected]!'O0HGH$0JA#[email protected]&VLH"#@@PY1/F-`
M2X(:L*!\[email protected]"9*$Y>H0KP(AI%!;@$2``-FP!"0`2U4!_C",=!$+T(:``%B
M]!Z645MX%@<%"$`!690.;-$NRC);:'#[email protected]@!8``7\A%5X)=7*.J$P0\T1ZP
M$AS%/(04;^"*7H0JVA04Z46X`VV44,!1\CA'9T`=-9/,`H_J43[:/A'I&GVD
MD?2-3E(MRD4M:1"PH]%1D^Y1_NDF]&AP1`%[email protected]`?M6A/)0!F=%$8!C?K"--!$>ZD,_9D'E``[email protected]
M`3>!!2S07:J$"@4E+:5#@`:$44X:`YIH'*VD0Z`&C%-O2DJ[:#IMH662!*@$
M<_I-Z:@XA:;8M)PJQ'-:2HF`.H6F_)2._M,[>@.,@`P%D)4"[email protected])3'2I/9P`W
M;18"@2"@T'4J`VJ`"[email protected]!-Z`&-%%LN`5BP#K=H1%U#4Q4X(E#C<(UE0$W(`7H
M48;H&BO%[email protected]'^NCM0CVI1F"GLE0.VD239!JE`0>U*A8*I+`HG.*>V76^L";0QP.Y2*5H(^T(4E4;
M*HHV\"%10`S8I"#@!/B*.7$"1B-7[0B,]*UVA(`Z`\2`&5BGI?23LE,Y6DK)
M``[`JV$`!NA5NDH&[FD:E:=0U9'NU4I*!O[I8<4)>I6C&E9?"%*!:E&UJ4>U
MH2I5A.I3%2M'=:R4=:;6U)MZ4E/J2N6LD16FVH"A:EE':V;-J9NUIZ96H,I"
MH6EEE:B7E;3*TZ1Z6F6K8G6JOA`'\-*PJ$AA(F+LDF)U4935L[I*U6J<8*MN
MU9'&5<5*5^[email protected][6>[email protected]"_"[email protected]%*W75KH45K\Z`Q%H2""MH9:73U+/"U,GZ
M1D6K4M+K6^2M3:FD;[email protected]"9UM\[email protected]"L($*Z.U)A>40V[3*5H,WVFC]5-S`!J
MRF>L*3;5B-NTB=9([email protected]`*.`%V`D_H"3[A)P"%H"`4PO,Q,@I#&BDF175T
M`YA"4W`*3W$"0(6H(!4,\E2DBE71*EX%8)P#)^"AKMA)P0;<@(JUD2WVQ=(*
M6X$K0BJO6*MN([email protected]\5V(19:\LBZVR9I;*6EF%.&7+P(L]%I-"60Q*;`@MTNI#
M#:05E)!V2-?8`FJ"752D;=;!IH$N0#]U0`[email protected]`6W5SJ;9%FM`Z241Z(D&,=)*
MVDE+:2NMI;VTF#;3:MI-RVD[K:?]M*`VU(K:44MJ2ZVI/;6H-M6JVE7+:ENM
MJWVUL#;6RMI92VMKK:V]M;@VU^K:7ZVO_;7`[email protected]*VV%+;(NML3VVR#;9
M*MMERVR;K;-]MM`VVDK;:4MMJZVUO;;8-MMJVVW+;;NMM_VVX#;8K;D]M^@VW:K;=`
end


[next article]
From: [email protected] (H.J. Lu)

Subject: Re: X, Mice, shared libraries, etc.
Message-ID: <[email protected]>
Date: 9 May 92 02:25:06 GMT
References: <[email protected]>
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 45

In article <[email protected]> [email protected] writes:
>Well, I've got pre-0.96 running (THANK YOU LINUS for telling us about the
>shared library permissions), and X11R5 is chugging along happily... well,
>mostly.
>
>I've got a Mouse Systems White Mouse (serial). This little gem has a switch
>on the bottom to switch between Mouse Systems (3 button) mode and Microsoft
>(2 button) mode. Needless to say, I prefer the 3 button mode. Unfortunately,
>I can't seem to get Xconfig right for a mouse systems mouse. My last try
>was:
> MouseSystems "/dev/ttys1"
> Baud 9600

First ot all, I don't think your mouse supports 9600 baud. Check your
manunal. I will use 1200.

Most of mice need some kind of initialization. I am not sure what Linux
has done to the serial port. You may have to play with it. I have a
testing program for SVR3, which will read the serial port and report
the mouse status. If necessary, I will post it.

>I've tried it with and without the baud line, and on different serial ports,
>but I can't get the thing to work. Has anyone had any success with the
>Mouse Systems mice?
>
>In the meantime, I'm just using X with the Microsoft mouse configuration.
>
>Now about the shared libraries... I've got my old Lib92.04.06 from MCC 0.95c+,
>and now I've got Lib92.04.12 that came with X. Do I need both? How does
>a program choose which library to start up? Will it just use the most
>recent? Are they backwards compatible?
>
>Lastly, I've pulled xdvi off of the net, and I'm configuring that for Linux.
>So the big question is: is our version of X a complete implementation of
>X11R5, and are libX11.a, libXt.a, etc. complete?
>

I think it is complete.

>--
>Darren Senn Phone: (408) 479-1521
>[email protected] Snail: 1785 Halterman #1
>Wasurenaide -- doko e itte mo soko ni anata wa iru yo. Santa Cruz, Ca 95062

H.J.


[next article]
From: [email protected] (Court Demas)

Subject: X Problems..
Message-ID:
Date: 9 May 92 03:58:29 GMT
Organization: Sophomore, Math/Computer Science, Carnegie Mellon, Pittsburgh, PA
Lines: 43


Ugh. Everything has been running really well until I tried to put X on
my system..


I downloaded and installed the X386 release. I followed all (I hope)
of the instructions and proceeded to type 'startx'. I first got the
same error everybody else has been getting (no tty0), so I checked
c.o.l. tried 'mknod /dev/tty0 c ? ?' - whatever somebody had posted as a
solution. I tried startx again only to have my screen enter some
strange mode with vertical white-striped garbage all over the place.
Reboot. I tried a number of different modes (800x600, 640x480, etc) and
the only difference I saw was that the screen when black and hung in
800x600 (instead of the white stuff).

Could it have anything to do with my tty0? (I'm totally clueless about
Unix devices). My system is fairly standard:

-Zeos 486/33
-8mb
-Microsoft Mouse
-Samsung MultiSync (from a few years ago - does 800x600 Windows fine,
and 1024x768 @30hz or something- gross)
-Diamond Speedstar+ w/1mb
(seems like I have a Hi-Color too, but I'm not supposed to! I ordered
the normal one, but it came with a 32k driver and when I used it it
worked! I have modified the colors in windows and there is absolutely
no noticable dithering of any colors I try. Strange..)


After all the posting about Tseng4000 boards, I figured I be fine! Has
anybody else had problems like this? [I used the bootimage that was
included in the X386 release (I figured it would be my best bet).]



thanks,
-court


on another note:
I am going to try to get the Tierra simulator running. Has anybody
tried this yet?


[next article]
From: [email protected] (Ken Corey - Operator)

Subject: 20MB drive & wdxt-gen2 controller on 386?
Message-ID: <[email protected]>
Date: 9 May 92 04:56:49 GMT
Sender: [email protected] (UNT USENet Adminstrator)
Organization: University of North Texas
Lines: 25


Has anyone used a ST-125 drive and a Western Digital WDXT-GEN2 MFM controller
with a 386?

Looking up the Seagate ST-125 in TheRef42, I fount that it's 615 cyl, 4 head,
and 17 sectors.

The WDXT-GEN2 is a 1/2 length, XT/AT controller.

I've also found out that I've got an 8-bit future domain SCSI controller.
(The ST01/02 boot doesn't work)

I want to use the 20 MB to get Linux running, and then put it on the much larger
hard drive.

However, right now, the WDXT-GEN2 controller refuses to see the ST-125.

The cable that I've got has the twist in one connector, but not in the middle co
nnector. Is this standard for MFM drives?

There's one jumper on the controller, marked as J4.

The SCSI controller is mapped into CA00, so I'm assuming the MFM drive is at C80
0.

Can anyone suggest anything to try out here?

-Ken


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: 20MB drive & wdxt-gen2 controller on 386?
Message-ID: <[email protected]>
Date: 9 May 92 05:46:23 GMT
References: <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 19
Nntp-Posting-Host: ophelia.cs.colorado.edu

In article <[email protected]> [email protected] (Ken Cor
ey - Operator) writes:
>
>Has anyone used a ST-125 drive and a Western Digital WDXT-GEN2 MFM controller
>with a 386?
>

It won't work under Linux. The XT and AT controllers are not compatable
on a register level.

>I've also found out that I've got an 8-bit future domain SCSI controller.
>(The ST01/02 boot doesn't work)
>
>I want to use the 20 MB to get Linux running, and then put it on the much large
r hard drive.
>
>However, right now, the WDXT-GEN2 controller refuses to see the ST-125.

You need a different controller to use it with Linux. Talk to
local techies and find one pulled from equipment for an upgrade -
should set you back ~$20 at most.


[next article]
From: [email protected] (Erik Fichtner)

Subject: Mouse Wont Work! (was Re: X Problems..)
Message-ID:
Date: 9 May 92 06:37:53 GMT
References:
Organization: University of New Mexico, Albuquerque
Lines: 28

X/Windows! YAAAAY! Well.. Kinda. A friend of mine and myself are sitting
here fighting with X on Linux. The kernel is the pre-0.96 with obz's patches.

after a good bit of fighting, we've got X to display things correctly. Looks
very nice. Only the mice don't work. neither one. I've got a Logitech ClearCase
(the old C7 type with three buttons. flat square thing) and friend has a
Logitech Series 9 three button.

They're doing random things instead of actual mouse movements. rather unplesant.

The normal logitech mouse type doesn't function at all. They're currently set
to Mouseman and Mousesystems. and not working right. anyway....

Does anyone have any suggestions? We're confused totally.

PS: the friend here ([email protected]) says that the mouse is doing
the same incorrect response that it did with the selection-1.0 package. I
wouldn't know, because I didn't run selection.

Respond to the newsgroup incase someone else is frustrated by the same thing..
(or email [email protected] or [email protected])

Thanks in advance.

--
Techs [email protected]
Disclaimer: any inaccuracies are caused by the phase of the moon
"...whaddya mean I ain't kind... Just not *your* kind!" - Megadeth


[next article]
From: [email protected] (Oren Kastner)

Subject: Some questions...(long)
Summary: /dev/floppy0 and /dev/lp1
Message-ID: <[email protected]>
Date: 9 May 92 07:17:29 GMT
Sender: [email protected] (USENET News System)
Organization: Brandeis University
Lines: 97


This is somewhat long, so if you are impatient or generally not
interested, I suggest you move on right now.

^L

Howdy,
I was using Linux when I came across a problem that I didn't have the
first clue as to how to go about solving it. So, I figured I'd ask the
"Net's" opinion...
It all started a couple of nights ago when I was trying to move some
of my files to floppies (some tar.Zs) so I'd have more room for 0.96 when
it comes out next week. I mounted my /dev/floppy0 (5.25, 1.2M -- A: under
MS-DOS) disk drive on one of my mount points. I moved some of the files
over and things were cool... 'df' reported the hardisk info and the mounted
floppy info correctly. Next, I umounted floppy0, replaced it with another
disk and re-mounted it on the same mount point. I did an 'ls' on the newly
mounted disk, and to my surprise, the stats for the FIRST disk were given
back to me. I tried 'df' and again, the stats for the FIRST disk were given
to me. I was baffled, to say the least, but, I thought maybe I'd write
something to the disk and maybe things with "sync". Not a good idea... I
copied the files I intended to copy onto the disk and in "mid-copy", the
disk was full. The ACTUAL disk had more than enough space for the files, but
the FIRST one didn't -- and since the stats of the FIRST one were kept, the
filed didn't fit on the disk. I tried 'ls' -- and I don't exactly remember
what happened... It was strange enough, that I tried to fsck the disk and got
a bunch of errors. When I did 'ls' on the disk, again, if I remember
correctly, I had a couple of 0 length files and one large 600K+ file which
looked as if contained ALL of the files put together. This file was a tar.Z
file (well, the name was ๐Ÿ™‚ -- I tried to zcat | tar, but the file was
corrupted and no go. I played around with the "problem" (feature? ๐Ÿ™‚ a bit
more, and managed to trash a couple of disks so I had to re-mkfs them.
If I remember, a few months ago, in one of the DOS groups there was a
discussion of some floppy drives that don't "update" correctly (or something
to that effect) and you either need a driver or to physically enable/disable
something in the floppy drive... some wire, maybe (I'll have to call Zeos
and find out if they know anything about it -- as I've had A: report same
stats on two different disks before -- but at the time, I figgered maybe it
was one of my device drivers that was causing it to mess up...)
So tonight, I decided to check that I wasn't imagining anything and
decided to try to re-mount two disks and check if the stats for the second
were the same as the first. I mounted floppy0, copied some files to it,
did an 'ls' on it, 'df', umounted it, and remounted floppy0 with another
disk in the floppy drive. I did 'ls' and 'df' -- and just like the other
night, I had the same problem -- the stats for the second disk were the same
as the first! (This time I didn't trash anything - experience is EVERYTHING ๐Ÿ™‚
I umounted the floppy0. Out of curiousity (as usual), I decided I'd do a
'cat /dev/floppy0'. I did so and my virtual console hung. I couldn't
control-c, control-z, etc out of it. So, I logged onto another virtual console
(again, as root) and tried to kill the process. I tried signal -15 first,
and when that didn't work, I tried -9. When THAT didn't work, I tried from
1-15 and none worked. *THAT* is what got me to finally post. Shouldn't signal
9 kill ANYTHING unconditionally? Is the problem I encountered just that, or
am I misunderstanding something simple? Any explanations would be appreciated.
As long as I have your attention(s) (or not :), I was wondering...I got
the source code for banner and compiled it with no problem on Linux. It ran
fine and I decided I'd redirect to /dev/lp1. I used:
=> ./banner Oren > /dev/lp1
(the ./banner is because I was root and I don't have '.' in my path ๐Ÿ™‚
It worked perfectly and I got the banner on my printer. It was too large,
so I wanted to shrink the size of the banner. First I did a:
=> echo "^l" > /dev/lp1
(echoed a control-l, formfeed). That worked perfectly. Next, I did:
=> ./banner -w60 Oren > /dev/lp1
THIS is where the problem began. I got my prompt back immediately and nothing
on my printer (no output). I did a 'ps' -- and there wasn't any "banner" job.
I tried to repeat the command, but with the exact same effects (nothing
happening). So, I decided to redirect the output to a file and then cat the
file to /dev/lp1. When I tried to cat the file to /dev/lp1, I got an error
about lp1 being "busy" (if I remember correctly). I tried again, and same
error. I echoed a " " (space) to the printer, and that worked fine. Tried to
cat the file to the printer -- got the busy error, again. Echoed another " ",
and that went fine. That's where I left it... don't understand what I did to
mess it up.
Finally, the sources for the "cursive" program were on my school system.
Thus, I figure its ok if I distribute them to anyone who wishes... I didn't
have to change a thing (not that I'd know how ๐Ÿ™‚ and it seems to run perfectly.
If anyone is interested, you are welcome to the sources (unless someone tells
me that I shouldn't distribute them).
This is cursive, btw:
_ , _ _ ______
' ) / // // / /
/--/ _ // // __ --/ /_ _ __ _
/ (_

Thanks again for any info that might help me better understand what the heck
is going on ๐Ÿ™‚
Oren.

ObOfCourse: Thanks to Linus and "The Gang" for Linux and all their efforts!

--

Oren Kastner E-mail address:

Brandeis University [email protected]


[next article]
From: [email protected] (Hien Luu)

Subject: fs.h and stty.h
Message-ID: <[email protected]>
Date: 7 May 92 04:30:15 GMT
Distribution: usa
Organization: Station Zebra Corp, Sunnyvale CA Ph:408-739-1520
Lines: 6

I was trying to compile the code for ps, and the compiler
complained that it couldn't find /linux/fs.h and /linux/stty.h...
where can I get these files...

thanks in advance.
Hien


[next article]
From: [email protected] (Hien Luu)

Subject: Missing file
Message-ID: <[email protected]>
Date: 8 May 92 06:04:19 GMT
Distribution: usa
Organization: Station Zebra Corp, Sunnyvale CA Ph:408-739-1520
Lines: 10


I just noticed that file tzfile.c is missing from the
time.tar.Z package..if any one has this file, could you please
send it to me...

When I was trying to compiled the df.c, the compiler complains about
the constant SUPER_MAGIC, which is not defined any where in the df.c program..
does the above constant belong to any include file ...?

Thanks in advance..


[next article]
From: [email protected] (The Master of Symbolic Links)

Subject: Re: X386 (Was Re: 0.96 out next week)
Message-ID: <[email protected]>
Date: 9 May 92 10:23:38 GMT
References: <[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
<
Sender: [email protected] (USENET Newssystem)
Organization: Inst. fuer Informatik, Technische Univ. Muenchen, Germany
Lines: 26
In-Reply-To: [email protected]'s message of 7 May 92 15:53:18 GMT

>Why would you need at least 256 colors? Most of our SUN Sparcstations
>run X on big B&W displays and for most tasks, B&W is acceptable. On a
>SVGA, 16 colors would be very easy to provide up to 800x600 (i believe
>Win 3.1 ships with a generic 800x600 SVGA driver) and would be enough
>for many applications, except viewing GIF pictures of course ๐Ÿ™‚

Try to run Motif and then open a few color windows ....

>What chipset is used by the ATI VGA Wonder+ (i own one)? Is it a
>proprietary ATI chipset? If so, why is it not supported? Lack of
>features? Non-standard? It seems fast enough, although much slower
>than ET4000 i suppose...

Thats simple. The problem is:

a) I don't have access to an ATI VGA board
b) I don't have access to their databooks without trouble.

- Thomas
--
-------------------------------------------------------------------------------
e-mail: [email protected]

immer ?
nein, nicht immer ...
... aber immer oefter !


[next article]
From: [email protected] (The Master of Symbolic Links)

Subject: Re: VGA graphics library for Linux
Message-ID: <[email protected]>
Date: 9 May 92 10:27:20 GMT
References: <[email protected]>
<[email protected]>
<[email protected]>
Sender: [email protected] (USENET Newssystem)
Organization: Inst. fuer Informatik, Technische Univ. Muenchen, Germany
Lines: 22
In-Reply-To: [email protected]'s message of 8 May 92 07:21:37 GMT


>Yes, it would be really nice if would could use the VESA standard, but
>this requires acces to the video BIOS, which is not possible in Linux.
>I'm afraid we can't avoid hardcoding.

No, that's the facy thing behind it. SVPMI is VESA standard for protected
mode applications that don't have access to the BIOS. It's a plain
ascii file that has some data-sections to describe the HW (which kind
of banking, how many colors etc) and some programm sections (in a simple
to implement meta language) that set modes, banks, the RAMDAC and so on.

Try to call VESA and let you send the SVPMI standard document.

- Thomas

PS: Maybe there will be soon a sample implentation available for free ๐Ÿ˜‰
--
-------------------------------------------------------------------------------
e-mail: [email protected]

immer ?
nein, nicht immer ...
... aber immer oefter !


[next article]
From: [email protected]

Subject: FAQ next release
Message-ID: <[email protected]>
Date: 9 May 92 10:30:54 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 8


I will post the FAQ to comp.os and news.answers in the mid of this week.
This release will not be uploaded to the ftp sites.

I'm wondering if it's worth to keep on uploading stuff to tupac; any advice
wellcome

Marc


[next article]
From: [email protected] (Mika Pekka Liljeberg)

Subject: Re: X Problems..
Message-ID:
Date: 9 May 92 12:40:46 GMT
References:
Sender: [email protected] (Uutis Ankka)
Organization: Department of Computer Science, University of Helsinki, Finland
Lines: 80
In-Reply-To: [email protected]'s message of 9 May 92 03: 58:29 GMT

In article [email protected] (Court Demas
) wrote:
>
> Ugh. Everything has been running really well until I tried to put X on
> my system..
>
>
> I downloaded and installed the X386 release. I followed all (I hope)
> of the instructions and proceeded to type 'startx'. I first got the
> same error everybody else has been getting (no tty0), so I checked
> c.o.l. tried 'mknod /dev/tty0 c ? ?' - whatever somebody had posted as a
> solution. I tried startx again only to have my screen enter some
> strange mode with vertical white-striped garbage all over the place.
> Reboot. I tried a number of different modes (800x600, 640x480, etc) and
> the only difference I saw was that the screen when black and hung in
> 800x600 (instead of the white stuff).
>
> Could it have anything to do with my tty0? (I'm totally clueless about

Nope. Only console input is handled through tty0.

> Unix devices). My system is fairly standard:
>
> -Zeos 486/33
> -8mb
> -Microsoft Mouse
> -Samsung MultiSync (from a few years ago - does 800x600 Windows fine,
> and 1024x768 @30hz or something- gross)
> -Diamond Speedstar+ w/1mb
> (seems like I have a Hi-Color too, but I'm not supposed to! I ordered
> the normal one, but it came with a 32k driver and when I used it it
> worked! I have modified the colors in windows and there is absolutely
> no noticable dithering of any colors I try. Strange..)
>
>
> After all the posting about Tseng4000 boards, I figured I be fine! Has
> anybody else had problems like this? [I used the bootimage that was
> included in the X386 release (I figured it would be my best bet).]

Well, I got sick of bitching and moaning about my Trident 8800 board
(I was pretty fed up with it anyway) and bought a Boca Super X VGA ;-),
which has an ET4000AX chip an 1MB of memory on it, as well as Sierra
DACs (surprise!).

Right. I plugged it in and tried X. Well, X386 won't play ball with me
either! ๐Ÿ™ I tried the default modes and when they didn't work, I
calculated a bunch of new ones based on my monitor's (NEC Multisync II)
specs (automatic horizontal sync from 15.5 kHz to 35 kHz). I'm pretty
sure I did it right, too. I can easily get a stable picture, but instead
of a single beautiful one I get two ugly ones, side by side. Both contain
the visible desktop, only squeezed in half along the X axis. I suspect
it's the Sierra DAC chips that cause this behaviour. Can anyone confirm
this? How do I get around the problem?

And then there is the mousie, of course. ;-( I've got a Genius mouse,
which can act either as a Microsoft mouse or a MouseSystems mouse.
It works, too. Not under X386, though. In Microsoft mode the mouse
cursor gets whopping mad, moving in quick jerks in the same general
direction I'm pushing the mouse. I've got no control whatsoever over
the buttons. X seems to think someone's bouncing on them, though. ๐Ÿ˜‰
In MouseSystems mode I get no response at all. Is the X mouse driver
picky, or what? ๐Ÿ˜‰

>
>
>
> thanks,
> -court
>
>
> on another note:
> I am going to try to get the Tierra simulator running. Has anybody
> tried this yet?

Not me.

Mika
--
Mika Liljeberg Email: [email protected]
Helsinki University [email protected]
Dept. of Computer Science


[next article]
From: [email protected] (Denise Tree)

Subject: SCSI boot problem
Message-ID: <[email protected]>
Date: 9 May 92 14:10:34 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected] (Denise Tree)
Organization: The Internet
Lines: 28


Hi Linuxers,
I traded a Future Domain for a st01 and it's time for installation
#4...but I got trouble. I have the pre0.96 vmunix boot disk and it says:L

Host 0 is detected as a(n) Seagate ST-01/ST-02

1 host adapters detected
Detected scsi disk at host 1, ID 0, lun 0
Detected 1 disk 0 tapes toral
Warning:SCSI device at host 1, ID0, lun 0, failed READ CAPACITY
status = 1, message = 00, host = 00, driver = 00
sense not available
block size assumed to be 512 bytes, disk size 1 GB
cannot read partition table for sd0
Insert floppy and press ENTER


CC
129/1200 free blocks
270/400 free inodes
Ans at this point there is nothing more ....keyboard is not locked though.

Any help or suggestions would be greatly appreciated... I hope this is just
some dumbness on my part.
.........................
Denise Tree
[email protected]


[next article]
From: [email protected] (Peter Orbaek)

Subject: Re: Getty on the serial port?
Message-ID: <[email protected]>
Date: 9 May 92 13:55:53 GMT
References: <[email protected]>
Sender: [email protected] (Peter Orbaek)
Organization: DAIMI: Computer Science Department, Aarhus University, Denmark
Lines: 22

[email protected] (Alan B Clegg) writes:

>Well, on a similar tac, but a bit farther along, I have been able to get
>a getty running on my serial port, but would like to be able to go
>bi-directional with the getty only grabbing the port on Carrier-Detect
>when the port is not already in use. Has anyone ported or worked on
>a UUGETTY type program yet?

Yes. It's called modem. I will put it on ftp.daimi.aau.dk tonight.

The original port to Linux was done by Daniel Hagan ([email protected]),
and then I hacked it some more. Using it I have been able to log in via modem.

Look for ftp.daimi.aau.dk:/pub/Linux-source/modemctl.tar.Z tomorrow.

- Peter ([email protected])

--
Peter Orbaek ----------------- [email protected] | ///
Hasle Ringvej 122, DK-8200 Aarhus N, DENMARK | ///
| \\\///
"Strong typing is for people with weak memories" | \XX/


[next article]
From: [email protected] (Bill Bogstad)


Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <[email protected]>
Date: 9 May 92 14:38:43 GMT
References: <[email protected]>
Organization: Johns Hopkins Computer Science Department, Baltimore, MD
Lines: 66

In article <[email protected]> [email protected] writes:
>...
>What I thought I heard as the general consensus (although I may be
>biased since that's what *I* believe :-), wasn't that sources was bad
>and that you should upload binaries. Rather, the consensus was that
>people should upload sources or diffs whenever possible, but making
>binaries available shouldn't be considered a bad thing.
>...

I've never had any problems with binaries being distributed, but I
have complained in the past about sources for some packages either not being
clearly labeled as such and/or not being distributed at all. Well, a recent
posting in gnu.gcc.announce by Richard Stallman may be relevant to this

discussion. [BTW, I did not bring this to his attention. It isn't my
fault. :-)] To summarize: He considers that FTP sites that provide binaries
of programs covered under the GNU public license (GPL) without providing
FULL source code (not just diff files) "side by side" to be in violation of
the GPL.

I don't know if this is an official statement of policy for the Free
Software Foundation (FSF); but if anyone can make such a statement I would
think that he can. He specifically is commenting on GCC; but it would
appear that his comments would cover anything under the GPL for which he or
the FSF hold the copyright. I'm including the full text of his message
which appeared in gnu.gcc.announce below.

Bill Bogstad
[email protected]

Newsgroups: gnu.gcc.announce
From: [email protected] (Richard Stallman)
Subject: Executables and anon ftp
Message-ID: <[email protected]>
Sender: [email protected]
Organization: GNUs Not Usenet
Distribution: gnu
Date: Fri, 8 May 1992 10:31:22 GMT
Approved: [email protected]
Lines: 27

I've heard that there are ftp sites distributing executables of
versions of GCC, without full sources on line side by side with them.
This isn't allowed under the GPL, and for good reason: ensuring users
have easy access to the source code is one of the main aims of GNU.

This is not a matter of deliberate wrongdoing; the source code for
these versions is free and is circulating. (The people who wrote the
modified versions aren't in control of what ftp sites do.) But it is
still a significant matter because it can cause trouble for users.

Sometimes there are patch files available giving changes from an FSF
release, such as 2.1. But this is not sufficient. For example,
suppose the user copies the binaries and patches now, then tries to
use the patches in a few months when a problem arises. At that time
it will be hard to find a copy of GCC 2.1 sources to patch. Instead,
2.3 or 2.4 will be current--but the patches may not work in those
versions, due to the changes we will have made in GCC.

This is why the GPL says that the complete sources have to be made
available. In the case of distribution by anonymous FTP, users can
choose to copy just the executable, but they must have the option of
copying the source as well. So the source has to be on line alongside
the binary.

If you see an ftp site that has binaries available without complete
sources, please bring this to their attention.


[next article]
From: [email protected] (Alex R.N. Wetmore)

Subject: hard drive crash question
Message-ID: <[email protected]>
Date: 9 May 92 14:42:37 GMT
Organization: Freshman, Math/Computer Science, Carnegie Mellon, Pittsburgh, PA
Lines: 23

I made a pretty big mistake that crashed my Linux partition yesterday,
but I was wondering if someone could explain what caused the crash.
Here is what happened.

I had a floppy mounted on /dev/at0, and accidentally removed it before
umounting (silly mistake). I saw the drive start to sync, and quickly
closed the drive (I had just started pulling it out). I then umounted
(which didn't work, I kept getting error 16, which I believe is device
busy, but I wasn't in /mnt). So I synced and rebooted, but linux when
the kernel attempts to load init off of the hard drive it crashed (I hit
Scrl-Lock, and there where only two processes (0, which is swapper, and
1 which is init).

Any theories? I don't really care about loosing my HD, I got everything
back, but I am curious as to what happened here. I didn't loose the
whole thing (I booted off of floppy and fscked, and it reported lots of
errors, and after fixing it, it still had a lot of broken files, so I
just reformatted).

Also, hows MGR going? If it gets ported will there be an EGA driver (I
would be willing to help out on this).

alex


[next article]
From: [email protected] (M. Saggaf)

Subject: xv and xcalendar at tsx-11
Message-ID: <[email protected]>
Date: 9 May 92 15:09:08 GMT
Sender: [email protected] (News system)
Organization: Massachusetts Institute of Technology
Lines: 8
Nntp-Posting-Host: e40-008-7.mit.edu


I have uploaded two X programs to tsx-11: xv and xcalendar. (xv is
the best image viewing/manipulation program I've seen, great for
viewing gif pictures, and can handle jpeg also). Check them out.

--------------------
M. Saggaf
[email protected]


[next article]
From: [email protected] (Joel M. Hoffman)

Subject: Re: X and 9600 bps
Message-ID: <[email protected]>
Date: 9 May 92 15:50:52 GMT
References: <[email protected]> <[email protected]
du>
Sender: [email protected] (USENET News system)
Organization: University of Maryland at College Park
Lines: 21
Nntp-Posting-Host: wam.umd.edu

In article [email protected] (Al Clark) writes:
>In article <[email protected]> [email protected] (Joel M. Hoffman)
writes:
>>[...] I can't go above 4800 even without
>>X (direct RS-232 connections). [At 9600bbs I lose chars.]
>>
>>-Joel
>
>
>I have been using kermit with a Robotics Sporster 9600, using 19200. The only
>possibly unusual issue is that I've got a 16550 UART. I get the normal
>9600plus display update, with no observed errors. File transfers rarely show
>an error.


Perhaps the problem is that I'm using a direct serial connection? Is
it possible that the modems are doing some sort of buffering that Linux
needs? I know the serial ports can go up to 116,000-ish, because
Laplink does it. Why does Linux give up at 9600?

-Joel
([email protected])


[next article]
From: [email protected] (Britt Park)

Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <[email protected]>
Date: 9 May 92 16:54:04 GMT
References: <[email protected]> <[email protected]
cs.jhu.edu>
Sender: [email protected] (Mr News)
Organization: DSG, Stanford University, CA 94305, USA
Lines: 25


I think that a few comments about the GNU license are necessary. I do
not believe that the portion of the GNU license which states that modifications
to GNU software must fall under the GNU license has any legal basis. The
rights to the modifications to a program invest wholly with the author of those
modifications. As to the requirement that complete sources to a binary on an
archive site must be located on the same archive site; I don't believe this
has any basis either, at least not as the GNU license is currently phrased:

"You may copy and distibute the Program ... in object code or
executable form under the terms of Paragraphs 1 and 2 above provided that you
also do one of the following:
... Accompany it with the information you received as to where the
corresponding source code may be obtained."


Certainly I believe it's better for archive sites to make available
complete sources rather than just diffs. This is not, however, always
practicable. It costs a great deal in resources to maintain an archive site,
and GNU software, as we all know, weighs in on the hefty side. I'd prefer to
see archive sites with binaries only rather than archive sites which fold
because they don't have the space to hold all the sources.

Britt Park
[email protected]


[next article]
From: [email protected] (K. LIN)

Subject: Re: X windows video card support?
Message-ID: <[email protected]>
Date: 9 May 92 14:08:26 GMT
References: <[email protected]> <[email protected]
ik.TU-Muenchen.DE>
Sender: [email protected] (Netnews Administrator)
Organization: McGill University, Montreal, Canada
Lines: 7

Hi, there!
Is X386 support ATI VGA1024 card w/ 512k? Or is somebody working on it?
In the vga.dbase there is an entry for ATI VGAWONDER, is it the same thing?
If it is, is it supported by the linux port?
When I tried to run startx, I got a black screen for all the concoles.
But they still get input from the keyboard. I guess I'm in the dark.
--Kebang


[next article]
From: [email protected] (Jim Winstead Jr.)

Subject: Re: MGR question
Message-ID: <[email protected]>
Date: 9 May 92 17:00:11 GMT
References: <[email protected]>
Sender: [email protected] (The News System)
Organization: Harvey Mudd College, WIBSTR
Lines: 37

In article <[email protected]> [email protected] (Alex R.N. W
etmore) writes:
>Also, hows MGR going? If it gets ported will there be an EGA driver (I
>would be willing to help out on this).

If you want to know more about how MGR is coming along, join the MGR
mailing list, which is one of the channels of Mail-Net. To subscribe,
send a message to [email protected] with 'X-Mn-Admin:
join MGR' in the first line. (I think that's the right address -
either that or linux-activists-request.)

A quick summary: I know of only two people working on MGR at the
moment, myself and Dan Stromberg. He is working on a Hercules
version, and I'm working with the VGA code that Tommy Frandsen posted
to this newsgroup recently.

Dan is currently making loads of progress :), and I haven't made much
progress 'cause I haven't had time. ๐Ÿ™ I'll do what I can over the
next week, but I have finals coming up and lots of other stuff to
catch up on (I've also been playing Caste Wolfenstein 3D, which is
really addicting. :). MGR will likely be one of my summer projects.

(Oh yeah, I should really finish putting together a root disk, too.)

It will be possible to support different video systems (possibly even
SVGA) under MGR by writing your own video drivers. MGR for Linux will
(probably) be distributed as one mgr.a file (and a few clients), with
various video.a files available. For example, if you were running
under VGA, you would get the main MGR distribution, and the VGA
driver. To compile MGR, you would simply do a 'gcc -o mgr mgr.a
vga.a'. This makes things easier for development (since MGR really
doesn't seem geared to supporting multiple video cards), and keeps the
executable size down.
--
+ Jim Winstead Jr. (CSci '95)
| Harvey Mudd College
| [email protected]
+ This is all my words. Honest!


[next article]
From: [email protected] (H.J. Lu)

Subject: a testing program for serial mouse
Message-ID: <[email protected]rval.net.wsu.edu>
Date: 9 May 92 15:29:24 GMT
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 233
Originator: [email protected]

>
>
> I'm a linux user.... downloaded X11 the day it came out... I HAVE THREE
> MICE IN MY ROOM, and I CAN'T GET ANY TO WORK UNDER X11 !!!! I have a
> mouseman.... a logitech .... and a pc-mouse.... all three have three
> buttons... none have any switches on them.... I have no docs for any...
> and I have tried each mouse at almost every concievable option in Xconfig.
> I get erratic results or none (no movement) with defferent settings. The
> serial line is fine because I have used the serial line before: "cat
> /dev/ttys1" and works... also have used selection which I had to modify
> to work.....
>
> Anyway .... you seem to know about mice... do you know which of these
> three DEFINITELY should work... and what settings I should use? You mention
> ed initialization .... yes... I believe logitech requires some... what
> to I need to send to the port before X11 starts... The logitech is a
> series 9... the mouseman is a logitech too.... I have finally decided to
> get some expert opinion on this after 6 MORE hours of playing...
>

I think they should work. Before you run X11, try this test first. If
selection works for you, see what it does to initialization in the
kernel.

BTW, this test was hacked for SVR3, you may have to modify it. I am
still waiting for 0.96 and shared X11. I will run the test after I get
those.

> Thanxs,
> Hillel
>

H.J.
--
School of EECS Internet: [email protected]
Washington State University BITNET: [email protected]
Pullman, WA 99164 Phone: (509) 335-6470 (O)
USA (509) 334-6315 (H)
-------------------
begin 666 mouse.tar.Z
M'YV0+EZT>5-G3ID7`!(J7,BPH<.'$"-*[email protected]`(B+-VK4`&'[email protected]:-CA>'!E#
MY,B3%V%M#IHR#+M!D'6MUI`T:-$S6!*G68]N1,VC(F+$R
MAHP:-FZ\C.LQ1EZ=(&"0'4QXH5"B1I$JQ=KT:=3"D"-+GIQ5P0L5"D"H`#$E
MRQ005E[,P!&23!DX9=R8=D,'1!HW8Y*:GI-9Q0L%(U['KF/Z)(\Y=,BD>1/6
M!V[=LLOX-C.&-9OBQV$G]SV&3A[4T'-+YZU\Y.\\`?S&\
M^G,TQK7O[[email protected],.T,;^=_D4>=*311AG\S=?=?_:],$<:9[@1QGOQG>M"-^%5T8;8T0%5&\-MB)YW4,GA!G'[email protected]*,\48;<*3!!E1AB:!`
M&7AX:*)KK(%0XHD[*(";:6:\IAP115B1Q!!%?)&$$TE0D0`,0Y91I!M')KED
MDT,P\<0412000Y579JDDDU\\X40",I!I)`A(GMGD$T88D<`,[email protected]`"='
M'=75([email protected]/%&4678`54>(#[email protected]#%:#[email protected]:[email protected]&":
M'9R6$:1FMXU7!I%]TO%G=2#L4>EK="[email protected]!%DG,IC:RBAA`<+>;`@1AUTT/&&
M&W/LT.M)>M(*E8-LM!J&AR"\808(3J4Q!PA?*6?;K#TN.Y)[email protected]\Z!QAMLD,$"
M"&[4T<:[email protected]'IC'),.QRBWUY$J[@CG3ALL<@YL+*+MMB$&
M5*B"RRO`^(VQ1AAGE&'PP1?I6?'%&6_[0I^@1NQON`"+$09O^7HH+\=Z#B5'
MQAZ2T2=JJYJ\*\=PR-'4&RZR\?+!>[email protected]'!'&)!:*^JT86";+[(@%OO:
M&3K_>_`89K"!\=`MV-A^.%RKK[;F]*BEZEP';[email protected]@MBP:9'R!X1QKK,OX%VGDNFRSJP58Y!CZ'EOM
MM4XI5Z0!B0PRM[Y7KS77EJFG/:.:).1TT&
MU*:C[NUM)T\LKL#&;QRVQ_6"K-R,:CME..**]SK'L&9L04,7N5^^&7#3*L>]
M&1"S#0(;;YR1AH>M9XVRN!?6X;7+>_.]LK$8#B4MM>[email protected]>F(M3!NO:$WJM-:
[email protected]!CB$#V#0LQC&S,=`]!5%=:QSG;BXMP4<@$\!?;`4IB35E4^-85TJN!1I
[email protected]*[email protected]&["[email protected]@\(4F?&8D-,D?2O34!$[Y;`[6:LVW1D"D-]'0AD,@B0Y[
MQ<-)*6<*X/%0&[8UA$7!(6)$M)(1:]B$DPAPBMA2*M!5#FD9",I"0'LQ4G"L2)+AB#"[!TATE*QBQH4MP`0D-*")[email protected]#">S.5##O.ZHR0E*XV!BBZ'2M*%3_ZA:[email protected],3HEPA4XTH0%+""H87J8V+0BG.,=)SG*:\YSH3*M9YV:PQR*X"C6;"NEA1G.:V>!K$9HWB77EJPZ=/S<%^
M=)B##O)Y$B<$%&)[email protected]]05!0C1W4!0S5#&=V!X(BF`%]U9$H14?R3J!YM*3U
MC"A(-;/.FMKTICAM06UN4\U0V>$-E5.`.^'I47HJ%`4I4("L+I)!Q,CAD-VS
MP0>3=Y(TL,`-0FB>KA0X0((A"UZ_$E4>8'@1%=K`G:[BZ#OM"((>@``%9D6K
M'%[email protected]&&E+F(T&2D'!2C`:O/<[email protected]`$%TE376UN'&!6DX*D,7)<-
M4D#7'[email protected]!GD5TD74AP+7]``&RDH#A=B550\9+`TK6$%23](S6ET+!2*``1Y*
[email protected]`40B&!=D+W6%M+0AI+6L>:T(N.`[email protected]=!ZYK1"M:TH4WM:D<"
M7#H(5[:TM>VZ4"#;&US+!"#(+0AVV]OC5E>XQ#6NKCB[W-`ZM[0:BZYJ]\I1
M,P;WK;&=;6UO^U;NXL$,*@&O>,E+5R\[email protected]\XP*QY[6M=_*;WN.Q-PV?=RP/2
M:M4UTJ7O>?&+W?UN5[8P^&YXH3I>WE)65QN&[8-AF)MK42F,]!7IQR8(@J<8
M%[email protected]'M=JA!3[@WA4M&QBZ+A4E:[email protected]"G)+7A"`M\=F_+$/)O>X-536
MK5"V(Y`G5[FD)B#[email protected]"L`,9"&G(+5=:.N(&4A>79W$1:QY31U,1=\^:%:.
[email protected]@#PWO*'D&'DC27)_F[email protected]]<08-G'B;X=&8PM'')@'[email protected]*7ABC/.F,[>?*P`M8O.)[&SAB&-Z>Z1N=5P2#-AH=IF^@H:!:D=
M,ZO-#`)'6UK*S#,M7>%LZCF3%<\HV0SM#.(P-D#*#/:+5JC+\&?ZSN$.[VN=
MKA'M`[<9*VB4[A5].6>0&=9P"CK('P_3(`<7`=$,[email protected],#[email protected]%8I`.RK
M.0B#+IA`MBU)@0]\,(,4\.'>K,GWMB_-;W_+UB8"]P$-CEO5U](;66V][email protected]
M4`.Z7GQ;&G?)L<[email protected]!&,(2D%L4)K0OVTE+9!/"@*@>DH$,,YJHR0\<
MRP/#0"[email protected]!CH>[email protected]*"=%]<5&^[email protected]+5+9]CSH/A?ZSXO>U0+6`[email protected]";V"
M&*3YR:RF,NVLW.M'[email protected]`/;@4LOM_.YM06&KP`QRP([email protected]"B;OY)-^J-!GP8&_&SA6P"P\SF_MM:!`7
MG$(5GNS?W[IO-LL=\[email protected]%.+'E(!=[R"L]\F3?_=9\/@:X\J/#$1R_[5F]!
M!K67K>I9WRLQ%'8-C[_(N&GGQW,/(=W/$Z,3.1/%"U'1BJ,>B=\#[X?2=Z_?
M=$4]#&Z0?%%1WJTC00'L>\]!K]-U!8MU7;Z]OX49]+;\9'`]"%K0W_73'_CO
M%W]$,7^[QF;?0WPH87QE$`;(5VG+5VY'U`30%WV1TD53`!5GM"TZIGT$5&]F
MQW"TUWD;1WZ2=WZD5X#?%WQX`'[email protected]\`.LA8);('?IQG_L9WKWUWKV1G^<)WXX
M0%MIX,J&`[email protected]^``([email protected]'\FZ';FLBAT$'M!*((!UX('[email protected]#,'7IIGA\P'9M]X+<
MUGYI=G(9RD(:81W_?PX:*YX*R!8=R6&1_9XM07_VQWN,F&:Y%[email protected](.&_'IX1OQGSF!DA.2#2&A$C3MX%8V($8QWU<&'X^1X*]
M,GEJEG[^!X/NAW_Z1X/J)W]]^'OE=82A6&=[I0`I)@(?EVXE8#/FQXS.*%;0
MF%[K\G%@!2]Y,(C>[email protected]@*I(1QF`$)"8AF805,YE8[JB%-\MLW\Z)5([email protected]/*\4^O\3YI\"!IH`>_LRYE0`>:]#[email protected]$#7=$E)0^%!FHU+T
M>!([email protected]=8Z;,04D95(H%5$ZMQDH$9$$HI#K&)(B>4X[I0"[email protected]#N"(][email protected]`)&
M,RW3%6CS[email protected]%?9U$UW=FTP]U8S^9*QLH1PHB5H\B11DFY9IV2)[email protected]!IN`%LO
M,"KB81T_QU]/\`520`[email protected]`=OZ01([email protected]!D`64-5J8Q7:RM(0=RL"XB,`0T=R*"8Y;;9!`KQG:%!`)#H"930`67B9<1#PD\[@@(M$`.9
ME0!VUBLH\%-=YAI`[email protected]%A.A"OKHIE'4`14$`[email protected]*]5RS9.(C*URNO5C_W
[email protected]]FHFV(W`R6KZYYR5II,Q*[email protected],`/E1YU_Z%9C.)T94G_6
M&0,CIWW%*[email protected]\J6:OQIOEQYN91#[email protected]%[email protected]'X`1"(`4K
M=W?ZZ010$`12``)@XY[649_6XC584YP<@Z!Y4)]MJIR]0I]C\`5L\";%F:$A:B];8`54D`1-4`1=,&$EAQ(@^@4H:@5-\"0O*IZ5
MEII`E7;MX9JPV16R"0*:"2:WF9N\F5?*MYPHR(WCB87EB1):F9YH0B=&H',H
MP:.K^:.OF5BS.01&BIM.MIO6<6*_F8!1>A)3RI5M5:2V*::Z&9]EJJ0Q^F;L4VY!:A"XXINFDZ8C88R5\IG]]%:8%22J]DWH.)*4
M*I+MY$15Y`9%@C7XI!D)\"E6^$]PMJF"\CL&&37NV([email protected]`:`U$.J0()
M$`0MNJDH0#80U5+?ABY>]B_4:C;6"C3H4I.5@[email protected]*[email protected])N-9JZ(JZ:Y&W>&BWF
M&JQF4#9TT*U!DZY1J4D"XXK;A'[email protected]+3MJ_\ZJ&1Z2GCJC(L4SZ]4J*5
M0I4$F32#:K#KZC/[email protected]:45BGDUGPV!*<%:T*9Q*%J!J8W,'I#T$9!0`2/]Z]B
M!SEJYG3..1(JVSB5,Q(:%TLI>Z\NH&A5UK)X8+/T]:^*-K,7$7(OZZ\X&VQH
MY%;825](V(`82XI'-`7/][00^`6F^+,X>Z$="J8X0+*9^010(`0D:[(HB[7<
MI$DKNP8\:P0X<+-G&Q8R&XT:EV!N:Q`>J[-CQ[,_5[<2F[.21K-1Q[<>B[3)
M60./U[3'EK'F-D8<^Z]:&[)3T+5W-Z!MQ)^C1[E/0`1$,+9%<+*".ZYIR[-%
M4+0="[IQ"[BR1;=F:[?CBKD"[1_.[2!N[I]2[[email protected](!.>Q&*RT9%
MT+A9"[)N);*72Z!%8+EW5[*=6[8G$;.+IK8:9P.Z"[,XBW:HV[.QB[.N*[T\
MM[UO&[1R6[O/>[1O<#JFI6:&R[11JFH[]E;_BK!DT#)E<+$GX;#Q$[$>*[_T
M:[\CH;@P`0.G6+J:]+B`!P)"$,`[email protected][A::1<_=XJ."[('+`0.S%R'RKN/I[@R
M$$L"C!(13*$3O,%["Z48K)4TD&`=7+YO:\!N)00G/,(7G(1:F0,!#,'#"\(M
M3,,PS,`GL:UTP+&;E6*&62***0166$:J6K"FLS+SJ[[email protected]!ICD&2^LV=N,)G%
MAZCN"P(A%(7E=F1&^[;\J[!N5<&?6\`2W,)D[([email protected]*@"&O\59)
M=63_(EH,>Q'0>BR;^E91>UPI>93PA`)4VB1>206`#!4J2M$&XCH9)%%[email protected]%2/EXQ/T:MOH%ZVIF8T8WETK"NGR1F*+,A`P\AR\@5=\B5%
M\,:.9$RT7,NV?,NXG,L144E=(1`,J$4SHLME<1%GD1:E-$IO41*G]$8Q4!*!
MV!(O$1,T$`,;41,M,0.V%!C"#$F\E!B_Q!BA\QC;/,[D7!5.$`0MJF8)^01"
MH`2?X5;+]`8*,`1)Y%9G8"_[9RXC*@;S;`1Z>03OO']/L'^*4C4M8`;Y,B`M
M``>VJB,*(`1%4`10H&8B<``[email protected]`[email protected]`+GW*)TE6X9S<[N[&493<]T-=+^'`0`
M35MJ=,\S=(H`-$2G52S7,Y0'=52;4R\;!`O4"@[email protected]]A,=4[`4K&3!)LL[email protected]=ZP1)G;1MT!HQD`,YH!*G0Z3G\[email protected]@(`5O$(5LL"Y$\!IKP,6M`Q4/XP;K<@0?0G,1JI`S
MY#?RM"_8MJ1JN$5%\[email protected]`XRZ\8ZKJX]D<)[email protected],76Y%[1AED-AG,#7`BFU.01"[email protected]]O
M0"JF83-.,2U\\D]A(`;=32AE"[email protected],,MCLTA2%[email protected]_"#1DMH9^[email protected]
M8=[6$]YTP">IK=Z"W1J.^=[`G2W;@AI!J2T+XCD%WBD'"0?Q711S_%`2CH97
M,]S%/2"L$9#XK:O3HJO*X2`#8I!4L-BTT]B/C=N.:3K*T14V\QI\8EB.DHC:
MM(L4GI#1F5"RI45_#T3:BW>!NX`(7<>(ODN*.#=G<\LO;M5=[8+>)][email protected]?O4]XR\CY-`SI'LRU-3A2BDUS*#0?,[email protected]<1%)T!I%[3;>O2I\
M(@(IKBTBH#3O$^[email protected],]\U+))2!P,*(YT]O4_=F:[email protected]`1/T`1!\!E2\`1%
MP`1,`"=)(+5ZR:*?$02P#@)70*!2$`1.L*)%\!E7$"5(T-A%<`0$NKE4,-"G
M/NN<02=4L.MMQ`)\\[email protected]$\"1'``*X'NLL"@4V702;*^V]_NM)[email protected]">1
[email protected],`0H]^M!(`1)8--4D`7M?NJIONJ-[>JY#[email protected]/[email protected]`1/4*"^;N]3``5%,`1)@.O^CI=)T$9#8/`7
MSR>;"2914`4%[email protected]+H?O'W;N\PK^JU*>PH1P53X.I6P.[&/@55P`0K
[email protected];;@2MWD5N^[email protected]@0NTB!?.WF>X%"@6MGN\\[email protected]`2VR?4%^N^^SNTE
MGP1JDNY&8)[email protected]]7O+KLN]'8-.UZ01,0O7I?NKL;NH73P55\`1.+_8KJB;K
M0J"[email protected]?_8T_^^;N>]C7_;#[email protected]/1N=/(BQ?%28/%2H.IR7_EG
MW^R?T?-&$.W(2ZR;$03$)L``1>$^:OP3;]DP9KX`+\
MU`)M,&>(DQHY2VVE+E([email protected]`0+>)CI]@)[email protected]`7-+!#OHR!04>-I.7DO,"V8
M\@+M/@57_2)932-HP`)V4!,N0!>'_0+1#P-=JTHZ`!(Z(`,Y$%YOT!0EA0=7
M1`(E:8Y3_B+3\SZN:<05CMRG(=`%(F-"KP($&`$
MS,BJB$(3Z38TT,
M+1&M"SOP()(*(FH-,9`FF:0>00.%AE+9*QZ0=K0!%?A3ZLB,@"W8K00P`96`
M`DH`::ITZ48'2+I*EYIJH&(B`U0P![(>5=-33"#E<`,%L`P0P1[E`HG%'[email protected]
M:[email protected]&UD`%<`-K52*B)JJO-?1`<#[email protected]`<&()?8`6B"WUQ!$5`GPD#?F$'),$K
MP`2=8`J`@D5G"KZ5'[email protected]"KF`61$-;<`3JBU!Q`L%@"LR#++`,[email protected][email protected]`9EX`I<
[email protected]!P8)4C8'?V`=_`)"<`P:P3*`!(E%"1`"A?`)4ATI&`:H("-TA*0'$LHR
[email protected]"K4DE#(,J\'`LB`:1,TX2`>R$:[email protected]$WJ`.]F#!D$%@BQO6(S^1F["`>
MU(.K$+:\!B48"P]AT;%#(J`'U$*\,Y`[email protected]$7TB-H8HJ"$YXH63T`NB
M0"P1#)$%-"R&K&$'GD$MJ`S1$#[email protected]%'A)#F]$&W>$I3(68<`\&B#[X!P/A
M-ER";\400D&F$P['X9(IA]][email protected]*0X[;',ZD`OV0DKX!8%A'GR&`W%7
M[$!.R`]#(1H"B*M#($;#G6$0+X(U5(5\D!6*@&T("R.B+`1B%Z$B4L'/9`XU
M(CI$ANMP(#4J=_B-N)ART!,>$`2J0#ZB163(,)R)[email protected]`&=*1U:\#<@$
M/50(]F`))5**`$>J([email protected],:&NJ9/S$-;EH%>B:+UD,5R&8B#/"R!AU65CL#M\9JY(P2A2-T!*+X%S_B#[email protected],[$BF63)X8E8W2
[email protected]>J?-`R!8I`ZBFA$*)C,]#6V5]60-E*OL"@.`Z-<"XT24;B;05D5
MPM%8*0=DE5*,(TIH*08%IBB4B>0:N^,Z*4E=,!/&QGF"R9"**8R)I(S)G(6>
MQ%6LRG.)+Q)CWNBK:R16/&,)_!=)H64=&Z;40M(*F%E*(^[email protected]@*@TMG*
M+)6BK\R(_O(>DU-A.2Q>[email protected]!=D,<"5=;%@M`#9MYE)A&J1\22JW)C]6F%=#N"A-,NHPVF7-Z!83$R220FJ!5#>"#70Q9$28'$QQ
MR2SY<8P%B1>[email protected]*&1[W"S0++V0B(M#'3!,*HE*\$8(IE=^(M_N0%G3\"0
M&`+CFS;#M*DQR0_')!MYXQIX3).:.XK'BP4:)4-H:L^K25L>\DW&+2]#?_H0
M'#@S*\`ZMM43(\!((:B'Y$17UB#7P:CA(]:`PKXHJEQD\^GD#Y:V:/JVE2>G)1/IY;DVNV
MT:,[email protected]+*.'[email protected]'230VFV<3;3BE!,D8U>9^63=M\VK8%;B1DE+J
M/&DL=*-NQ`B[<3=!Y`I5'0^DA1*.OH%!_>;?#!W,,G`*SL$9/,$H6SX<;BEQ
M#-6M^3AT)30>%#6CHWJ%4ZPO'F4T*J>P]%:V3\CI.-/R%?F<[E1I]`33.3DI
M9^6H#Y?C/AY6I'`B,Z?F5`Y\!!2;CO:*.A`S\5"=HT/M=%S1ZL`ZO4+K;,C=TG48Y=EI'%7F5M18:RL!V_<\`$C[5D,H9'!;4AQN-X_J4&K$4FZ!<-H$6$+540#/@\G4CT\"'Z
MM"M0'.%@?RX1]/[email protected]'=IVG*(A!#BU""+4(_;B<7=M6M2+^@\P"D(!"'$2H,')DS[1FJR7^BI=EHMU*20KS;M4E_*R&U4:'I8LU6;5
[email protected]*A,@[email protected]M'[email protected]?#]H_O[-2"J..MCLIIQ$:G8CJ3/FNM=F$4B<=T1=3J`KM$=>9+Y,3
MPE%$Q!/U>"&ZPGBZ$]*\16_E!JHALNF(CFM\2)#E#[CX/JDG(T(];0A2(0')%'\I$.6B%)JGNZ1B<8F`.5$%49_0D^T63JG
M9]4R1:GSD,R!1"(M[V;>E)U"2/[email protected]+GX^2=6+-Q`A;!J3P!D*$BG1?A
ME.V)9#[email protected]/3J+C0Q_R369)1G9A#11)&=AAJ!"NYQWCI-^O"^XP7KY-$
M4=!X`4!#LD?"%BP)1+.+$-4O1/1*9LDBRD".:&_Y,'@`2QH:)5J=&J&C"J-;
ME(RZT"CJ6O#+T2EJ)6`.9%'0:3-MDHI5*FM*\$0>K92+8(]4PX+41VZ`'[email protected]$3%!!)!$G/!HV"!-H<"JZPC1#H6
M'ZGX5:2+=%*25:SJ%1[)]!'253I.2M))"F2(=)5=I1C#[email protected](-TE<4*?O
MH2OD!JRH&W]#/A6E`-F4_*-2"CS\L0PX)0/)NL;`L<%[email protected];@*F9.`
M$E0`+(5,L62GR))90DMJB3>UI77QEN+2%9`"=ND+X*[email protected],`4Q
M*3K$$)-B8DQNH,65I2W7+2;3'=(3*4\SJ8G-]2T\DU(,3:.I-,G0],,"Z0J7
MBEA#BC;EJ=Q$IGJ3**J?W$8XB3&"I4!G#^[email protected]?2>]<
MI^STG;:3&&*I.RD3=[email protected]/_PD$!"C*9:`JZ+A(4"%J0>$G!W4P3!2-(EZP4FO-)[email protected]`JZ;H(,RH,
MFE09Q55KU(IJ42^*N5C5&G6CG$".*C]RRJ'2*8AZI\#41'4R24I>W)F."FO^
M*$UM0#:5*Y(B:[email protected]&)<[email protected]:Y^J3!%4?F4196>Y(F"JBEEN95:V2M[4SNJ
MH2)6(!4A\=21&E..U4R=5-\5J/Y*8GFAT?.D9K&/Z,9P)/^;5*R4D%ZJKA"M
M&(0B][email protected]*K^A-TW%($I5HI!O$VPL$F)H[+R5+2*H3F+HI"KR)LG^U4Y
[email protected]`,*]`F'4M**:V.RRJVRM:(L<=XZQF85E9"7M$K;,5VM%5XK584JUT!Q']5
[email protected]@I7.,M8HHN%!:_$:WJM5_#U;>4KV#FP7M:_"ECTU;_BK#!&+=Z5Z+P(
MT!1^Q=?[BJVR$M5:EL)KA6%5X[6\R%[email protected]?Y+!763#<-*PX*O#=B\0
M>V'%5_;:6[9K<)VOYI&[#EK"+2VIAJ=]EM210BAU7CZMX12ZOM>S"%N?R
M7#<6;:',O*5QV-:%Q5ZT*W6%V+XU8HE6B9U=>FO)JECTE;0"D8LM83`V`D78
M#IMC+=..G5S(2WG-$`*5N397A6U>1W;(OJX#,[K2+,O,7JJ+PS)9V;&S8->3
M?;-)5F->6-REM*ZL#,NR<"EX!=F/552][,@"LY5+;)U9("MG/5;HFE[5ZXMU
M6"[email protected]=MD_U>=_;#.ME!RV>M+/NBH%GLUL0O)M:_'*RPK$KZZV"1VO+AOVXJ
M"*AA.>F&X:<)IL`B:TTUG
MP%#!6FY>&+&5M'V+A2$P9ZML$97BTF'/EH`1VF.+P+!MM96L]16B`,7!U$91
M0#U-3+;%B!%$#"1MQ39:''O&?&U+>%FK]:">I4?U#N-8&)ACYM&.][email protected]^;)(F3.],]VUCZ&`J'5[
[email protected];2&#S)H>LD3VCE392B*L3N"X8#M]Y(=%6TD147`&:A_6'`0BM=:NNU;VZ6#?K:MVMRW6[KM?]NF`W[(K=L4MVRZ[9/;MH-^VJW;7+=MNNVWV[
M<#?NRMVY2W?KKMV][email protected][^K=O`NOX3V\B#?Q*M[%
[email protected];K^-]O)`W\DK>R4MY*Z_EO;QH%Z[?
`
end


[next article]
From: [email protected] (Jim Burns)

Subject: Re: dvips/dvilj2p .. need some help
Message-ID: <[email protected]>
Date: 9 May 92 17:40:24 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 13

in article <[email protected]>, [email protected] (Brian Cho
jnowski) says:

> Also as a note, has anyone managed to 'tex' the a386 (gas docs) without
> getting a stream of errors?

If you mean a386.texinfo in gas-docs.tar.Z from tsx-11's pub/linux/info
directory, that is a subset of the full doc gas.texinfo in the same tar
file, and the latter tex's just fine.
--
BURNS,JIM (returned student)
Georgia Institute of Technology, 30178 Georgia Tech Station,
Atlanta Georgia, 30332 | Internet: [email protected]
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt0178a


[next article]
From: [email protected] (Michael Kraehe)

Subject: Re: Logitech Mouse and X11
Message-ID:
Date: 9 May 92 14:41:00 GMT
References:
Organization: MausNet
Lines: 9
X-Gateway: MausGate/News 1.05

> Now I would gladly give up this functionality to have my Logitech mouse work
> with X. I download the binaries, and EVERYTHING works, except for the MOUSE.
Hey man I saw your artikel getting glance on my eys ๐Ÿ™‚

How can I get an X11 for Linux 0.95c, perhaps patches for the MIT-Tape
or the Thomas Roell-Server. And can somebody tell me witch SVGA an Mouse
I should use ๐Ÿ™‚

Tschaui Michael


[next article]
From: [email protected] (Mark Holden)

Subject: Re: X Problems..
Message-ID: <[email protected]>
Date: 9 May 92 19:43:19 GMT
References: lsinki.FI>
Organization: Saint Cloud State University
Lines: 35
Nntp-Posting-Host: 134.29.38.2

In article , [email protected]
FI (Mika Pekka Liljeberg) writes:
|> sure I did it right, too. I can easily get a stable picture, but instead
|> of a single beautiful one I get two ugly ones, side by side. Both contain
|> the visible desktop, only squeezed in half along the X axis. I suspect
|> it's the Sierra DAC chips that cause this behaviour. Can anyone confirm

I doubt it's the Sierra DAC, the same thing has happened to me with Diamond
Speedstar w/o Sierra DAC, when I had the incorrect configuration in Xconfig.
With the Diamond Speedstar+ and the Gateway 2000 Crystal Scan monitor, it's
a simple process to set up X386, and I'd done it before under Mach anyhow ๐Ÿ™‚
The important thing is to get the correct clocks for your card, the ones that
are included in the Xconfig probably aren't right for yours. To see what
they are (as far as X386 is concerned) take the "clocks" linne out of Xconfig,
and run xinit, preferably from a login on one of the serial ports (I used my
Linux machine to set up my Mach machine, then reversed the process later ๐Ÿ™‚
then make sure you have a line in modedb that corresponds to one of the clocks
that X386 shows (probably the last one).

|> And then there is the mousie, of course. ;-( I've got a Genius mouse,
|> which can act either as a Microsoft mouse or a MouseSystems mouse.
|> It works, too. Not under X386, though. In Microsoft mode the mouse
|> cursor gets whopping mad, moving in quick jerks in the same general
|> direction I'm pushing the mouse. I've got no control whatsoever over
|> the buttons. X seems to think someone's bouncing on them, though. ๐Ÿ˜‰
|> In MouseSystems mode I get no response at all. Is the X mouse driver
|> picky, or what? ๐Ÿ˜‰

I've run into severe wierdness with mice under x386. On Mach, my
Microsoft Mouse works _almost_ correctly, except that it doesn't recognise
a button press until you move the mouse, and then doesn't notice that the
button is being held down anyhow. The same behavior exists under Linux,
except that it does recognise that the button is being held down after you
move the mouse. So, I can use my mouse under Linux, but I had to punt back
to X11R4 under Mach . . . Also, I've tried a Logitech Mouseman, with
absolutely no luck whatsoever.


[next article]
From: [email protected] (Eric Youngdale)

Subject: Problem with tar reader in rootimage 0.95a
Message-ID: <[email protected]>
Date: 9 May 92 18:42:29 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 33

Linux Activists:

I just installed linux on my 486-33 system over the past
couple of days, and first of all I wanted to congradulate you all.
I had been planning to wait for hurd to reach a state where it was useful,
but linux fills the bill quite nicely. I must admit that my experience
in managing a SVr4 unix machine came in *very* handy during installation.

There are a couple of rough edges that could be polished off. To
make it easier for people who work on these things, I will make one point here,
and save the rest for other mail messages.

First of all the tar reader in the rootimage-0.95a apparently has a bug
in it. I am sorry that I do not have enough utilities going to be able
to come up with a patch, but it appears that a patch is not needed (I will
explain later). I noticed the problem when trying to add the 2.1-lib.tar.Z
and 2.1-misc.tar.Z distributions to my system. Having used the command

tar xZf /dev/PS0

to unpack, and then having done the installation, I found that the gcc
image (i.e. the compiler driver itself) was adding a unparsable option
to the list of options for ld. On the screen it looked almost line
four dashes, except that the second one was a little box shaped
creature. The cc1 image was also corrupt, because it would report a
syntax error for the simplest programs, complaining about the '(' in
a function definiition or prototype. I tried making a second compiler disk and
installing it, and no dice. It was not until I tried using the tar on the
utils.tar.Z distribution that I was able to unpack a compiler that would work
for me.


-Eric


[next article]
From: [email protected] (Eric Youngdale)

Subject: Problem building kernel 0.95c+ (and pre-0.96?)
Message-ID: <[email protected]>
Date: 9 May 92 18:47:42 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 23

Linux Activists:

The file tools/build in linux-0.95c.tar.Z distribution will not compile
because of the line:

#include

Basically the compiler cannot find linux/fs.h because there was no -Iinclude
specified for this file in the Makefile. I tried adding a -Iinclude, which
helped a little, but this created other problems because some of the data types
in linux/fs.h were declared somewhere else. To fix this, I ended up adding
the lines:


#define MAJOR(a) (((unsigned)(a))>>8)
#define MINOR(a) ((a)&0xff)

to build.c and removing the above include in order to get the thing to fly.

I looked at the sources to pre-0.96, and it looked like situation
had not changed.

-Eric


[next article]
From: [email protected] (thomas dunbar)

Subject: .96 testimage and Emacs
Message-ID: <[email protected]>
Date: 9 May 92 18:51:08 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 20

I'm having problems show up more and more with Emacs:
abbrevs L12 - linux v 0.12
L95 - linux v 0.95c+
L96 - linux v 0.96 testimage
E57 - GNU Emacs 18.57
E58 - " 18.58
Enu " " newemacs with fixes wrt gcc 2.1 lib

when i was using L12 and E57, there were no problems. nor have i had
any trouble with the L95 and E57 combo. however with L95, i'd be using
E58 and i'd get a login: prompt as i was editing a file..very confusing
(to me anyway) and so i went back to E57. When Enu was posted, i tried
it with even less success: it loaded and displayed the first line of
it's default info msg and then locked the computer.
Just now i installed L96 and find that E57 now give the login: problem.
Sigh. micro-emacs works ok but it doesnt recognize some of the keymappings
i'd been using and in fact it doesnt recognize Quote (Ctrl-q) at all.
Unfortunately, i dont have disk space in Linux to recompile/debug Emacs.
Not that i have the skill to find the problem anyway. Are other GNU Emacs
users having these problems?


[next article]
From: [email protected] (Eric Youngdale)

Subject: Lack of send_break for serial I/O
Message-ID: <[email protected]>
Date: 9 May 92 18:53:31 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 15

Linux Activists:

I was having trouble with Kermit, because the hangup command was not
working, and I am guessing that the underlying problem is that the function
send_break in linux/kerel/chr_dev/tty_ioctl.c is essentially a null function.

Is there a good reason that this has not yet been implemented? Are
there kernel patches to implement this? Is someone working on this? Finally,
if the answer to the above questions is "no", would you be interested in
having me do the job?

-Eric

(I will be out of town all next week, so I would not be able to get to it until
after that).


[next article]
From: [email protected] (Michael Pereckas)

Subject: Re: Some questions...(long)
Message-ID:
Date: 9 May 92 19:10:00 GMT
References: <[email protected]>
Sender: [email protected] (News)
Organization: University of Illinois at Urbana
Lines: 24

[email protected] (Oren Kastner) writes:

> It all started a couple of nights ago when I was trying to move some
>of my files to floppies (some tar.Zs) so I'd have more room for 0.96 when
>it comes out next week. I mounted my /dev/floppy0 (5.25, 1.2M -- A: under
>MS-DOS) disk drive on one of my mount points. I moved some of the files
>over and things were cool... 'df' reported the hardisk info and the mounted
>floppy info correctly. Next, I umounted floppy0, replaced it with another
>disk and re-mounted it on the same mount point. I did an 'ls' on the newly
>mounted disk, and to my surprise, the stats for the FIRST disk were given
>back to me. I tried 'df' and again, the stats for the FIRST disk were given
>to me. I was baffled, to say the least, but, I thought maybe I'd write

Happens to me all the time. Does anyone dissagree with the following
statement?

"It is impossible to have Linux filesystems on floppy disks and keep
them from being badly corrupted."

This is with kernels earlier than 0.96 (which is not yet released).
If you dissagree, what do you know that I don't?


Michael Pereckas


[next article]
From: [email protected] (Al Clark)

Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <[email protected]>
Date: 9 May 92 18:54:25 GMT
References: <[email protected]> <[email protected]
cs.jhu.edu> <[email protected]>
Organization: Netcom - Online Communication Services (408 241-9760 guest)
Lines: 38

In article <[email protected]> [email protected]
(Britt Park) writes:
>
> I think that a few comments about the GNU license are necessary. I do
>not believe that the portion of the GNU license which states that modifications
>to GNU software must fall under the GNU license has any legal basis. The
>rights to the modifications to a program invest wholly with the author of those
>modifications. As to the requirement that complete sources to a binary on an
>archive site must be located on the same archive site; I don't believe this
>has any basis either, at least not as the GNU license is currently phrased:
>
> "You may copy and distibute the Program ... in object code or
>executable form under the terms of Paragraphs 1 and 2 above provided that you
>also do one of the following:
> ... Accompany it with the information you received as to where the
>corresponding source code may be obtained."
>
>
> Certainly I believe it's better for archive sites to make available
>complete sources rather than just diffs. This is not, however, always
>practicable. It costs a great deal in resources to maintain an archive site,
>and GNU software, as we all know, weighs in on the hefty side. I'd prefer to
>see archive sites with binaries only rather than archive sites which fold
>because they don't have the space to hold all the sources.
>
> Britt Park
> [email protected]

Not being a lawyer, I have no idea whether or not the above has any validity.
However, I would suggest that people be very careful about following it.
FSF and rms have a lot invested, and I suspect will be militant about
enforcing their copyright (copyleft). I have been told that one of the rights
protected by copyright is the right to modify, thus modifying a copyrighted
file without permission is in itself a copyright violation.

Be careful when you get into this legal stuff.
--
Al - [email protected] - My opinions are my own.
*** Practice random kindness and senseless acts of beauty! ***


[next article]
From: [email protected] (Paul Fink)

Subject: Tape drives ?
Message-ID:
Date: 9 May 92 19:34:37 GMT
Organization: Thinking Machines Corporation, Cambridge MA, USA
Lines: 9
NNTP-Posting-Host: custard.think.com

In looking through the FAQ's and other docs it appears that there is not
support for any tapedrives, yet.

True?


--
----------------------
Paul Fink [email protected] 617 234-4150


[next article]
From: [email protected] (Eric Youngdale)

Subject: Re: send_break
Message-ID: <[email protected]>
Date: 9 May 92 19:52:54 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 16

Linux Activists:

| I was having trouble with Kermit, because the hangup command was not
|working, and I am guessing that the underlying problem is that the function
|send_break in linux/kerel/chr_dev/tty_ioctl.c is essentially a null function.

I realized that the above statement was far too vague. The problem
that I am having is that the DTR line is not dropped, and as a result the modem
does not hang up the phone. The only way that I can get the modem to hang up
the phone is to power-cycle the modem, which is a real pain.

I saw that Linus had commented in comp.os.linux that 0.96 will drop DTR
when the serial line is closed. If I am reading this correctly, then all I
need to do is to wait until 0.96 comes out.

-Eric


[next article]
From: [email protected] (Brian D Chase)

Subject: Where can you find X for linux???
Message-ID: <[email protected]>
Date: 9 May 92 20:03:08 GMT
Sender: [email protected] (The News Administrator)
Reply-To: [email protected] (Brian D Chase)
Organization: Rose-Hulman Institute of Technology
Lines: 10
Nntp-Posting-Host: o231-14.nextwork.rose-hulman.edu


I've been seeing a whole lot of postings on X for linux yet I haven't seen
anybody mention yet where to get it. Soooooo I thought that I'd just ask:
where does one find the elusive X for linux?

------------------------------------------------------------------
Brian D. Chase Internet: [email protected]
(they call him Sid) ph: (812)877-8812
Rose-Hulman IT, Box#258, 5500 Wabash Ave, Terre Haute, IN 47803
=================[ bit a byte O' bitter batter ]==================


[next article]
From: [email protected] (Ari Lemmke)

Subject: Re: MGR for Linux
Message-ID:
Date: 10 May 92 02:24:20 GMT
References: <[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
Sender: [email protected] (Usenet pseudouser id)
Organization: Helsinki University of Technology, Finland
Lines: 10
In-Reply-To: [email protected]'s message of Wed, 6 May 1992 00:58:0
0 GMT
Nntp-Posting-Host: deathstar.cs.hut.fi


In article <[email protected]> [email protected]
nt.edu (Jim Winstead Jr.) writes:
: On a related note, I've created a MGR channel on Mail-Net. To join,
: send mail to [email protected] with 'X-Mn-Admin: MGR' on
^^^^^^^
Should be 'X-Mn-Admin: join MGR' ...

: + Jim Winstead Jr. (CSci '95)

arl


[next article]
From: [email protected] (Eric Youngdale)

Subject: Re: GCC 2.1, probably FAQ, but . . .
Keywords: gcc 2.1
Message-ID: <[email protected]>
Date: 9 May 92 20:00:41 GMT
References: <[email protected]> <[email protected]
itoba.ca>
Sender: [email protected]
Organization: Naval Research Laboratory
Lines: 24


>In article <[email protected]>, [email protected] (
Budi Rahardjo) writes:
>> [email protected] (Networking Research) writes:
>> :
>> : Ok, I know I've seen references to this in the past, generally
>> : as the "infamous 'parse error before ('." Well, anyhow, I'm trying to get
>> : Gcc 2.1 to work on my Linux machine (0.95c+) and I'm getting all sorts of
>> : 'parse error before (' messages, in stdio.h, in code, everywhere. I've chec
ked
>> : the ftp sites for info and looked back in comp.os.linux as far as my nntp
>> : server will allow, and I haven't been able to find the solution. Anybody
>> : want to help me out here?

Here is the solution to the problem. The tar reader has a bug in it. G
et the tar reader
that comes with utils.tar.Z, and then use this tar reader to unpack gcc 2.1. I
went throught
the same thing, and once I changed tar readers, everything worked like a charm.

The tar reader that is corrupt is the one that originated on the rootima
ge-0.95a distribution.

-Eric
--
Eric Youngdale
Code 6551
Naval Research lab
Washington DC 20375


[next article]
From: [email protected] (Britt Park)

Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <[email protected]>
Date: 9 May 92 20:23:07 GMT
References: <[email protected]> <[email protected]
cs.jhu.edu> <[email protected]> <[email protected]
om>
Sender: [email protected] (Mr News)
Organization: DSG, Stanford University, CA 94305, USA
Lines: 29



Perhaps I should clarify a little. I am not an attorney either. I do
live with one, however, and do know the rudiments of intellectual property law.
As to my first point in the original posting: to distribute or sell modified
copyrighted software is certainly a copyright violation. I do not believe that
selling or distributing the modifications or a program which performs the
modifications is. If through some Herculean effort I were to figure out what
bytes had to be changed in MS-Windows, for instance, to change it in some
pleasing way, there is no copyright violation in doing so or in distributing a
program which would automatically make the changes. In the old days there were
numerous patches of this type to Wordstar under CPM. As far as I know none of
these were illegal.

As to my second point, the GNU Copyright states clearly that one is
allowed to distribute diffs or binaries as long as it is made clear where the
sources the sources may be obtained. The author of the diffs or the compiler
of the binaries is obliged to make the sources available. An archive site is
obliged to make instructions as to where to find that source available.
(Paragraph 4 of the GNU GENERAL PUBLIC LICENSE).

I must admit that I'm puzzled by R.S.'s statement. It seems to be
against the stated purpose of the FSF, the stamping out of Software Hoarding.
Making burdensome demands of overtaxed bulletin boards and archive sites makes
the propagation of Free Software more difficult not less.


Britt Park
[email protected]


[next article]
From: [email protected] (Eric Youngdale)

Subject: Re: Trident (SVGA cards)
Message-ID: <[email protected]ra.nrl.navy.mil>
Date: 9 May 92 20:21:51 GMT
References: <[email protected]>
Sender: [email protected]
Organization: Naval Research Laboratory
Lines: 12


> working for this card, or are we all doomed $180 for the ET4000

For what it is worth, last weekend I saw ET4000 boards for 89$. This wa
s at one of those
computer flea-markets where the low-price vendors all show up and sell their war
es. If I had known at the time that X was out for Linux, I might have even pick
ed one up.

-Eric
--
Eric Youngdale
Code 6551
Naval Research lab
Washington DC 20375


[next article]
From: [email protected] (Magnus Alvestad)

Subject: Subgroups?
Message-ID:
Date: 9 May 92 20:38:33 GMT
Sender: [email protected] (Bergen University Newsaccount)
Reply-To: [email protected]
Organization: AlveSoft
Lines: 18

Would it be an idea to create a few subgroups to comp.os.linux so as
to separate posts concerning X11, mgr (might be some volume in a few
weeks) and maybe some other topic I can't think of? SCSI?

Btw: I am coordinating effort on making screen-card drivers for X386, mail
[email protected] if you are making a driver for a new card,
want to help making a driver, want to know if someone else is working
on this or that card.

-Magnus





--
| Magnus Alvestad | I am a prolog program. I get all my opinions
| [email protected] | from a small organization in Birmingham.


[next article]
From: [email protected] (Steve Robbins)

Subject: Smail query
Message-ID: <[email protected]>
Date: 8 May 92 19:44:57 GMT
Organization: The chiral cage of chaos
Lines: 35

I'm posting this because I'd also like to know the real answer. I hope this
doesn't get lost in the X-tasy flooding comp.os.linux these days...

From [email protected] comp.os.linux <[email protected]>

I'm having a heck of a time guessing about this...
Can someone tell me how to set up mail from scratch
on the 0.95+ interim release?

"smail" assumes that I've already *got* "mail", but I *don't*!
I'd be tickled pink if I could just get _local_ mail working...

As I read the smail documentation, you *don't* need anything other than
smail. Unless, of course, you want to read the mail using anything more
sophisticated than "more /usr/spool/mail/steve" ๐Ÿ™‚

All I have is smail to do the delivering, and elm to do the reading.
However, I keep hearing references to 'lmail' (or procmail) as being required,
which puzzles me. As I read the documentation, you can set up smail to use
an 'lmail' program for delivering local mail if you already have a local
mailer that you wish to retain for some reason, but it isn't necessary.

As I said, I've compiled smail and it works. I had major hair-pulling
sessions with building smail, but I'm told this is due to a bad awk program.
Smail also gives GP 0000 errors, but I'm virtually certain they come from
a bad libc.a (due to bsearch() and fseek() at least). We'll see.

Please help... Don't be worried that your reply will be one
of many, 'cause I didn't get *ANY* responses to my last request.

Thanks in advance,

Rick Miller [email protected] [email protected] [email protected]

Steve Z


[next article]
From: V5068U%[email protected] (jim wiegand)

Subject: burst (mail de-digestifier) - anyone?
Message-ID: <[email protected]>
Date: 9 May 92 21:07:59 GMT
Sender: [email protected] (Mr Background)
Reply-To: V5068U%[email protected]
Organization: The Internet
Lines: 7

Hi All:
Does anyone have (or know of) a PD source for burst? I use a VM mailer
that gloms everything into one big file and I really need to sort my mail out.
Any pointers would be most appreciated.

PS. X?? emacs?? TeX?? AWESOME WORK ALL! Dead Operating System will soon
be GONE!!


[next article]
From: V5068U%[email protected] (jim wiegand)

Subject: MATLAB Unix Port - Does it Exist?
Message-ID: <[email protected]>
Date: 9 May 92 21:18:03 GMT
Sender: [email protected] (Mr Background)
Reply-To: V5068U%[email protected]
Organization: The Internet
Lines: 12

Hi All:
Another call to help me not reinvent the wheel. I know there is an old
PD version of MATLAB on the net for NeXt machines that uses the original
pre-commercial sources from 1982. Is there any newer version around?
I have the old sources but from what I can see it is really a primitive
beast.
This is the last $%$! piece of software that keeps me using DOS. If I
have to, I will rewrite and update the PD Matlab and release it ...
Yes, Yet Another Freeware Clone!
Any pointers most welcome!

jim [email protected]


[next article]
From: [email protected] (Ben Combee)

Subject: Problems compiling UMoria 5.4 (MCC .95c+, GCC 2.1)
Message-ID: <[email protected]>
Date: 9 May 92 21:44:19 GMT
Organization: Satellite of Love, Gizmonic Institute
Lines: 15

I've recently tried to post UMoria 5.4 to Linux. It compiles fine,
except that I had to comment out references to SIGBUS and SIGSYS in
the signal handlers, and change the signal handling routines from type
int to type void so I wouldn't get GCC warnings. Still, after getting
it all linked and installed, when run, I get a GP error 0000. Are there
known bugs in the MCC shared or static libraries that might cause this
to happen? Thanks.

(BTW, the Curses routines appear to work pretty well. I compiled a few
test apps, and they ran nicely.)
--
Benjamin L. Combee | -===-
Internet: [email protected] | = ___ "Home of the Big G Burger"
1991-1992 joint enrollee | - = -- MST3K, Comedy Central
C, Modula-2, C++ resource | -===-


[next article]
From: [email protected] (Adam Goldberg)

Subject: Re: X and 9600 bps (ANSWERED!)
Message-ID:
Date: 9 May 92 21:42:10 GMT
References: <[email protected]> <[email protected]
du> <[email protected]>
Sender: [email protected] (USENET News System)
Organization: Iowa State University, Ames IA
Lines: 89
Originator: [email protected]

In article <[email protected]> [email protected] (Joel M. Hoffman)
writes:
>[Max 4800 baud]

In article [email protected] (Al Clark) writes:
>I have been using kermit with a Robotics Sporster 9600, using 19200. The only
>possibly unusual issue is that I've got a 16550 UART. I get the normal
>9600plus display update, with no observed errors. File transfers rarely show
>an error.

In <[email protected]> [email protected] (Joel M. Hoffman) writes
:
>Perhaps the problem is that I'm using a direct serial connection? Is
>it possible that the modems are doing some sort of buffering that Linux
>needs? I know the serial ports can go up to 116,000-ish, because
>Laplink does it. Why does Linux give up at 9600?

Even though I may sound like I have the definitive answer, please don't
take the following as gospel (though I believe it to be basically
true):

Here's the story. Joel is able to get only 4800 baud, using a direct
serial connection between two computers. Since he didn't mention
otherwise, he probably doesn't have 16550 UARTS. [For those not in the
know, most UARTS are either 8250s or 16450s. They are functionally
identical, though different internally. These are just plain-jane
regular UARTS. A 16550 is upwardly compatible from an 8250 or 16450
(just as a 386 is upwardly compatible with a 286), but it has 16-byte
FIFOs for both input and output.]

Al, on the other hand, is able to get a 19,200 connection. He has a
16550. Bingo.

Each time a character arrives at the serial port (be it from a modem,
direct connection or whatever), a 8250 (or 16450, or 16550 with the
FIFOs disabled (the power-up condition)) generates an interrupt. At
4800 baud, this is approximately 500 times each second. At 9600 it
would be 1000 times/second. On some processors, this wouldn't be a
problem, but the x86 line is fairly slow in acting on an interrupt (due
to the segemented architecture, or whatever). Plus, since Linux is a
protected-mode OS, there is overhead involved in task-switching to the
interrupt handler (device driver). Add to this the fact that Linux
doesn't always enable interrupts as soon as it can (NOTE: Linus has
said that 0.96 will be a little beter in this regard), and you get
problems trying to receive 1000 characters/second.

But! (Here's the good news)

A 16550, with it's 16-byte FIFOs can be programmed to interrupt every x
characters received (where 1<=x<=16). Say the device driver tells the
16550 to interrupt every 8 characters (a good choice since it allows
room in the FIFO for a couple more characters in case the device driver
is slow responding to the interrupt). When the interrupt does occur,
the device driver just grabs whatever is available. Since an interrupt
is generated every 8 characters instead of every 1, at 4800 baud, you
get (appx) 62 interrupts (instead of 500). Or, at 19,200 you get
apprx 370 (which is still less than 500, but each interupt takes
a little longer to deal with, so is perhaps as time-consuming as
the 500 1-character interrupts).

What a deal, eh? You get 19,200 with the processor overhead of 4800.
What's the catch, you ask?

Here's the catches.

1) A 16550 costs between US$15-25. Some motherboards have serial ports
'built-in' (ie, a 8250 on the motherboard) (ugh), or in worse cases,
consolidated in a VLSI chip. In this case, you need to disable the
motherboard serial port, and buy a $20 serial board (see below).

2) Some IO boards which have serial ports on them have the 8250 or
16450 soldered (not socketed) on the board. Make sure you aren't
buying one of these. (I got a 2 serial port (both socketed), 1 Game
Port, 1 parallel port card for $15...look in computer shopper). Most
boards don't come with a 16550 (though some do--ask), it's no big deal
to swap the 16550 instead of the 8250/16450 (they're pin-compatible) ,
though you may have a hard time buying only 1 16550 (or you may not).

3) (big catch) The software needs to be 16550-aware, and enable the
FIFOs. I believe Linux already does this.

One thing I don't know (and some of you probably picked up on this
already) is what happens when, say, 'ls ' comes over the
serial line. That's only 3 characters. Do you need to type 5 more
in order for the first 3 to be recognized? I don't know how this
is worked out. (Does anyone know?)

--
Adam Goldberg ! Ask me how IOWA STATE UNIVERSITY is
[email protected] ! censoring my usenet access!!
..!uunet!iastate.edu!goldberg ! >>I don't need a BIG BROTHER!!<<


[next article]
From: [email protected] (Doug Evans)

Subject: X386: R4 or R5?
Message-ID: <[email protected]>
Date: 9 May 92 20:17:27 GMT
Organization: Edmonton, Alberta
Lines: 11

Is the X386 that has been ported to Linux R4 or R5?
Or, put another way, is it X386 1.1b or 1.2 (or ?) ?

Secondly, Thomas, are the vga drivers architected the same in 1.2
as they are in 1.1b? Or, put another way, if I have a driver for 1.1b,
how usable is it with 1.2? How much work is involved in changing a 1.1b
vga driver to a 1.2 critter?

Lastly, Thomas, if I want to post my driver, which is for 1.1b, to say,
alt.sources, can I do it (re: copyrights, etc.)? Must I post diffs
(I'd prefer to post the whole thing: driver.c and ban


[next article]
From: [email protected] (Steven M. Palm)

Subject: New User
Message-ID:
Date: 9 May 92 07:51:24 GMT
Organization: Currently Lacking Organization
Lines: 22
X-NewsSoftware: Amiga GRn V1.29 4/25/92 (Beta) by Mike Schwartz

Thank you one and all on the development effort!

I've been following this group for a short while, and decided to take the
plunge. I found it almost effortless to repartition the system and install
the system, using the standard 0.95a distribution.

Perhaps I should have used the mcc-interim release?

Regarding future upgrades, what will have to be done to upgrade an already
installed system? Just a new bootdisk and binaries on the harddrive?

I put up EMacs. Seems to work fine. (Hooray! :). I've even managed to move
mtools over by using rawrite and then tar x
I had seen talk a while back about getting news/mail up. What's the current
progress? That was my main reason for putting Linux up, outside of wanting
to get into a Un*x like environment.

--
-- [email protected] -----------------------------------
-- Steven M. Palm - This whole world is umop ap!sdn -
-- Milwaukee, WI -----------------------------------


[next article]
From: [email protected] (Steven M. Palm)

Subject: Swap file on the same partition as the root filesystem?
Message-ID:
Date: 9 May 92 16:34:46 GMT
Organization: Currently Lacking Organization
Lines: 8
X-NewsSoftware: Amiga GRn V1.29 4/25/92 (Beta) by Mike Schwartz

I installed linux, but only created one non-dos partition. Can I set up a
swap file on this same partition, or do I have to break it down into a third
one?

--
-- [email protected] -----------------------------------
-- Steven M. Palm - This whole world is umop ap!sdn -
-- Milwaukee, WI -----------------------------------


[next article]
From: [email protected] (Peter MacDonald)

Subject: X, ET4000 and mice
Message-ID: <[email protected]>
Date: 9 May 92 22:23:27 GMT
Sender: [email protected]
Organization: University of Victoria, Victoria, BC, CANADA
Lines: 35
Nntp-Posting-Host: sanjuan.uvic.ca


Contrary to popular belief, having an ET4000 doesn't automatically
let you run X386. I have seen reports of some ET4000 cards that just
don't work.

Myself, I set up X under DELL UNIX at work using an ET4000 Ammazing card:
no problem. Now I am trying it at home with an ET4000 vga2max card with Linix.
Using the default "640x480" mode, it gives me four images on the screen,
complete with four pointers, four xterms, etc. I have mucked around for 4-5
hours trying different monitor settings, but no joy. It seems that Clock must
be wrong (half?) or something. I also found a 800x600 mode that gives me
6 windows! The card seems to have something called Hicolor mode (32K colors
I think).

Also, I have one of those switchable mouse (ACE). The logitech setting
works not at all, while with the MS setting, the buttons work but the mouse
movement doesn't. Using another alternate mouse, a logitech only (2button)
causes eratic movement.

Fortunately, I am just borrowing this equipment from work, so am not
saddled with it. But, the moral is: If you think just getting an
ET4000 card will let you run X with no problems, maybe.
Configuring the monitor section alone can be very tough. Anyone know
why you can install a VGA card under DOS, and it can display just fine
without knowing anything about the monitor? The card just uses various
Video Clock, Horz Sync, and Vert Sync combos. If the monitor doesn't
support it, you get the "Outer Limits" on screen.

But not X386. I like flexibility, but I think maybe a more usable
configuration would be one that just lets you supply the Clock/VSync/HSync,
and it would try some intelligent monitor timings. Actually, have tried
going through the video tutorial by Chin Fang. Even tried developing
a spreadsheet under oleo to do the calculations. But the real solution
is to use C code, and then incorporate it into a server as a configuration
option.


[next article]
From: [email protected] (H.J. Lu)

Subject: a testing program for mouse
Message-ID: <[email protected]>
Date: 9 May 92 21:13:25 GMT
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 196
Originator: [email protected]

Here is a testing program for serial mice. I am DEFINITELY sure some
of you have to change the serial_open () to make your mice work. That
is what I did. You have to know something about serial port and your
mice.

H.J.
---------
begin 644 x11r5.tar.Z
M'YV0+E[@B1%#3HT7`!(J7,BPH<.'$"-*[email protected]`(B+-VK4`&'Q([email protected],#AZO!A#
MY,B3(4'4D`%#1L8:-&"4Q'[email protected]+-2KJW,FSIT\`=>;0"2/G(@`T;.K\A/B&
M#IHR"D80/5!9JL:*=Z?&F29,B.'DO"'6E#
M!LB+*UO6O`'CQLP8+%=:A)&VL.&%08<6Y8A4*=:F3Z,>GDRYLN6L+U0H`*$"
MQ)`W[email protected];[email protected]=HFC)L\[email protected]""O,T<^:D>>,&!)TW((*6Z?TU=.\V;\BD,9.G
MMWJAH\74H3,>A',R(.;H9N,=#?;\;YA!QQU$E6&?&_BE0<<M"L\T'@AEEU`>'
M'&_8D0899>#[email protected]&(D:[email protected]%4&F,0"0<<91!%8G=A[`9"
M>6F48=M]_0$)@I%.A1DD=4F.5MII3EH8Y8$^^F<;E7*[email protected]#7YY)MA3%EE
M4<#E5\>4&EKHQAD-/ACAA&Y4>&%O8/YXVI!-AC$A"`'&-EMSMN&VGYZG'1>>
[email protected]`[email protected]!E87:'8II41Z*P0:4"G:X:AC``7>&=^"]1T=\\UV(J1G]&3A'
[email protected]`,6",*+3KTQ7X-S4#G&>E#V=L=H=-#7W8W;P7I==A>Z<)%LM'6:FY;-"6>;
MDR#(4<:-7E((Y(4,4E>DLY,^):B"18Y5Z[#]_7>[email protected]`2Z.Z.&!W;X88CZB0M"
M$J?]=V..._;8H`BUD3B'",O"B"\(9>`!;W:ODAC6K#TN2]2*C4(G78-4(/%$
M$T%,<=L313#!A&])3#$$$T$DT83-0>@,PA5!2"%%$$Y0D401-E^11,RW%7%$
MTD3$]D1L2/0,PA1/&$$%TE(4P4*#23CQMS6&#[email protected]!!--4!"%$$G53D07<3F1MQ-1.0#U%;VF#X,03#19A11%-?SVX
MT3'/7//-.>\[email protected][email protected];\804C"\^!11%#)%$T90WGD390U`!PNL-#O&$
M$U,4$445G#M=M&]!T'P$U+_#SO3B1#0?Q/-1#TX%V)H#7O4453#AM!-O&R'%
MS*D_,07?1H!01?%G4U=]X=%7=_[CV_=V!1)%Q.Q]Y4R#6^^2,+S?M4]X35M:
M[WH3N2/4[7EJ0]WK?N<_*<#L=4Y[POL$F$$G]"9I/7/;[[email protected][Y-L'((C-P`
M"[email protected]/?<6+7H-H)P7728%[email protected][P&-K&1K0C1X4R#2("$*O%(#CH`P0LF
MIH878([email protected]+[`3'[email protected]'[email protected](`,9\""0&>#`!B\(W!1>\)6PC*4L:&"!'3[B
M`[email protected]@`$870("&,!`!S+1`4':]8:[email protected]*`((@,!":3S`@48<@3K05`9B/4%
MX34!"[email protected]#W!>0H(`1\,@,P#%0(V<&2=9)89*'7,\E03`K-]0!#PH0RKS&``([
MO&%'[email protected]?`E!A34P0W:.8,;4C:'VMH39M<&8KW\"&>9$%!"@0`2U!4(+5E"`&(E!``[email protected]`XQQ#`4H*`Y14M#*8R:3
M!2(@PSEMB4O2[!(_O2P4.X-)AQ2D8`<*Z$,J'<5*5\(2.++$U'SN0,][email protected]^V,RG)7&8SGSF':'YAFM6TT#6SB:\[<-.;X!0G.M168*X"G/EM[2H;R,Z#[U^<^`&E*5;@I//769,G4"[email protected])!4XM;VK/G!8*!/Q4
M)C,3P%2BXL>H(/B"'0":`&="4YJN#.D9L2D"I7([email protected],$X1]"`,Y]1J
M"LI94W16]:$OW>DO_U:D6C:I&
MJ3I4O"I6L.+T:D?!2DUK&BB;P#'I-],*@[6V]:UQM<-M\%M3A44TI*+Y&VCY/`%Y`2W#,.]2''S60?D+DNYX'$*=IQKGXYF4RAIH"YG"EG)
MX1BHN"$HY2GG>\C=[G)49'F7+?4S!G;.(0_$R1(;4#`#@[email protected]^)AY0L^8D<8
M*+.2HEQD5K]P!"=480C._"E!-XS03')[email protected])0KXV1MCLD&9)WEI;'\)T)[email protected]
MU<4`9>9%-.M1D'J6I/,[email protected]!"@H)OIH!C22PG7+'IXIF"@+78C&I/!3K059(8
MJ2;^J$')D.*%KGBB%X7Q2&3,SQJ/6;$XUNB..=KCL/Y8FT&^`I&-C&00*/F<
M3>X-E+^\URF+V,K.+/$NI>G:*P5)J7SNYV;._-0RJ_FP32W5:?QJ9AYS5JS"
M+:L;A#SG&!S9SN),@$?M=&\5$-I$B]DC9"Y4CA>MZT;%[email protected]<[email protected]&-5M'C5ILGEK:HK[(
MGEN]Z%?;]MBSKG*MOR#H,IR8R[WE+G!E!5ZI+MK26NZL2,$K`NG&U\_B-JRM
M97EN7F^WNV[X[G`?#6\?S_N]%J)[email protected]=_K`TK5M)$A'+OSU"@3.G
M0`[email protected]/@+[#2!";"Y\8Y_'`0]Z,%'E#F"BV"\O6W8,#0KCASUF`'#;I!N'[email protected]$
M`AX<>`XO*+=9?*#;[/XWW;_%CL#9"=UZ3Y?E+T>Z=][email protected]_-[U1=!Z#EGS
M]3B=9M_O5-
M/C*2GT0"?R]#^,(;_O"(3[SB*;,[email protected]@[email protected];QF/V(`[email protected]+A37_$MP"XRH,$,\,(2EVB$)3<@[email protected][U"DF,E1B3E*Q`!BJPS[WN=[^0S`0/
M-$I:TQM7TYHQP89M23I/>M;3'@=AYU=ID`]](([email protected]_'Q*O;896+(,!B8%U:LSBI*0O"R4IPQMZ#E0
MM3`B8B`E`A4G`ALJPB(IPRPQM(B548B6!HF`:*!9>`B=F,B:0828-DB9+PB9[0B3M!W]0]2UXTAT>R"=^(H+=
M,0>#PD^[email protected]^-,B_HB5M,B8&(AZEMH2LE$RNS,BUT8"N_A"MNP(7>YRO`PG_=D2F4(G\%8R`,B"_0(BW4,@;[email protected]:
M`[email protected][email protected]\X3F\BGIMJ"S^ES`!""(#Z#`0HS%Z2#$3>#[email protected]<$XM1!VA0S,V$LS=',S6"5S97(P590P5;$S,\%#9CDS1F
[email protected](T)Q\S"/5#=%<#=)DS=.`SU]\S>!,SA-8SB(,S6+PS2.`SF2DSN6
[email protected]=,P6?LS.W.#JZ:#2GDSJW(P2L`P(V%#[email protected]/HSN\4T(61!T(
M5#S'DSRWLS/5XSSB*#U.0#[email protected]'HSWM$P3T,T%0<#]3,SD=V3_\`SL`Y`0<1$`ZV3<(1`4*1`4,9#4/Q#E#($&P\P05
M=$%2D$$;%`0KY$%P(P4A1#XC5$+MMY(PVTT/16#;7-T1%A"I0D41+A"--]$0Q$$4*0D50<459M$5=]$5A-`1CA`=>
M9`-)\`8N8(=K%`-M]$9Y*4Q$=^!$AP($B$U%]&9R"1TXU?
M0(_;0W9UA',ZQW,[email protected]`5F$4Z(E'-)P9I8,#'?\9JJ.9L&HFD>LB"[email protected][
[email protected][[email protected])S!J9M/9IA?!`4X,@9S`)S`(9S3>9CJ0P3:
M*9O*^61`[email protected]`=,Z"7`:5_)R9KH(1[[email protected]%]4I<[email protected]@"-%D4[<(11;
M0DQ%H0+PX6X780;[email protected]=-H!]%H"[email protected]*]-2;S\1W)[email protected]@@>]4:)DD`<$.B,'[email protected]<)*A0,JF?JAZ))
M5`)DX!YXL*,]ZB!Y`*01.J$FFBU'N*+N`:/A=J%;[email protected]`:[email protected]'4%*2[
M=`?D80;5=`:_=A$E>J6]D:5FYA&%)F-CP*5A<`;#=A'P40=C,"G7`1[>$89F
MAE4G<1(J,`8`[email protected]&7LM!UQNF`:[email protected]"$NT`;[email protected]#\#R%TQLFD"U,
MNFB4NICDIJ:[HG))T&%"(`5+``)\\#`[email protected])`P([M,'*$>0/*E`"[email protected]:9*JM\P*D<1Y50`%`,)Y^6&H:8&B"RJG(PD*IY.A+%"AVZ
[email protected]::BG)U9&:Y^@4)I:S6:[email protected]`56X#1_TP4]L*[email protected]\*[email protected]@1I0ZXQ
M8&:[email protected]?,4AS8=*7LI&-;[email protected],V4)D7<:VQNJ;4&K"[,JI"P*]VQ*QBX"YA
ML`9WJJ\@0`,XX*]U"JU?0+`#.ZT&*[$)>Q$+6R4.NVAC`+&A1[$`.ZTJA[$&
M6[+F.B8,&[(7,;(C`AB5N6B7%`;E10=)Y!$GFZQ;[email protected](`0TV[(?V[!F)E`7
M\:?V&:@[email protected]>$>IB&BJB*[email protected]:,VAN.6CR%MB[!W%+/4*@(J$`K$H
M^[.\&K1[J[!X:[8CPK([RTHJQ[[email protected]%J[)!R[**"[)Y:R!#[email protected]
MRP8Z.Q*0^V0J(*%P.[@8.[>'*[0L<;>MT+7G>A%[email protected]`(@46$6UJ=)"[email protected]/^T51R[N+VJB$V3]!H+6SRK7,NJH:
MYJA-H$'%XZ?.*ZA.BP*Z*[75:[5#$+[O8S8KU[W#:G.YI0"^1QT:^J%E8)T0
M`@(MT`(-M.P8H<`[email protected]:+-ZL%`TJ9"A:?-VJS-`0=;0`-=P+SY,6)[email protected]$NP4UT`4H
[email protected]@:\[email protected]*UFL)`[email protected]$P19,\0H.03+,ZI(4`50\#,@X&3BU1E-`"4X
M,@7(PE][email protected]`.BZAD<]P10(`1!/,1%?,2>D<1+#`)-_,0[(\5?UP215P93
[email protected]&#TH:A:?!)<[,6C:JIEXP1DW,=)\[email protected],9$7`1&_,4_\P1*O#-,[,10#,?%
MU01#`Q5=8AL4M\6$R<[email protected](QC7,:><<:*K,:.S,:0_,:],<6I\[email protected]+OTFLDH
MD45Q"$!1R+!YY?!%]`*]N&K;81,(@[!$\G*\[email protected]:DP1'^05I
M,S5)I,,>L<);$`,XK'(S4*;3W"=;(`/7_'G:?!'4/`/?_*[2[+&+FZ?2++,&
MHLQ6P,Q%\`7#`[HAG*#H"[6[[email protected]@1SW++WY/+5$4`8Z$B6]\01?H(Q7
M`#NCBM"-DS-!D`7^U'-U=*M3=1'XJ9]D-03.,87]W!T47`8*)[email protected](A
M,`5P&B79T5,)4+OUC$WI2[UE0`5Y0"[email protected]$`(J!P5?(`1-P$[[email protected]$*536J5D
[email protected]`)HVQLS[0)[email protected]+-D(`5``GZ:[email protected]+/;4V325=X-(70=1T0*4UMO7-0O7Z&NZ94/=8T?=5ED-5M_=9:/=1S7=1?C;RQF]>(F<]E_=11O:5_7=7ZX=9P;4C9YG7_W-CJ6].#C7([
M_04H>01)<*LC(=0C<;OTL=G3J\]4&Z0B,`5&ZFEQ?1)<[=49M]2/W=>,W=J[
M'=EI?09;P-.4?-G9YLN8_:I;O-1S8"EP0!9]W7,J=Q$69KO8DKML[=J^*P)/
M,-N%W7":S=S.#=U1+=TD40/LM"S7S=H!O;Y/[email protected]>[email protected]:#B?3)E$-T\,-TN
MD=ZJC=V!K=VPO03Q3=(@`WT$2-_9W=SVC=_330/!:]VXR]Z=O=U,,.#9-M\R
MG>#C?=_EG=\7T:_\O=Z4_=IZ)@46?A$8+N&*J>#/S>'[email protected]@`/H#>&K/>+;
MW00G7N`C$MX:ON`=KG*`,>/J'>$V#MM1D.,-9\M*GMHBGMV]"]M.8.%(JZIY
MNM35)!2YC*2X1*W+>LZ&_=]ET`;E%=534"$&HJU>7K0P2^4>PM0,\Q"['N#,^-A)+%0^<>P>/_;=6?G7+5(=I/0-H63>#]K>(`KF=5D..W;=>Z
MS=?`/;22K=8\/=I)8-Q3;KIL,("*_N0]E:=J'L[#ZQ$P;=)?L])>,@?A9DCW
MVQD97`8A2B&V\;\!/!)28$3%TB1E8%($3>[email protected],Q1RD\4M#U`29P4!NV,23#
MGE[Z\NN-0DP(XIS71U]/&NNSWBAS,'#X>DLH3,MRD,MFL`67]\*+)L)[email protected]
MX`9"D`?T,0>]D:/]Z:-+:F87BNXI#`?B#@5<;N\R7%E%!:!0)>Y;@`/F'J__
M_G9899MO8,-=X/#4>LS%!0)H0`9?P"EK0/$6#TLN!P<77QL93P8>S_%Z\AKO
[email protected]`<@D``[email protected]`M`/,M/[%&CPM7,FC<8*PO,-9W_5?S_5)O_0S(/;%Q02L[,IH$/1#7_0QO_9*W_5+[_1]L/.=
MD]1%WAOI)/`JD`[email protected]=F0%$*7`8!8FKB
M[D\3;=(PG(DH0"+EFJII0-'L[NY>@OHKL`)E.Q+W.Q*=,[email protected]"1/MM_()4J-[=7_Q^,\=3[X[-`?77&;8$.VEJ^.'\'1'[email protected]/([9].!]L4C*E`$
[email protected][`T42S*"(E3S!AO0%X0N`5:\%\*_O\!683ZMX$RX&F;"BFZ.GFE4+J'_V
[email protected]?[LAC=P!WI$O),1)W#Y/078H"RN"&S0#\@%+)R7?-4=JH1V8`-Y``(V*U[&
M[-K+\2-_9:`%*$"HT#'*!+N0=]M"]Y&*($+[KEZ?2U#Z#O/Q.YU61V;5R!$G
MZ.3RC;[email protected],-&#L,#;)Q-L&&U;I;58,CSK)M`7)&TOC`J%*#6X`-SC^N
MYT^\W%0)@V:`WVFK(#9>UD`:Z$SCSP&6/Y2'W*299D.$BE!9F4$T"`@%H1O$
M$=\!#K8V0#<'84`=)(-XT,E5ML'V\*R9^DMYXL3WB)K:UP#]G]QK9=BO[A&^
M)N`<_A\O:P%^8T>[email protected]"!4F:9IB%[F-W_9);$M60('.98'.,XF4,8S(:[email protected]
M00YGH$Q$0MHR!Y)(&QH*"((HMH>I'`\K$.EP63V&77)%B80SSGQ8AABMX^L0UVG5I8QO>!_$X1,4#J>!`>TZ]*`J&$600([email protected]`@M>"41U$9P`#8PM_JD7
M0KO3,"[email protected]_*[email protected]/4ERI&`[email protected](_[(`[email protected]`/3#[email protected])%0-QR]Z\*P8'OUR+\
MX3`T$%@P.LS"0I(`_%P<=&UN;?X1-YLAOXC535H4P
MVZ"V(X6B;(/*02>])05X/LP7"[email protected]'<$#+(&=^``?$+%,VTD858KNRM&!+%<6
M9]5;A`'HK;;)M>K4H?17MJ,#*``+PKLZTAN6%X&C,@F`[[email protected]*9,*L&&%GF`..
M$!(VP)=X]DR=(9%FB'`+8,+7]\TZH;D;"9=PW]G!4!0*\R`K;&%U4#*&G3(`
MPZ:8$E1V84Y6P`8S4%[X0R1\@'K*"P8S!5$-Y
MMF!97#,[email protected]:4"H.PE;A%I-OG6"3MPC6X2/?W$&L!,>P`-`@`[email protected]
M)[email protected]%HS\VB8C`!?\7(O2C\:%;6(3LBC6VQY,&`(#,@":1^_F'\<=]Z,0M!WD?1\*H\W+K#&)-Q2'@'"VC+QP1(!"/?08000G)(G<\BW[`0H:MG&Y):T)^R&I6!U<`-A&2H9&MNG2F')[email protected]!N3:!%NDA*]C4LF=D3CEYRT&$Z,=FLXM[[@
M>[R02J]'[email protected]\H!\M)(.#F`[email protected])0
M2LE"[email protected]):D[D>K,=*G(GA8F;1DNTV71<$XJ1OV8(^_DCL20>C)/_4B/8"53
M`**$D(K2?QU)`.4J+66:XY*5<"1H*`XE%`RCB$*,BI'>":DEB0G['>W*+=?N
M,.FOSW"H2`,V*5%A"#D^&M>(TU1.M*2-BV;$#31CM[:R"$%[`91J#H`3,U/?
M6EQT`W(UH,L-QJ<8YL8M9%TN2^Z`278%"EB1/25;*DL-=IU0P)UK9L^,"MS'6+?!'":M!\0`-L&0,\`,B('&J/0$9*J"9:1O!^R`Y"@P0=2OW';$M<^G)+$E^F2(M`Y_.`M2D"I8'\VB>NS`7.D_7O`@TX`;0$950>C)"V&P)JF<&W``:,!BNYF60
M/8OA*-2>QS`D)(/;O)MXTR?8+^SS-;*`S;`"+^`P.8AW47,HQ)403O?GNFPG
MZ>01?,Z"[email protected]#IST9B,9I!L9`HV`#D3,ZC:?&&:<&6^9[email protected]$1W/RG,;YQ61!N`G=RI<5J("8$[&:?HI`_;1S&A
[email protected]`5+`[email protected][9^[email protected]@G(0,X-D=2M3Q=!)]2N[\%Z/IS)S`
MU&!Y1:?MM+.+F>?41Q%(`#[email protected]?#(6[CD\$H`,4)_;DWS*N00P`UQ=9X`6OP).
MU18D`1;0`X9`&+].#L`&PK=HL,!AVI8?*->[email protected],Q;F6H^1-Q2E8"*8:
M!5MA5&8&WS4K%M4>%&-E;).=H;=`A8##'X#*+"$6-"BDY3$-FJ><@NZ;#M3$D(]X`[email protected]:(G:,(0&!YZ&H+88UL1]*&)IK#BHAC*7U&,%E^-BLXRAE<"[email protected](Q8=0
MHNX1"-B0*8BB!/L5SJ&C*`A#[email protected](U79K*C#I1_".ZMQR2HG`J>%4N/[email protected]:2?`@R$%&`Q43GI!;R?PFE7%^WB8L3=M=6B'XL09D0;[email protected]'^ZA[",DLX>5>GQ,)Y[&'EDH).&4!X!2C')R%)#`$."N5'C
M\^+L:`\%8C^4];$Z3][email protected]/!1E(I7FG$J:I^:`?!AW+DR1GH3Q8N9,[email protected]
M$44A$W4Z7%@&7AD2_:4IK%V&%-I(3.W#?*`-0X$FOKA,Y$9$([email protected];[email protected]*Q
M9PH'I.D9-:;/%)OF,[email protected]$&-"A.<<4:&]0`0/5!A`^*/`KLN?+
MF8HC(24LN5M&Q7K>=+PN#_4+#(TA$!P8UX
M5*BGO\J>U8.?7Y+0D;:+,'I6ZM>YA72/OVC/F.K31D+FL:F?M`H"@&J9$"8$^[email protected];Z%%(E%H\47SW15X(S8051.`WQ=)8]N3O%3)N5"A!H!*V!
MZDMYFL)((;:LH&2U6?VV+#I69QF+(V]L%:W.,NUH'3,H.&U6+M1+P%`L-4,'
MIQNPH9T!A^H&J+!#LVH0$Z=DSLPITAMZ3IM#.C6A_J^=_D1H"$^]X%(EI$`,
ML;[+>,GJ#$D?^')QT*!RAIG&4.5`GYII"A2T3B^'&IH&W="(J#8UCEG4+&92
M6^N*[*A`%:3R,I'J&31$2<6HME6C-H&1H%)7ZB1KJ9=,IV943$=38ZN;Y*;>
M%*8.NI[J$7ZJ<0U\H)*7$55!.79V:JKS5JL.?'I7H(0\T*=T-0+T\GS"``50
M>*3FXRE>-0(.+*:\Z1,LCQW9FF_A),B%%!83:FI>F`%V!"[email protected])I*$#`/VW0]
M]-4PP$VCT!AL3]U,L!`[email protected]'.MT^E<182)4A`Z++UP`';`#7V"*JIP(HT0HC/)2
MEWCUQ=;8&_L%>.R.G:+`,'DI+^I`8^>#C<6Q!(;4N:G!&J>0%P,)'[email protected],
M+9MEMY>0G0]*4\**V3%[&=SK02A$F(0LD%F)D#[email protected]\]#4&>!?RR9,
ML`$VX"_$`)AP8`G#FJT*"Y;V.(:K<'OLYI\]M$S3"30/U*%R0IH">`)"0`G8
MC$8[QUP`>!"Q\O4-*(`AP%%5SAGP5OZKE\P*,;!I.^V68"6;[email protected]@`C2,P:?W7
[email protected]!&4**P%K;&05#A`B5`[email protected]=!0!1?AD#D`X40`D``4HVK^!:@0)"H"TDO96
M!5M.RTZ6K:J]'E.`G;0`\!!LAVT1:+;'-M)&6V-;-X1`M!4G0*";SH8G$Z'R
ME`BHMHLV4-F&![%+E&*X+0[4%@7H6BAP81"MO;VW^#;?ZMM]RV_[K;_]MP`W
'X`K<@=LT`=F&
`
end


[next article]
From: [email protected] (Genesys)

Subject: Sound Blaster
Message-ID: <[email protected]>
Date: 8 May 92 06:24:42 GMT
Organization: Computing Technology Services, Michigan Technological Univ.
Lines: 6

I've seen (and have) the source to integrate the Sound Blaster with the kernel.
Has anyone ported this to Linux? The current source(s) is for either SYS V or
BSD. Since I'm new to this, I'm not sure how to go about adding it to the
kernel and whatnot. I'd appreciate any help anyone could give. Thanks.

Joel Lucsy ([email protected])


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: X and 9600 bps (ANSWERED!)
Message-ID: <[email protected]>
Date: 9 May 92 23:39:10 GMT
References: <[email protected]> [email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 90
Nntp-Posting-Host: romeo.cs.colorado.edu

In article [email protected] (Adam
Goldberg) writes:
>In article <[email protected]> [email protected] (Joel M. Hoffman)
writes:
>>[Max 4800 baud]
>
>In article [email protected] (Al Clark) writes:
>>I have been using kermit with a Robotics Sporster 9600, using 19200. The only
>>possibly unusual issue is that I've got a 16550 UART. I get the normal
>>9600plus display update, with no observed errors. File transfers rarely show
>>an error.
>
>In <[email protected]> [email protected] (Joel M. Hoffman) write
s:
>>Perhaps the problem is that I'm using a direct serial connection? Is
>>it possible that the modems are doing some sort of buffering that Linux
>>needs? I know the serial ports can go up to 116,000-ish, because
>>Laplink does it. Why does Linux give up at 9600?
>
>Even though I may sound like I have the definitive answer, please don't
>take the following as gospel (though I believe it to be basically
>true):
>
>Here's the story. Joel is able to get only 4800 baud, using a direct
>serial connection between two computers. Since he didn't mention
>otherwise, he probably doesn't have 16550 UARTS. [For those not in the
>know, most UARTS are either 8250s or 16450s. They are functionally
>identical, though different internally. These are just plain-jane
>regular UARTS. A 16550 is upwardly compatible from an 8250 or 16450
>(just as a 386 is upwardly compatible with a 286), but it has 16-byte
>FIFOs for both input and output.]
>
>Al, on the other hand, is able to get a 19,200 connection. He has a
>16550. Bingo.
>
>What a deal, eh? You get 19,200 with the processor overhead of 4800.
>What's the catch, you ask?
>
>Here's the catches.
>
>1) A 16550 costs between US$15-25. Some motherboards have serial ports
>'built-in' (ie, a 8250 on the motherboard) (ugh), or in worse cases,
>consolidated in a VLSI chip. In this case, you need to disable the
>motherboard serial port, and buy a $20 serial board (see below).
>

There is also a clone available from TI? for ~$10.

>2) Some IO boards which have serial ports on them have the 8250 or
>16450 soldered (not socketed) on the board. Make sure you aren't
>buying one of these. (I got a 2 serial port (both socketed), 1 Game
>Port, 1 parallel port card for $15...look in computer shopper).

An IO board is going to be double sided at worst. If you
are proficient with a soldering iron, and desoldering braid
this isn't a problem.

Some IO boards will have a higher scale of integration :
these MAY be a problem.

National semi makes single and dual 16550's available in
chip carrier packages, as well as a serial / parallel
chip.

>boards don't come with a 16550 (though some do--ask), it's no big deal
>to swap the 16550 instead of the 8250/16450 (they're pin-compatible) ,

For all practical purposes yes. The 16550 adds a few signals
to support DMA, and one obscure pin (not used in PC implementations)
had a meaning change.

>3) (big catch) The software needs to be 16550-aware, and enable the
>FIFOs. I believe Linux already does this.

Wrong. Read the code in

kernel/chr_drv/rs_io.s
kernel/chr_drv/serial.c

The changes are quite trivial - I won't go into details here
because my brain is scrambled from programming an odd,
non-standard duart on a single board 8088 -

>One thing I don't know (and some of you probably picked up on this
>already) is what happens when, say, 'ls ' comes over the
>serial line. That's only 3 characters. Do you need to type 5 more
>in order for the first 3 to be recognized? I don't know how this
>is worked out. (Does anyone know?)

No. The 16550 will generate an interrupt condition when
- the FIFO trigger level has been reached
- there is atleasst one character in the FIFO
and 4 character times elapse without receiveing another.


[next article]
From: [email protected]

Subject: Serial ports and speed under linux.
Message-ID: <[email protected]>
Date: 9 May 92 23:47:32 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 19

Let me add my machine setup to the list for comparison.:
386-33/Linux--No 387--et4000--14"non-interlaced--1.2MB from HP--1.44MB Teac\
Seagate ST-3120 100MB--4MB Ram(soon to add 4 4x9's) generic IDE/floppy \
controller, 4 serial ports consisting of: 8250, 8251, 16450, 16451.\
I don't have any 16550's(yet)

I have an XT ~5ft from raskol(the 386). I built my own LapLink(tm) cable.
I have used ALL ports to log in at 19,200 at one time or another. Running
zmodem to transfer files twixt the two, I noticed that when the HD kicks in
(ie. sync) The zmodem xfer would lose bytes. Otherwise, it works fine.
On another note, I have used xcomm (xc) to connect to the XT at 38,400
No problems except for the losing bytes when HD kicks in.

My question for people with high-speed problems: How much load was on the
system? The mor load I put on mine(especially re-compileing kernel), the
more errors zmodem caught.

I only have a 2400 baud modem, so I can't say anything about high-speed
via modem.


[next article]
From: [email protected] (Krishna Balasubramanian)

Subject: X386 quickfix for non-microsoft mouse
Message-ID: <[email protected]>
Date: 9 May 92 23:58:19 GMT
Sender: [email protected] (NETnews )
Organization: The Ohio State University Dept. of Computer and Info. Science
Lines: 72


/*
It appears that along the way obz's x386 port is setting istrip

and hence eating up the hibit which is needed for syncing non
microsoft mouses.

Heres a little program picked out of x386 source which will set
the appropriate cflags and -istrip.
Seems to get my genius mouse going in mouse-systems mode.
you could probably run it out of the ~/.xinitrc
although I've only tested from an xterm.
Make sure you have a ~/.twmrc that makes some keyboard action
possible. ... does anyone have one with decent keyboard bindings?

You might want to change some of the cflags (or iflags) ...
look in /usr/include/termios.h for possible values
and Roell's x386/x386Io.c for expected mouse behaviour.

I suspect linux does not do anything with some (many?) of them.
simply compile the following and run it in an xterm.

krishna
*/

#include
#include

main()
{
char mousdev[] = "/dev/tty65";
struct termio tty;
int mfd = -1, i= 0, sbuf = 64, nbuf =0;
unsigned char c = 0, in [64];
unsigned short cflag;
unsigned short iflag = 0x01; /* ignore break */

if ((mfd = open (mousdev, O_RDWR /* | O_NDELAY*/)) < 0)
{
printf ("could not open device \n");
return 0;
}
(void) ioctl (mfd, TCGETA, &tty);
printf ("iflag = %04x \t oflag = %04x \t cflag = %04x \t lflag = %04x\n",
tty.c_iflag, tty.c_oflag, tty.c_cflag, tty.c_lflag);
tty.c_iflag &= iflag;
(void) ioctl (mfd, TCSETAW, &tty);
sleep (1);
(void) ioctl (mfd, TCGETA, &tty);
printf ("iflag = %04x \t oflag = %04x \t cflag = %04x \t lflag = %04x\n",
tty.c_iflag, tty.c_oflag, tty.c_cflag, tty.c_lflag);
cflag = B1200 | CREAD | CS8 | CLOCAL | HUPCL ;
tty.c_cflag = cflag;
tty.c_cc[VMIN] = 1;
tty.c_cc[VTIME] = 0;
(void) ioctl (mfd, TCSETAW, &tty);
sleep (1);
(void) ioctl (mfd, TCGETA, &tty);
printf ("iflag = %04x \t oflag = %04x \t cflag = %04x \t lflag = %04x\n",
tty.c_iflag, tty.c_oflag, tty.c_cflag, tty.c_lflag);
/*
while ((c = getchar ()) != 'q')
{
if ((nbuf = read (mfd, &in, sbuf)) > 0)
{
for (i= 0; i< nbuf; i++) printf ("%02x ", in[i]);
}
}
*/

close (mfd);
return 0;
}


[next article]
From: [email protected] (H.J. Lu)

Subject: Re: Problems compiling UMoria 5.4: please join GCC
Message-ID: <[email protected]>
Date: 9 May 92 22:12:56 GMT
References: <[email protected]>
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 34

In article <[email protected]> [email protected] (Ben Combee) writes
:
>I've recently tried to post UMoria 5.4 to Linux. It compiles fine,
>except that I had to comment out references to SIGBUS and SIGSYS in
>the signal handlers, and change the signal handling routines from type
>int to type void so I wouldn't get GCC warnings. Still, after getting
>it all linked and installed, when run, I get a GP error 0000. Are there
>known bugs in the MCC shared or static libraries that might cause this
>to happen? Thanks.
>
>(BTW, the Curses routines appear to work pretty well. I compiled a few
>test apps, and they ran nicely.)
>--
>Benjamin L. Combee | -===-
>Internet: [email protected] | = ___ "Home of the Big G Burger"
>1991-1992 joint enrollee | - = -- MST3K, Comedy Central
>C, Modula-2, C++ resource | -===-

I hate to see people complain about the old bugs which have been fixed
in the new release. If you are porting something to Linux or want to
use the latest stuffs and are not afraid to ftp twice a day to get a
bug fix, please join the GCC channel on the Linux mailing list.

BTW, gcc 2.101 for Linux will be released shortly after 0.96 is
out.

H.J.
--
P.S. To join GCC, send mail to

[email protected]

with only one line

Mn-Admin: join GCCC


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: Problem building kernel 0.95c+ (and pre-0.96?)
Message-ID: <[email protected]>
Date: 9 May 92 23:44:44 GMT
References: <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 29
Nntp-Posting-Host: romeo.cs.colorado.edu

In article <[email protected]> [email protected]
writes:
>Linux Activists:
>
>The file tools/build in linux-0.95c.tar.Z distribution will not compile
>because of the line:
>
> #include
>
>Basically the compiler cannot find linux/fs.h because there was no -Iinclude
>specified for this file in the Makefile. I tried adding a -Iinclude, which
>helped a little, but this created other problems because some of the data types
>in linux/fs.h were declared somewhere else. To fix this, I ended up adding
>the lines:
>
>
> #define MAJOR(a) (((unsigned)(a))>>8)
> #define MINOR(a) ((a)&0xff)
>
>to build.c and removing the above include in order to get the thing to fly.
>
> I looked at the sources to pre-0.96, and it looked like situation
>had not changed.
>
>-Eric

This is NOT a problem. You should also have a linux directory
under /usr/include. To keep this current, a symbolic link from
/usr/src/linux/include/linux to /usr/include/linux is
probably the best thing.


[next article]
From: [email protected]

Subject: Re: Smail Query
Message-ID: <[email protected]>
Date: 10 May 92 00:15:21 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 41

After finally getting all the programs that smail is dependent upon(libsdbm.a)
I got smail to compile quietly. To test it, I logged in as a user and root:
root> smail jim
This is a test message.
. <=required to exit edit mode (also exits attmail's edit mode\
hence the indentation. ๐Ÿ™‚
root> (approx 5sec later, GP fault. Hmmmm.)
jim> more /usr/spool/mail/jim
This is a test message.
jim> (there was some header, but it was weird.)
I then remembered that emacs can mail, looked it up in emacs-help,
and tried that. It gave me a To: and Subject: field automagically.
I sent it, and no GP fault. Hmmmm. Maybe it requires the headers?
Nope, I still get a GP fault when running smail directly. If I use
emacs to create the mail, no problem. That is why I am looking for elm
sources. I want a decent front-end to smail. I am not a great fan of emacs,
it is a system hog, and takes too long to load. I use mg for quick edits.
I only use emacs for e-mail locally only. When I get around to it,
I will create some default key-bindings, and fix it to use the alt key,
But I've been busy working on my new prototype serial board. It is still
in design phase. It will have 4 16550's, and be set as ttys1-4 default,
with an option to jumper for ttys5-8(one jumper) It will also have a printer
port, lp1 default, lp2 if the (one and only)jumper is shorted. It will share
an interrupt(ports 1-4 on IRQ4, 5-8 on IRQ3 lp1 on IRQ7 lp2 on IRQ5)
With two of these boards, you can have 8 serial lines, and two printers,
and only use(lose) 4 IRQ's total. Everyone I have asked to help, or
described it to wants one already, So I have some incentive to finish it.
(besides wanting to have linux FULLY usable, and make a little money.)
If you are interested, let me know, and I will keep you up-to-date on
development.
The development of this card is VERY linux oriented(I wonder why?), so
to make them usable under DOS(I have customers that are still misguided), I
have to write a few special drivers. ๐Ÿ˜‰ That will come after the card is
developed. Making them work under linux will be very easy(I think).
I will set up a (small) list about it. (outgoing only, you can't have
everything) ๐Ÿ™‚
--Jim Gifford
PS. I want to know one thing:
WHERE CAN I FIND ELM SOURCES?????????? please help. please e-mail. ๐Ÿ˜‰
[email protected]
...!uunet!attmail.com!jgifford


[next article]
From: [email protected] (david nugent)

Subject: clock.exe
Summary: Looking for ftp site to obtain CLOCK.EXE from X386 distribution
Keywords: X386,SVGA,Genoa,ET3000
Message-ID: <[email protected]>
Date: 10 May 92 00:03:13 GMT
Reply-To: [email protected]
Organization: The silliest things can happen, if you only ask!
Lines: 20

I need to figure out clock timings for my SVGA card (a Genoa 5300, which
has an ET3000). The card is supported, and X sorta works but it's a real
mess... ๐Ÿ™‚ Anyway, I no longer have docs for this card and figure that
the next best thing was to obtain CLOCK.EXE mentioned in the vga
tutorial. Does anyone have this item separate and apart from downloading
the complete X386 distribution? ftp site? (I would suggest that there
may be many others in a similar situation, so it might be a good idea to
post it on one of the Linux sites).

Alternatively, if anyone has the clocks entry for this card, I'd love
to have them. ๐Ÿ™‚ The monitor is in the database, and apart from the
obvious sync problem with the card it seems to display semi-readable
graphics, but in this state it's just not usable.


..............................................................................
david nugent Public Access Usenet "Only Nixon could go to China"
[email protected] +61-3-792-3507 - old Vulcan proverb
3:632/[email protected], 58:4100/[email protected], 199:4242/[email protected], 33:300/[email protected]
PO Box 260, Endeavour Hills, Victoria, Australia, 3802.


[next article]
From: [email protected] (Michael Hamilton)

Subject: Re: X-Windows Story
Message-ID: <[email protected]>
Date: 9 May 92 23:01:09 GMT
References: <[email protected]>
Organization: Wellington City Council, Wellington, NZ
Lines: 84

In article <[email protected]> [email protected] writ
es:
>First of all, my kudos to Linus and everybody who has done such a great job
>with LINUX over the past ~year.
>...

Couldn't agree more. And here's my experiences with X386 so far, which
I hope might help others.

I found clock.exe at capella.eetech.mcgill.ca in pub/X386/clock.tar.Z
It only seems to be written for et3000/4000, it hangs my machine.

I don't have an ET4000, I've just placed an order for a Tseng ET4000
(NZ$332 ~ US$150 - sounds cheaper in US$ ๐Ÿ™‚ with 1MB, with a proviso
that it should work before I accept it. Is there more than one kind
of Tseng, and if yes, do all of them work. Any info beyond that in
the vga database would be most welcome.

Is there anyone out there with the skill to create a B&W X386, as I
would assume this would consume much less memory and swap space, and
is just fine for many applications (my X-terminal at work is B&W, and
I seldom miss colour).

Anyway...

I'm a bit impatient, so I tried X386 on my existing card, an Hewlett
Packard Headland Video 7, of some kind, with 256 KB - I presume this
is not a supported card, anyone got any info on this brand of card, as
I've got nothing technical? My monitor is a Philips 7 CM 3209. I
found that if I selected et4000, 640x400, and lied about having 512
KB, I could get a ghosting and wrapping image. I've also found that
the following gives a crisp image, but with wrapping and alternating 1
inch blue an white striping horizontally across the screen:

WARNING - during playing, my monitor sometimes received some pretty
weird signals, and I often switched it off just in case it was being
over stressed. I don't fully understand H/W, so I didn't want to risk
it.

vga256
chipset "pvga1"
clocks 25
modes "640x400"
vram 256

Modedb
"640x400" 25 640 648 768 800 400 412 414 449

This is not usable, but you can certainly see, and operate, pieces of
windows, which is quite exciting. I.e. The clock comes up, as does
Xterm. With some guess work, you an mouse into the mess of wrapping
and stripes that confuses the real X windows, and the right response
takes place. Response is quite nice - snappy even - although I didn't
try any scrolling text. Panning seems quite snappy. I've got 8MB
plus another 8MB of swap, and I got the impression that I was eating
much of the 8MB of memory, as logging into an attached vt320, caused
the disk to work much more than normal.

I found that X complained about /dev/tty0, so I tried creating one,
I also tried linking /dev/tty1 to /dev/tty0, both of which got me going.
But I don't seem to be able to type into the Xterm window.

If you instruct twm to exit, the system either goes into a loop, or
hangs, I don't know which. I have a vt320 connected to /dev/ttys?,
which definitly which hangs when I leave X, so it is either a hang or
maybe the kernel goes into a spin. Normally I exit X, by using the
reboot command from the vt320. I always sync before entering X, so
that I can power down if anything bad happens. I also find that it's
a good idea to start X from the vt320, taht way I can see the error
messages.

It all looks very promising. A great effort - congrat's to all involved.









--
Michael Hamilton, Computer Services Section, Wellington City Council, P.O. Box
2199, Wellington, New Zealand. Phone: (64) (4)801-3300 FAX: (64) (4)801-3020
Domain: [email protected] PSImail: PSI%0530147000090::HAMILTON


[next article]
From: [email protected] (Toshiyasu Morita)

Subject: Re: X and 9600 bps (ANSWERED!)
Message-ID: <[email protected]>
Date: 10 May 92 01:14:42 GMT
References: <[email protected]> [email protected]>
Organization: Netcom - Online Communication Services (408 241-9760 guest)
Lines: 152

Adam Goldberg writes:

>One thing I don't know (and some of you probably picked up on this
>already) is what happens when, say, 'ls ' comes over the
>serial line. That's only 3 characters. Do you need to type 5 more
>in order for the first 3 to be recognized? I don't know how this
>is worked out. (Does anyone know?)

(this is in reference to 16550 UARTS, BTW) ๐Ÿ™‚

I picked up a file called "16550.C" which explains the 16550 quite succinctly...

/******************************************************************************
16550
a utility that iturns on the FIFO of a NS 16550 chip
by John Welch, of Igloo Computers
Public Domain Software
Hobbyists, feel free to use, abuse, pirate and re-distribute this program.
Commercial use prohibited without consent of the author.
Send fan mail to [email protected]
Send flames to [email protected]/dev/null
******************************************************************************/
#include
#include
#include

int fd;
void outb();
extern int errno;

main()
{


/*
I've waded through the NS 16550 docs for several days now,
and what follows seems accurate enough.
At any rate, the basic idea is that a 16550 with FIFO enabled
has (*FINALLY!*) cured igloo's problems with losing newsfeeds. I
thought I'd make this available to the net in the hopes that it may
help other people who've been singing the 'microport lost character
blues.' As microport has been unable to cure this, we finally fixed
it ourselves. By the way, if you aren't tech-y, this may bore you
to death. I've got it written to patch tty0 and tty1, and since on a
16450 or earlier this register isn't writable it should cause no
problems. Just un-shar it by typing
sh 16550.shar
and compile this with
cc -o 16550 -O 16550.c
It might be a good idea to chmod this to 700, to prevent users from
flushing the buffers on you. Adding this to the inittab file or some
such place should be all thats needed. Run it once at boot-up and
forget it.
A bit of history here: Microport's serial port drivers are
brain-dead. Apparently the problem is with the high-water mark in
the kernel, and so it's *not* fixable with new drivers. What this means
is that you lose characters sometimes. It is particularly noticable
while doing a 9600-baud newsfeed on one serial line while another
user on the other line is cat-ing a text file at 300 baud. Usually
when this happens, you get many errors during the newsfeed and as
often as not you lose the feed altogether. Microport has been told of
this problem many times, and they've piddled and fuddled and moped
around trying to fix it. At first they said we should get faster
hardware because an 8Mhz AT couldn't handle 9600 baud. Fine, we found
a place that loaned us a 20Mhz 0-wait 4Mbyte 386 board for a weekend.
Guess what??? The problem was just as bad as before. Then they sent
us new driver to install, which did not help the problem. They then
suggested that we need a smart serial card, to the tune of around
$1500 or so. They think that 9600 baud cannot be done on these machines.
That's bull-squat, folks! XENIX does 9600 baud just fine. Microport is
unable and unwilling to fix this problem. We had to explore alternate
ways of doing it ourselves, preferrably without costing us hobbyists
an arm and a leg. We feel we've found an acceptable solution with the
16550 UART chip.
The 16550 is a half-way step between a 'smart' serial card
and the usual 16450-type 'dumb' card. Using the FIFOs in the 16550
can not only prevent lost characters, but can result in more efficient
CPU utilization, with less time wasted in processor overhead to read
each character sent. It does this by a device called a FIFO buffer, a
First-In, First-Out scheme. When receiving characters, the 16550 will
only interrupt the CPU when one of two events happens: When enough
characters have been received (you can define 'enough' to be 1, 4, 8
or 14) or when at least 1 character has been received and there has
not been another character come through for 4 times the time 'width'
of a character. In this manner, when the CPU finally gets interrupted,
much more data can be dumped to the CPU at once, cutting down on all
the overhead involved in context-swiching and so forth.
In addition, if the CPU takes too long to read the FIFO, the
16550 will send its own flow-control to throttle back the incoming
data, preventing buffer over-runs that have plagued microport from day
one.
The 16550 chip is a drop-in replacement for the 16450, and it
costs about $25 or so (not much more than the 16450). With this chip
and a trivial amount of software, you can not only cure microports
brain-dead serial device driver problems, but you can also enjoy most
of the benefits of a 'smart' card with very little cost.
What follows is an explaination of what this program does and
why it does it. If you have multi-port 'dumb' boards with 8250's,
16450's or whatever, you *should* be able to replace those chips with
16550s and modify this program to set (base+2) for each port.
tty0's base address is 3f8. The FIFO control register is at
base+2 (3fa). The byte written at this address is defined as follows:

Bits 7 & 6 define at what level the 16550 interrupts the CPU. 11 is 14
bytes deep, 10 is 8 bytes deep, 01 is 4 bytes deep and 00 is 1 byte deep

Bits 5 & 4 are not used, so I set them to 00.

Bit 3 defines the performance of some pins that are not used on the
16450, so it's not likely to be of consequence to anything we do. I've
chosen to keep them performing the way a 16450 would, setting this bit
0.
Bit 2 clears and resets the transmit FIFO.
Bit 1 clears and resets the Receive FIFO.
Bit 0 turns the FIFO buffering on.
To turn on the FIFO at tty1 to a 4-byte level, one should
write a 0x47 to port 0x2fa. To set tty0 at the maximum FIFO level,
send 0xc7 to port 0x3fa. To disable FIFOs at tty1 send 0 to 0x2fa.
*/


if ((fd = open("/dev/mem",O_RDWR)) == -1) /* open mem dev for read/write */
{
perror("Open /dev/mem");
exit(1);
}


outb(0x3fa,0x87); /* this turns on tty0's FIFO, 8 characters deep */
outb(0x2fa,0x87); /* this turns on tty1's FIFO, 8 characters deep */

close(fd); /* all done setting up FIFOs. */
}


void outb(portno, data)
int portno;
unsigned char data;
{
io_op_t iop;

iop.io_port = portno;
iop.io_byte = data;
errno = 0; /* clear error indicator */
ioctl(fd, IOCIOP_WB, &iop); /* write the data to that port */
if (errno)
printf("Error setting port %04x\n",portno);
/* send to stdout so they can re-direct easier */
}

Toshi Morita
[email protected]


[next article]
From: [email protected] (Court Demas)

Subject: 0.96 Test
Message-ID:
Date: 10 May 92 03:17:25 GMT
Organization: Sophomore, Math/Computer Science, Carnegie Mellon, Pittsburgh, PA
Lines: 9


I know that the testimage for v0.96 on tsx-11 was just a 'test' and
all, but I thought it might be worth mentioning that 'ps' doesn't seem
to work on my system with it (type 'ps -aux' and get a kmem error or
something). Also, I get a bunch of info-stuff printed out after I sync
the disk - I don't know what it's trying to tell me though..


-court


[next article]
From: [email protected]

Subject: X & Paradise Cards
Message-ID: <[email protected]>
Date: 10 May 92 04:27:51 GMT
Organization: Memphis State University
Lines: 4

What are the settings for the "640x400x256" mode for 256k Paradise cards???
I've tried a couple modes except, I recieve an extremely screwed up display.
Is there anyway to curtail 256 colores into, oh about 16 so I can display a
higher resolution????


[next article]
From: [email protected] (david nugent)

Subject: Re: elm? dos emul.
Message-ID:
Date: 10 May 92 00:28:55 GMT
References: <[email protected]>
Reply-To: [email protected]
Organization: Unique Computing Pty Ltd, Melbourne, Australia
Lines: 13

[email protected] writes:

> I need a reference to where to get elm. I have smail, but need a front-end.
> Please e-mail. Thanks.

Did you want me to send what I have here?


..............................................................................
david nugent Public Access Usenet "Only Nixon could go to China"
[email protected] +61-3-792-3507 - old Vulcan proverb
3:632/[email protected], 58:4100/[email protected], 199:4242/[email protected], 33:300/[email protected]
PO Box 260, Endeavour Hills, Victoria, Australia, 3802.


[next article]
From: [email protected] (Chin Fang)

Subject: clock.pas. Please archive it (was Re: clock.exe)
Keywords: X386,SVGA,Genoa,ET3000
Message-ID: <[email protected]>
Date: 10 May 92 04:26:21 GMT
References: <[email protected]>
Sender: [email protected] (Mr News)
Organization: DSG, Stanford University, CA 94305, USA
Lines: 120

This group reminds me the posting activities in comp.unix.sysv386 in the
early 1991, X386 related posts sometimes outnumbered OS types ๐Ÿ™‚ No wonder
Thomas Roell makes excusrions to this group every now and then ๐Ÿ™‚ ๐Ÿ™‚ ๐Ÿ™‚

Anyways, I happen to have a copy of clock.pas, the Turbo Pascal code
written by Thomas for X386 1.1 in 1990, on my RISC 6000 320 (Err..it's not
running linux, not yet anyway ๐Ÿ™‚ So here it is. Please archive it at linux
ftp site.

Have fun ๐Ÿ™‚

Chin Fang
[email protected]
----------------------------------------------------------------------------
(*
* Copyright 1990,91 by Thomas Roell, Dinkelscherben, Germany.
*
* Permission to use, copy, modify, distribute, and sell this software and its
* documentation for any purpose is hereby granted without fee, provided that
* the above copyright notice appear in all copies and that both that
* copyright notice and this permission notice appear in supporting
* documentation, and that the name of Thomas Roell not be used in
* advertising or publicity pertaining to distribution of the software without
* specific, written prior permission. Thomas Roell makes no representations
* about the suitability of this software for any purpose. It is provided
* "as is" without express or implied warranty.
*
* THOMAS ROELL DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
* INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO
* EVENT SHALL THOMAS ROELL BE LIABLE FOR ANY SPECIAL, INDIRECT OR
* CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE,
* DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER
* TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR
* PERFORMANCE OF THIS SOFTWARE.
*
* Author: Thomas Roell, [email protected]
*
* $Header: /proj/X11/mit/server/ddx/at386/etc/RCS/clock.pas,v 1.3 90/11/08 17:5
4:22 root Exp $
*)


USES dos;

VAR
reg: registers;
no,i: integer;
norm: longint;
clock: array [0..7] of real;

function test(c:byte): longint;
var i: longint;
j: integer;
begin
port[$03C2] := c;
i := 0;
while (port[$03DA] and $08) = $00 do;
for j:=0 to 10 do
begin
while (port[$03DA] and $08) = $08 do;
while ((port[$03DA] and $08) = $00) and (i<1000000) do inc(i);
end;
test := i;
end;

BEGIN
reg.ax := $0012;
intr($10,reg);

port[$03C4] := $01;
port[$03C5] := port[$03C5] or $20;
port[$03D4] := $11;
port[$03D5] := port[$03D5] and $7F;

norm := test($23);
clock[0] := 25.175;
clock[1] := (clock[0] / test($27)) * norm;
clock[2] := (clock[0] / test($2B)) * norm;
clock[3] := (clock[0] / test($2F)) * norm;
no := 4;

if paramcount = 1 then
begin
if paramstr(1) = 'ET3000' then
begin
port[$03bf] := $03;
port[$03d8] := $a0;
port[$03d4] := $24;
port[$03d5] := $02;
clock[4] := (clock[0] / test($23)) * norm;
clock[5] := (clock[0] / test($27)) * norm;
clock[6] := (clock[0] / test($2B)) * norm;
clock[7] := (clock[0] / test($2F)) * norm;
no := 8;
end
else if paramstr(1) = 'ET4000' then
begin
port[$03bf] := $03;
port[$03d8] := $a0;
port[$03d4] := $34;
port[$03d5] := $02;
clock[4] := (clock[0] / test($23)) * norm;
clock[5] := (clock[0] / test($27)) * norm;
clock[6] := (clock[0] / test($2B)) * norm;
clock[7] := (clock[0] / test($2F)) * norm;
no := 8;
end;
end;

reg.ax := $0003;
intr($10,reg);
writeln('CLOCK: A simple program to determine your VGA''s quartz oscillators');
writeln;
writeln('usage: clock [ET3000|ET4000]');
writeln;
writeln;
write(' detected clocks: ');
for i:=0 to no-1 do
write(clock[i]:2:1,' ');
writeln;
END.


[next article]
From: [email protected] (Mike Me)

Subject: Re: X and supported VGA cards.
Message-ID: <[email protected]>
Date: 10 May 92 02:46:16 GMT
References: <[email protected]>
Sender: [email protected]
Organization: George Washington University
Lines: 20

A quick and probably dumb question.....

I sence that speed is a major problem as far as X-display cards are
concerned.

I have a Hercules Graphics station card that uses the TI 34010 TIGA
accellerator / kick-butt etc. chip.

If there would be a TIGA driver for X, all TIGA boards would work with
it, since TIGA is a standard for all TI-340x0 boards.....

Is ther TIGA support? Is anyone working on it?
and is anyone interested in TIGA support?

I can get my hands on all the required manuals / libraries etc.
maybee a summer project if there is ineterest, and people to help
me....

Thanx,
MikeE


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: 0.96 Test
Message-ID: <[email protected]>
Date: 10 May 92 04:27:01 GMT
References:
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 20
Nntp-Posting-Host: romeo.cs.colorado.edu

In article [email protected] (Court Demas
) writes:
>
> I know that the testimage for v0.96 on tsx-11 was just a 'test' and
>all, but I thought it might be worth mentioning that 'ps' doesn't seem
>to work on my system with it (type 'ps -aux' and get a kmem error or
>something). Also, I get a bunch of info-stuff printed out after I sync
>the disk - I don't know what it's trying to tell me though..
>
>

1. Did you do a ps -U to rebuild the kernel nlist database?
2. Did you recompile ps if any header files changed?

Programs like ps, top, netstat, uptime, w, vmstat etc directly
read kmem, and examine kernel memory. If the structures are
different, things stop working. If the system uses a kernel memory
nlist file (like Linux, /etc/psdatabase, or 4.4 BSD (4.3 Reno or later) and
/var/run/kvm_vmunix.{dir,pag} this must be updated anytime a single
symbol moves (even by a single byte) or else ps, or whatever
looks at garbage.


[next article]
From: [email protected] (Craig)

Subject: X-Windows Needed Badly!!
Message-ID: <[email protected]>
Date: 10 May 92 01:25:26 GMT
Sender: [email protected] (USENET News System)
Distribution: alt.os.linux
Organization: Montana State University, Dept. of Computer Science, Bozeman MT 59
717
Lines: 18

Everyone and anyone:
I figure that this type of thread has been started many times, but I need
to ask it again. I need to get X-Windows up and running on my 386-25 PC with
Linux. If anyone can EMAIL me a response to this I'd really appreciate it. I
need to get this running as soon as possible, so if anyone could tell me if it
is even currently possible, I'd be grateful. All I need is a rough outline on
what steps I need to take to set it up, and which files are necessary.( FTP
addresses for the files would also be appreciated.) If anyone else is interested
in this, I'll post a summary as soon as I get the info. Thanks in advance!
Craig Collinson
[email protected]

P.S I currently have the bare bones Linux setup on a 40MB partition, if disk
size is important.


--------
The Only person without a .SIG


[next article]
From: [email protected] (Terry Dawson)

Subject: Re: X and 9600 bps (ANSWERED!)
Message-ID:
Date: 10 May 92 03:52:17 GMT
References: <[email protected]> <[email protected]
du> <[email protected]> [email protected]>
Sender: [email protected]
Organization: Sydney University Computing Service, Sydney, NSW, Australia
Lines: 16
Nntp-Posting-Host: extro.ucc.su.oz.au

[email protected] (Adam Goldberg) writes:

>One thing I don't know (and some of you probably picked up on this
>already) is what happens when, say, 'ls ' comes over the
>serial line. That's only 3 characters. Do you need to type 5 more
>in order for the first 3 to be recognized? I don't know how this
>is worked out. (Does anyone know?)

I believe the 16550 has a built in FIFO timer, and this timesout, generating
an interrupt so that the data can be serviced ..

Terry

--
Terry Dawson, [email protected], vk2ktj%[email protected]
+61 2 925 1556 (voice), +61 2 922 5973 (fax). __\*/__ 0^Ooooo _____


[next article]
From: [email protected] (Terry Dawson)

Subject: RTS/CTS flow control on serial ports
Summary: How do I make it work correctly ?
Keywords: RTS CTS Flow control
Message-ID:
Date: 10 May 92 03:58:46 GMT
Sender: [email protected]
Organization: Sydney University Computing Service, Sydney, NSW, Australia
Lines: 21
Nntp-Posting-Host: extro.ucc.su.oz.au

I am having trouble getting RTS/CTS flow control to work correctly with Linux.
I have a device that I must use RTS/CTS flow control (I want it to be
transparent, and its a slow packet device), and I have set it to use hardware
handshaking, and have a properly built null-modem cable, but am unsure of how
to configure the serial device to behave accordingly.

I have used:
getty /dev/ttys0
stty 2400 sane >/dev/ttys0
stty crtscts >/dev/ttys0

and all seems to work just fine, except the flow control.

What am I doing wrong ?

Terry


--
Terry Dawson, [email protected], vk2ktj%[email protected]
+61 2 925 1556 (voice), +61 2 922 5973 (fax). __\*/__ 0^Ooooo _____


[next article]
From: [email protected] (Douglas E. Quale)

Subject: Re: Password encryption bogus?
Message-ID: <[email protected]>
Date: 10 May 92 06:18:12 GMT
References: <[email protected]> <[email protected]
k.rwth-aachen.de> <[email protected]>
Sender: [email protected] (The News)
Organization: Undergraduate Projects Lab, UW Madison
Lines: 13

In article <[email protected]> [email protected] (Drew
Eckhardt) writes:
>
>UFC is also freely available, export legal, and standard. We don't
>want more than 8 characters for compatability reasons.
>

Well, anything that doesn't allow more than 8 characters is broken.
Since we have the source, we should fix it. Many Unices have more than
8 significant chars in their passwords. 16 is a nice, round number.

--
Doug Quale
[email protected]


[next article]
From: [email protected] (Denise Tree)

Subject: Need SCSI install help (a lot!)
Message-ID: <[email protected]>
Date: 10 May 92 06:14:45 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected] (Denise Tree)
Organization: The Internet
Lines: 35


Well it's gonna be an X11 Christmas.... wow

I'm having major problems installing Linux on my scsi drive, silly me,
I thought getting a st-01 would solve my problems.... The problem is I
can't get the scsi fdisk to work in order to get the partition names and
block sizes for mlkfs. Here's what I've done so far:

1. Used the dos pfdisk tp make 3 more non-dos partitions in addition
to my bootable dos partition. Pf disk command "L" then returns
ID first last
1 128 385 476
2 128 477 769
3 128 770 1022
4 6 0 384

etc. and so on.

2. So then I boot with vmunix(pre0.96) and remove some stuff from
the 95a root floppy so I can uncompress tar and do "tar xvf
/dev/at1" on the disk that I rawrote the scsi.fdisk.tar to...

3. ^^^^^^^^^^^^ (Thinking I'm pretty clever at this point) I
try out fdisk and get anly "can't open /dev/hd0" or some-
thing to that effect. That's it...no /dev/hd0....nothing else.

My next plan of action is to get the 0.12 rootimage because maybe the problem
is the dreaded "0.95a tar" problem....eh????

Could someone who has succesfully gone through this lend a hand????Please?????
...................
DEnise Tree
[email protected]^H^Het.towson.edu
(go ahead and reply with some advice...it'll give ya a nice warm friendly
feeling.. like registering shareware ( sorry for typos, its late, I'm nuts,
and this lamo comm program has onlt vt52)


[next article]
From: [email protected] (Erik Green)

Subject: Paradise settings
Message-ID:
Date: 10 May 92 05:36:04 GMT
Distribution: comp.os.linux
Organization: Long polymers in double-helix formation
Lines: 138
Nntp-Posting-Host: att2.cs.mankato.msus.edu


Time for me to make another loooong post with some mild flames,
heated opinions, somewhat useful information - you know, the usual.


In an earlier article, [email protected] (Chin Fang) writes:

>In article <[email protected]>, [email protected]>
edu writes:
>|> What are the settings for the "640x400x256" mode for 256k Paradise cards???
>|> I've tried a couple modes except, I recieve an extremely screwed up display
>|> Is there anyway to curtail 256 colores into, oh about 16 so I can display a
>|> higher resolution????

>The following simple calculation shows that 256K is NOT sufficient for the
>resolution that you desire

>640x480/1024 = 300K > 256K

[Tangent begins here]

Actually, this assumes 8 bits per pixel. For a Sierra hicolor, which is
15 bits per pixel(stop me if i'm wrong), you need twice the memory for
this res in 15-bit mode. OTOH, "standard" 640x480x16 mode uses 4 bits
per pixel, and therefore 153600 bytes of video memory. Theoretically,
a "standard" VGA card w/256k(262144 bytes) of video RAM could do either
800x600x16 (240000 bytes) or 640x400x256 (256000 bytes), assuming you
had software capable of setting those modes, perhaps by directly
programming the video hardware (which we have to do anyway in protected mode).

[Tangent ends]

>Please read line # 427 ~ 430 of X386's CONFIG.

Please read line #1 of the post you quoted. He's asking for 640x400x256,
not 640x480.

I, as I have posted before, am running at 640x400 with a cheap card that has
the paradise chipset. I don't remember the clock settings offhand, but I
do know it's the only 640x400 entry in "vga.dbase". It probably could
be improved a little, because the picture is just a little to the right
of center on the screen.

Also, the original X386 software is designed to use cards in
256 color mode only, and then in a packed format in video memory. To
port/create a driver for a card that uses the packed format and has a
256-color mode(higher res than 320x200 would be sane) _and_ has a
good set of docs for its chips should be "easy" - you just have to
re-create a finite set of functions that deal with the card.

To do a driver for a card that doesn't use the packed format would
require re-writing parts of the X386 software not directly related
to the driver.

Zorst: Could you ship X eventually in a .a format, and allow people
to link in their own video drivers, like jwinstead is going
to do for MGR?

>I would like to encourage ANYONE who is new to X386 carefully study
>the very informative X385.man by Thomas Roell for MIT X11R5. If you
>don't have it. It's available for anon ftp from export.lcs.mit.edu
>cd /pub/R5untarred/mit/server/ddx/x386. Then simply get it.

Good idea.

-------------------------------------------------------------
> SuperVGA Cards Max Res ChipSet
> -------------------------------- -------- -------
> Compuadd Hi-Rez card w/1meg 1024x768 ET4000
> Diamond SpeedStar 1024x768 ET4000
> EIZO MD-10 800x600 ET3000
> GENOA 5300/5400 800x600 ET3000
> GENOA 6400 800x600 GVGA
> Optima Mega/1024 1024x768 ET4000
> Orchid ProDesigner 800x600 ET3000
> Orchid ProDesigner II/1024 1024x768 ET4000
> Paradise VGA Professional 640x480 PVGA1A
> Paradise VGA 1024 640x480 WD90C00
> Sigma Legend 1024x768 ET4000
> STB PowerGraph w/1meg 1024x768 ET4000
> Swan SVGA equipped with VCO chip 1024x768 ET4000
> TRICOM Mega/1024 1024x768 ET4000

AND probably also these chipsets in 640x400 mode.(Paradise PVGA1A confirmed)

----------------------------< Clip and Save>----------------------------------



Here's a simple test under MSDOS to see if you have the above chipsets:

DEBUG[Enter]
-d c000:0000

... and read the text you see on the right side of the screen. Usually it's
a copyright and the Chipset ID.(This is the right BIOS location for most
cards, for some you may have to look around)

If you don't have one of the above chipsets, wait. There's enough people
pissed off that they can't run X that there'll be more drivers showing up
in a few weeks.

Notes:

1) The X software can run on a great many hardware platforms. Don't
however, expect that someone will spend time porting drivers for
CGA or EGA, at least in the near future. Hercules, maybe - the extra
res and cheap hardware are nice pluses.

2) As long as you can tolerate it, _no_ speed is too slow. All those
people saying X is intolerable on anything less than system NNNN with
resolution NNNN are making subjective judgements which I'll say are
true IF you have access to another machine that runs X better.(Whew)

Not just access, actually. Total control. Most people are running
Linux on their home machines, and having X at home that you can do
whatever you want to with outweighs the speed and resolution consider-
ations.

Finally, keep in mind that this release of X was supposed to be a limited
audience release - future editions should be a little more robust.

-Erik

PS: To the guy who posted that X started and made a bunch of white lines and
other garbage on the screen: mine does this too. About thirty seconds
later it changes to the gray basketweave that is standard, and thirty
after that it goes to slate blue.
"Sometimes, ya just gotta wait"

PPS: Anybody want to trade me a decent video card for my old 286/12
motherboard? It's got 1MB 80ns/Award BIOS/Scores 3333 Dhrystones/
norton 9.2.

--
Erik "Longshot" Green
[email protected](vax1 | att2.cs | krypton | theory.cs).mankato.msus.edu
"Ash nazg durbatuluk, ash nazg gimbatul, ash nazg
thrakatuluk agh burzum-ishi krimpatul!"


[next article]
From: [email protected] (Orest Zborowski COMP)

Subject: Re: X Problems..
Message-ID: <[email protected]>
Date: 10 May 92 06:11:41 GMT
References: <[email protected]>
Organization: Printer Products Eastman Kodak
Lines: 90

[email protected] (Mark Holden) writes:
>In article , [email protected]
.FI (Mika Pekka Liljeberg) writes:
>|> sure I did it right, too. I can easily get a stable picture, but instead
>|> of a single beautiful one I get two ugly ones, side by side. Both contain
>|> the visible desktop, only squeezed in half along the X axis. I suspect
>|> it's the Sierra DAC chips that cause this behaviour. Can anyone confirm
>
>I doubt it's the Sierra DAC, the same thing has happened to me with Diamond
>Speedstar w/o Sierra DAC, when I had the incorrect configuration in Xconfig.
>With the Diamond Speedstar+ and the Gateway 2000 Crystal Scan monitor, it's
>a simple process to set up X386, and I'd done it before under Mach anyhow ๐Ÿ™‚
>The important thing is to get the correct clocks for your card, the ones that
>are included in the Xconfig probably aren't right for yours. To see what
>they are (as far as X386 is concerned) take the "clocks" linne out of Xconfig,
>and run xinit, preferably from a login on one of the serial ports (I used my
>Linux machine to set up my Mach machine, then reversed the process later ๐Ÿ™‚
>then make sure you have a line in modedb that corresponds to one of the clocks
>that X386 shows (probably the last one).

that's right. it's probably up there on the x386 faq! i hoped i was clear
enough in the readme that you need to examine the xconfig very carefully.
the clock recognition in linux's x386 isn't stable because i had no clean
way to disable interrupts while the detection was going on. that's why i
provided the vga.dbase file with all the clocks. i'm afraid you have to

1. run x386 by itself, redirecting stderr to a file (i.e. x386 2>/tmp/OUT)
and then exit it to see what clocks it reads (after commenting out the
clocks in the provided xconfig)

2. run the clock.exe program hlu posted, which is much more stable because
there are no context switches in dos

3. lookup your card in the vga.dbase and plug it in, then try various
configurations until one works.

even after you get what you think are the right clocks, you still may not
get a stable display. this is because there could be multiple entries for
each resolution, with different clock values. since x386 chooses the last
one from the list (not the first, as advertised), that may well be wrong
(it was for me). so simply comment out the extras. this is also true for
entries with the same clock but different timings, as for the 1024x768 mode.

try with the lowest clock frequencies (25mhz) and at 640x480. this is
probably the safest bet (i ran this way for a while until i discovered
the multiple timing problems above) to get things to work. even at this
resolution you can still have a virtual space thats 1152x900, and pan
around.

>
>|> And then there is the mousie, of course. ;-( I've got a Genius mouse,
>|> which can act either as a Microsoft mouse or a MouseSystems mouse.
>|> It works, too. Not under X386, though. In Microsoft mode the mouse
>|> cursor gets whopping mad, moving in quick jerks in the same general
>|> direction I'm pushing the mouse. I've got no control whatsoever over
>|> the buttons. X seems to think someone's bouncing on them, though. ๐Ÿ˜‰
>|> In MouseSystems mode I get no response at all. Is the X mouse driver
>|> picky, or what? ๐Ÿ˜‰
>
> I've run into severe wierdness with mice under x386. On Mach, my
>Microsoft Mouse works _almost_ correctly, except that it doesn't recognise
>a button press until you move the mouse, and then doesn't notice that the
>button is being held down anyhow. The same behavior exists under Linux,
>except that it does recognise that the button is being held down after you
>move the mouse. So, I can use my mouse under Linux, but I had to punt back
>to X11R4 under Mach . . . Also, I've tried a Logitech Mouseman, with
>absolutely no luck whatsoever.

with two button mice x386 needs to do three button emulation (there is a
line in xconfig for that too). inside the server there is a state machine
which treats a single press as "ok, now what? the other button, to make the
missing middle one? or something else?". when you move the mouse, or release
it, it gives up waiting for the "middle" button. i guess there could be
a small timeout which could say "enough already, its a press". since i have
to contend with this, i'll look into how it can be done.

as for other mice, i don't have them, but i know you need to be careful about
the baud rate and bits-per-sample. i thought the kernel default was to
use cs8, passing 8 bit, but that may be incorrect, or may be modified if
you're sharing the line with a modem, etc. the server only uses cs7 with
the microsoft mouse, but the kernel doesn't honor this, so i plugged in
an istrip for this case (only if cs7 was asked for). one person wrote to
me and said that turning off istrip made his mouse work, so you may want
to try "stty -istrip >/dev/ttys2" in your startx script. if this is the
case, i may add the logic to turn off istrip if its cs8 mode.

zorst
[[email protected]]
--
zorst (orest zborowski)
[reply to [email protected]]


[next article]
From: [email protected] (Orest Zborowski COMP)

Subject: Re: X, ET4000 and mice
Message-ID: <1992May10.062117[email protected]>
Date: 10 May 92 06:21:17 GMT
References: <[email protected]>
Organization: Printer Products Eastman Kodak
Lines: 52

[email protected] (Peter MacDonald) writes:
>
>Contrary to popular belief, having an ET4000 doesn't automatically
>let you run X386. I have seen reports of some ET4000 cards that just
>don't work.
>
>Myself, I set up X under DELL UNIX at work using an ET4000 Ammazing card:
>no problem. Now I am trying it at home with an ET4000 vga2max card with Linix.
>Using the default "640x480" mode, it gives me four images on the screen,
>complete with four pointers, four xterms, etc. I have mucked around for 4-5
>hours trying different monitor settings, but no joy. It seems that Clock must
>be wrong (half?) or something. I also found a 800x600 mode that gives me
>6 windows! The card seems to have something called Hicolor mode (32K colors
>I think).

ah! but i already thought of this! one of the few nonstandard patches i
have applied to the server is the hiclock patch (which was discussed in
comp.unix.sysv386). if you find your clock halues are high, try adding
the line
vendor "hiclock"
in the vga256 section. i'd be interested to know if this works.

>
>Also, I have one of those switchable mouse (ACE). The logitech setting
>works not at all, while with the MS setting, the buttons work but the mouse
>movement doesn't. Using another alternate mouse, a logitech only (2button)
>causes eratic movement.
>
>Fortunately, I am just borrowing this equipment from work, so am not
>saddled with it. But, the moral is: If you think just getting an
>ET4000 card will let you run X with no problems, maybe.
>Configuring the monitor section alone can be very tough. Anyone know
>why you can install a VGA card under DOS, and it can display just fine
>without knowing anything about the monitor? The card just uses various
>Video Clock, Horz Sync, and Vert Sync combos. If the monitor doesn't
>support it, you get the "Outer Limits" on screen.
>
>But not X386. I like flexibility, but I think maybe a more usable
>configuration would be one that just lets you supply the Clock/VSync/HSync,
>and it would try some intelligent monitor timings. Actually, have tried
>going through the video tutorial by Chin Fang. Even tried developing
>a spreadsheet under oleo to do the calculations. But the real solution
>is to use C code, and then incorporate it into a server as a configuration
>option.

what a great idea! let me know when you're done! ๐Ÿ™‚

zorst
[[email protected]]
--
zorst (orest zborowski)
[reply to [email protected]]


[next article]
From: [email protected] (Orest Zborowski COMP)

Subject: Re: Subgroups?
Message-ID: <[email protected]>
Date: 10 May 92 06:27:56 GMT
References:
Organization: Printer Products Eastman Kodak
Lines: 33

[email protected] (Magnus Alvestad) writes:
>Would it be an idea to create a few subgroups to comp.os.linux so as
>to separate posts concerning X11, mgr (might be some volume in a few
>weeks) and maybe some other topic I can't think of? SCSI?
>
>Btw: I am coordinating effort on making screen-card drivers for X386, mail
>[email protected] if you are making a driver for a new card,
>want to help making a driver, want to know if someone else is working
>on this or that card.
>
>-Magnus
>
>

the "official" channel of communications on x11 is with the mail server.
thats where i post all tech stuff and updates. i want to avoid posting
to the world about all sorts of x questions, etc, esp while i still
consider x386 unfinished.

to join, send mail to [email protected] with the
line
X-MN-Admin: join x11
in the body of your text.

>
>
>
>--
>| Magnus Alvestad | I am a prolog program. I get all my opinions
>| [email protected] | from a small organization in Birmingham.
--
zorst (orest zborowski)
[reply to [email protected]]


[next article]
From: [email protected] (Orest Zborowski COMP)

Subject: Re: X386: R4 or R5?
Message-ID: <[email protected]>
Date: 10 May 92 06:32:57 GMT
References: <[email protected]>
Organization: Printer Products Eastman Kodak
Lines: 20

[email protected] (Doug Evans) writes:
>Is the X386 that has been ported to Linux R4 or R5?
>Or, put another way, is it X386 1.1b or 1.2 (or ?) ?

the linux x386 port was based on the stock distribution from x11r5, from
mit and was done by me, not thomas, although in terms of quantity of
effort, his was more inspiration while mine was more perspiration ๐Ÿ™‚

>
>Secondly, Thomas, are the vga drivers architected the same in 1.2
>as they are in 1.1b? Or, put another way, if I have a driver for 1.1b,
>how usable is it with 1.2? How much work is involved in changing a 1.1b
>vga driver to a 1.2 critter?

diffs for the linux distribution are available on banjo.concert.net, and
they don't touch much of the meat of the vga drivers, except to change
all asm global names from foo to _foo.
--
zorst (orest zborowski)
[reply to [email protected]]


[next article]
From: [email protected] (Court Demas)

Subject: Long Filenames
Message-ID: <[email protected]>
Date: 10 May 92 07:52:33 GMT
Organization: Sophomore, Math/Computer Science, Carnegie Mellon, Pittsburgh, PA
Lines: 9


What's the deal with the filename-length limit under Linux? It feels
weird having to rename so much stuff just so I can get a ".tar.Z" in
there. Is there any way to get the full length filenames like most
other Unixes (and OS/2 :-))???


still trying to get X up,
-court


[next article]
From: [email protected] (H.J. Lu)

Subject: Re: X, ET4000 and mice
Message-ID: <[email protected]>
Date: 10 May 92 07:40:07 GMT
References: <[email protected]> <[email protected]
.com>
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 21

In article <[email protected]> [email protected] (Orest Zbor
owski COMP) writes:
>[email protected] (Peter MacDonald) writes:
>>
>>But not X386. I like flexibility, but I think maybe a more usable
>>configuration would be one that just lets you supply the Clock/VSync/HSync,
>>and it would try some intelligent monitor timings. Actually, have tried
>>going through the video tutorial by Chin Fang. Even tried developing
>>a spreadsheet under oleo to do the calculations. But the real solution
>>is to use C code, and then incorporate it into a server as a configuration
>>option.
>
>what a great idea! let me know when you're done! ๐Ÿ™‚
>
>zorst
>[[email protected]]
>--

I believe somebody posted it a while ago on comp.unix.sysv386. Just aak
it over there.

H.J.


[next article]
From: [email protected] (Joe Waters)

Subject: X and Logitech mice : RESOLVED
Message-ID:
Date: 10 May 92 09:43:31 GMT
Organization: The University of New Mexico, Albuquerque, NM
Lines: 42

Okay. We finally got a Logitech mouse to work with our X system, so I thought
I'd share our experiences with you... We are using a Logitech 3 button Serial,
hooked up to /dev/tty65... Yesterday (I think) Krishna posted the mouseinfo.c
program to one of the linux groups. This is a program which tells some mice to
go into mouse systems 5bit compressed data format from whatever they were using
(on a Logitech, it defaults to 3bit compressed). After much fighting and arguin
g
with X, we found that this actually works with a Logitech - the problem is, if
you call it when you're not in X and then start X up, the X mouse driver confuse
s
the mouse so that it doesn't work. Krishna addressed this in the short intro to
his source where it says "run this from in an xterm" - the problem is, without
a working mouse, we had no way to START UP an xterm... The solution is actually
really simple - add this program to the beginning of the /usr/lib/X11/xinit/xini
trc
file - it will execute once X has already "corrupted" (for lack of a better
word) your logitech mouse. The only problem here is that the actual source
spews stuff to stdout - the problem is that that hoses the X display at the
bottom of the screen. Once again, the solution is simple - reroute the output
to /dev/null or some file if you REALLY REALLY want to keep the output ๐Ÿ™‚
So the command in your xinitrc file looks like this:

/PATH/mouseinfo > /dev/null

which will work. The mouse settings that correspond to this from the
Xconfig file are

mousesystems "/dev/tty65"
BaudRate 1200
# Samplerate 150
#emulate3buttons

Do this and everything works just peachy.

Linus, obz, and everybody else involved in this thing, you
did an excellent job. I never thought I would see X running
fully on a 386 with good speed. Mondo huge kudos.

Joe
--
--
/-------------------------+-------------------------------------------------\
| Joe Waters ("Falc") | "Most any problem can be solved with duct tape |
|[email protected]| and Froot Loops . . . or a good heavy club." |
\-------------------------+-------------------------------------------------/


[next article]
From: [email protected] (Thomas Gardner)

Subject: What about EISA?
Message-ID: <[email protected]>
Date: 10 May 92 09:11:12 GMT
Sender: [email protected]
Reply-To: [email protected] (Thomas Gardner)
Organization: Case Western Reserve University, Cleveland, OH (USA)
Lines: 38
Nntp-Posting-Host: cwns16.ins.cwru.edu


I am what you would call VERY new to Linux (I just got some ZIP files from
a buddy who thought I might be interested about 16 or so hours ago, and
have been reading ever since). I haven't even been home to try them out
yet!

Anywho, here is the gist of the story. I have just bought (hasn't even
been delivered yet) a 486 with an EISA bus as a play-toy (hopefully some
consulting work as well), and I sure would like to get Linux running on it
- it sounds like a WONDERFUL OS, as long as I don't run into quite as many
problems as I have been reading about. Unfortunately, (although I didn't
think so at the time I bought it) it has a EISA SCSI controller for the
hard drive! I am not what you would call a wealthy man, so I am sure that
I cannot be the only one in the world who would like to get something like
this rockin'-n-rollin'. If so, I certainly won't be for too long!

I sure would hate to have to put my current MINIX stuff onto it now that I
know that something so much nicer is available. This is especially true
since I was planning on dropping a couple of my current 30 meg segates into
it and just living with the performance drag until I could get a device
driver written for the SCSI drive.

I would think that even this would be a reasonable interim fix for Linux.
Has anyone tried porting some of the MINIX drivers over? I know this sort
of thing couldn't be released to the net without Andy's permission, but it
sure would be nice to know that it would not be too tough to do.

What's the deal??? Is anyone working on EISA stuff??? Is anyone planning
to???

If not, I am a software engineer for Tecmar (tape drives, memory boards,
laservaults, etc.) thus, I am at least mildly familiar with drivers and the
like, and I would be willing to help move things along, but having never
dealt with this particular issue, I would need some help!

Tanx in advance for any responses.
- Thom.


[next article]
From: [email protected] (Chin Fang)

Subject: Re: X, ET4000 and mice
Message-ID: <[email protected]>
Date: 10 May 92 09:22:56 GMT
References: <[email protected]> <[email protected]
.com>
Sender: [email protected] (Mr News)
Organization: DSG, Stanford University, CA 94305, USA
Lines: 79

In article <[email protected]>, [email protected] (Orest Zbo
rowski COMP) writes:
|> [email protected] (Peter MacDonald) writes:

|> >But not X386. I like flexibility, but I think maybe a more usable
|> >configuration would be one that just lets you supply the Clock/VSync/HSync,
|> >and it would try some intelligent monitor timings. Actually, have tried
|> >going through the video tutorial by Chin Fang. Even tried developing
|> >a spreadsheet under oleo to do the calculations. But the real solution
|> >is to use C code, and then incorporate it into a server as a configuration
|> >option.

|> what a great idea! let me know when you're done! ๐Ÿ™‚

I whole heartedly agree the timing stuff should be done in a manner that's
transparent to general users. But I will tell a bit of background why the
tutorial was written in the manner it is ...

The way I look at the timing calculation is basically a mult-objective
optimization problem. (Let's use non-interlace and multisync monitors for
discussion purpose) It is because

(1) if you want high resolution => you get low refresh rate
(2) on the other hand, if you want high refresh rate => you can't get
high resolution

So, some compromise has to be reached (thus multi-objective), and this
compromise is *really* subjective, varying from person to person, quite
tough to meet using a program ๐Ÿ™

And, there are *real* constraints as well. The biggest constraint usually is
a monitor's horizontal sync upper limit, this is the main constraint for most
people. Normally most people have a video adapter which has far more capability
than his/her monitor (primarily due to cost). In addition, some early SVGA
cards using crystals don't have certain desirable dot-clocks, and that's
another constraint. Third one is the available display memory, which is
getting less important as 1MB equipped SVGAs and better are becoming more
common. There are few others that I will not elaborate here.

In short, this is a classical constrained optimization problem. One way
to solve it is to use Danzig's simplex method [a] or a simplex method
with BFGS update nonlinear programming [b] afterwards for polishing (geez,
what a fussy guy :). However, the big trouble I had (and still having!) is
that there is no way I can 100% sure about the constraint values that *my
program* gets. That is, it's not that hard to write a C code,
probing the hardware, and obtaining constraints discussed above. But
without extensive testing and validation, the timing suggested by such a code
would be almost untrust-worthy as a random guess as soon as it is used on
a unknown hardware. Thus it would be really imprudent to release it.

Therefore, I often look at the descriptions (in fact, basic algorithm in
disguise) in the timing tutorial with regret. Simply because mathematically
it's a very easy problem, but it is also ill-posed due to the fact that part
of the problem formulation data are not guaranteed to be i) available and
ii) accurate. Given the situation of the commodity PC market, I doubt such
a code would work for many ๐Ÿ™

David Wexelblat at AT&T Bell Laboratories started collecting timing info
for various hardware a while back, primarily in comp.unix.sysv386. Early
on I naively thought that soon this database would grow very significantly
and thus I would have a good base to train a neural network code [c],
which, after trained using data in this database, would provide a good guess
(more precisely, interpolation, as neural network <=> glorified data
interpolator). But after months, even the latest version of X386 timing
data base is not as extensive as I would like to see. So another regret...

The problem is still nagging me however ๐Ÿ™‚ ...

Have a nice weekend.

Chin Fang
[email protected]
-----------------------------------------------------------------------------
PS. yes, this is a nerdy post. But I believe some people must be thinking
more or less along the same line, so I list two references below

[a] D.G.Luenberger, Linear and Nonlinear Programming. Addison-Wesley,
Reading, Mass, 2nd, 1984, Chapter 3.
[b] op.cit, Chapter 9.
[c] Burnod, Adaptive Neural Network. Prentice Hall. 1992


[next article]
From: [email protected] (BURNS)

Subject: Re: elle(1) and rc(1) ports uploaded to tsx-11.mit.edu
Summary: link errors resolved
Message-ID: <[email protected]>
Date: 10 May 92 08:13:59 GMT
References: <[email protected]> <[email protected]>
Organization: Georgia Institute of Technology
Lines: 198

I found out why I was getting undefineds from elle's make. There was a
syntax problem with the '-sh -c if ...' line in Makefile that prevented
ranlib from processing the made libraries. That, plus adding -ltermcap to
the link line solved everything. To remake things in mid-stream, I also had
to reorganize things in sbmake. New make files follow:

---------
Makefile:
---------
# ELLE Makefile.
# The main trickery to notice is that this Makefile invokes itself
# in order to build ELLE. This allows it to generate sections of
# the makefile dynamically, namely:
# makecf.rl - defines the RANLIB macro to use "ranlib" if system has it
# makecf.fun - defines the FUN_OFILES macro to specify which modules
# are needed in order to furnish all of the desired
# functions for a specific ELLE configuration.

# Basic definitions

CFLAGS = -c -O

CONFS = makecf.rl makecf.fun defprf.c eefdef.h eefidx.h

CORE_OFILES = eemain.o eecmds.o eesite.o eevini.o\
eedisp.o eeterm.o eeerr.o eeques.o\
eebuff.o eefile.o eefed.o eeedit.o eebit.o

.c.o:
cc $(CFLAGS) $*.c
ar rv elib.a $*.o

# ------------------------------------------------------------------
# Standard ELLE configuration: "elle".
# Default if no target given to make.
# Note that "xelle" is created, not "elle".

elle: $(CONFS)
make -f makecf.rl -f makecf.fun -f Makefile B-elle

B-elle: $(CORE_OFILES) $(FUN_OFILES) elle.h eesite.h libsb.a
$(RANLIB) elib.a
cc -o xelle eemain.o elib.a libsb.a -ltermcap


# Don't flush these files if interrupted, dammit!

.PRECIOUS: ellec deffun.e defprf.e $(CONFS) libsb.a

# Configuration setup stuff, for files listed by CONFS.
# If any of the files is missing or outdated, all must be
# re-generated. We assume the *.t files always exist.
# A forced re-compilation of eecmds.c must also be done to
# ensure that the latest .h files are included in ELLE.

makecf.fun defprf.c eefdef.h eefidx.h : ellec defprf.e deffun.e
cat deffun.e defprf.e | ./ellec -CMconf > makecf.fun
cat deffun.e defprf.e | ./ellec -Pconf > defprf.c
cat deffun.e defprf.e | ./ellec -Fconf > eefdef.h
cat deffun.e defprf.e | ./ellec -FXconf > eefidx.h
rm -f eecmds.o

# The following modules make use of eefidx.h and thus must also
# be recompiled if the configuration is changed.

eebuff.o eeerr.o eehelp.o eejust.o eemain.o eeques.o eef1.o : eefidx.h
cc $(CFLAGS) $*.c
ar rv elib.a $*.o


# RANLIB definition stuff. The idea here is to automatically check to see
# whether a system has the "ranlib" program, and set up the RANLIB
# macro appropriately (use ranlib if possible; else substitute a dummy
# program, "echo" in this case). The existence of "makecf.rl"
# indicates that this setup code has been executed; it should never
# need to be done again.
makecf.rl:
echo "RANLIB = echo" > makecf.rl
-sh -c "if [ ranlib ]; then (echo \"RANLIB = ranlib\" > makecf.rl;) fi"

# ELLE profile compiler. Needed to generate makecf files!
# Although eefdef.h and defprf.c are included by ELLEC, they
# are not listed as dependencies in order to avoid loops (see
# their target entries). That is OK because their information is not
# used when generating the makecf files; it only furnishes default
# values needed when an ELLE user compiles a user profile.

ellec: ellec.c
cc -o ellec -O ellec.c

# SB library, used by ELLE.
# This target should not be directly invoked by the user since it
# requires that RANLIB be defined; for direct invocation, better to
# just do a MAKE of "sb" in the sbmake file.

libsb.a:
make -f sbmake sbnoran
$(RANLIB) libsb.a

# ------------------------------------------------------------------
# ELLE Variants (system or configuration dependent)

# SUN workstation system/configuration.
# Needs extra libraries to support window hacking; libsb.a comes last
# so that "valloc" refs can be satisfied from SB library instead of
# C library.
# Plus special patch to binary, very installation-dependent!

sunelle: $(CONFS)
make -f makecf.rl -f makecf.fun -f Makefile B-sunelle

B-sunelle: $(CORE_OFILES) $(FUN_OFILES) eesun.o elle.h eesite.h libsb.a
$(RANLIB) elib.a
cc -o xelle -u _main elib.a -ltermlib\
-lsuntool -lsunwindow -lpixrect\
libsb.a
echo "tool_select+148?W my_select" > esunpat.sh
adb -w xelle < esunpat.sh

# APOLLO system.
# Linker doesn't understand libraries as regular arguments.
# May have troubles with duplication of CORE and FUN files.
# System has no termlib - uses TERMCAP emulation that comes with AUX.

SB_OFILES = sbstr.o sbm.o sberr.o

apolloelle: $(CONFS)
make -f makecf.rl -f makecf.fun -f Makefile B-apolloelle

B-apolloelle: $(CORE_OFILES) $(FUN_OFILES) elle.h eesite.h libsb.a
cc -o xelle $(CORE_OFILES) $(FUN_OFILES) $(SB_OFILES) -ltermcap

# IBM PC/IX system.
# Just needs -ltermcap instead of -ltermlib.
# Does not have/need "ranlib" by the way.

pcixelle: $(CONFS)
make -f makecf.rl -f makecf.fun -f Makefile B-pcixelle

B-pcixelle: $(CORE_OFILES) $(FUN_OFILES) elle.h eesite.h libsb.a
cc -i -o xelle -u _main elib.a libsb.a -ltermcap

---------
sbmake:
---------
# SB library makefile.
# Only funny stuff is handling for the "bcopy" routine, where
# we try to use the system version if any exists.
#
# On PDP-11 systems only, BCOPYSUF can be set to ".s" to use
# an assembly-language version. The bcopy.s here is
# better than the BSD2.9 version and can replace it.

CFLAGS = -c -O

OFILES = sbstr.o sbvall.o sbm.o sberr.o sbbcpy.o

BCOPYSUF = .c

.c.o:
cc $(CFLAGS) $*.c
# ar rv libsb.a $*.o

# Default entry - build library and attempt ranlib, but since some
# systems don't have ranlib, ignore error in latter.
sb: libsb.a $(OFILES)
-sh -c "ranlib libsb.a"

# Note that ELLE's makefile invokes this entry, since it does the ranlib
# itself if necessary.
sbnoran: libsb.a $(OFILES)

libsb.a: $(OFILES)
ar rv libsb.a $(OFILES)

# BCOPY special-case handling. If system already appears to have
# a "bcopy" routine, we use that for best efficiency (normally it
# is written in assembler to take advantage of things like
# block move instructions).
# Otherwise, we use our own C-language version.

sbbcpy.o: sbbcpy$(BCOPYSUF)
cc -c -O sbbcpy$(BCOPYSUF)
rm -f a.out bnull.c bnull.o bfind.c bfind.o bfind
echo "sbm_null(){}" > bnull.c
cc -c bnull.c
echo "main(){exit(0);bcopy();}" > bfind.c
-cc -o bfind bfind.c
-sh -c "if ./bfind;\
then echo Using system bcopy; (mv bnull.o sbbcpy.o)\
else echo Using SB bcopy; fi"
rm -f bfind bfind.c bfind.o a.out bnull.c bnull.o
ar rv libsb.a sbbcpy.o
--
BURNS,JIM (returned student)
Georgia Institute of Technology, 30178 Georgia Tech Station,
Atlanta Georgia, 30332 | Internet: [email protected]
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt0178a


[next article]
From: [email protected] (James Henrickson)

Subject: Re: password
Message-ID: <[email protected]>
Date: 8 May 92 23:45:27 GMT
References: <[email protected]> <[email protected]
cso.uiuc.edu>
Organization: State University of New York -- Institute of Technology
Lines: 24

[email protected] (THE Upholder of Truth) writes:

>In answer to the above question, it's in the FAQ (part2) and the binaries
>are located on most FTP sites in a file called shadow-bin.tar.Z (I found
>it on tsx-11.mit.edu).

>However, after messing with these binaries for about 6 hours now, I have
>yet to get useradd to properly add users. it *CLAIMS* that the users
>are added, and the /etc/passwd file is modified, but neither login nor
>usermod or userdel will recognize that the user was added (and the passwd
>binary modifies the /etc/passwd and /etc/shadow files, but not properly...
>I sets a passwd, but that passwd won't let me in.. I had to reinstall from
>scratch 'cause I didnt' test it from another virtual term...)

Like so many others, I got bitten by shadow-bin.tar.Z. Keeping a boot floppy
and file system saved me from a lot of horror. I found a few goodies in
misc.tar.Z, part of mcc-interim. I now have passwords, but in the old manner,
not shadow passwords. I find it much better than no passwords, though. ๐Ÿ™‚

--
James L. Henrickson, Software Engineer * "machine learning for
Critical Technologies Inc. * autonomous vehicles, digital
311 Turner St Suite 303 * signal processing & parallel
Utica, NY 13501 [email protected] * computing"


[next article]
From: [email protected] (H.J. Lu)

Subject: This is for Orest
Message-ID: <[email protected]>
Date: 10 May 92 07:49:56 GMT
Sender: [email protected] (USENET News System)
Distribution: usa
Organization: Washington State University
Lines: 15
Originator: [email protected]

Hi, Orest,

I tried to send you mails several times. I am not sure they reached you
or not. This is my last resort.

I think I have fixed the problem with the shared libs for X11. Please
get the latest gcc 2.1. The shared libs for X11 now are in it. It is in
the usual places and still uses old LSL prefix.

BTW, which X11 lib will be always linked, beside libX11.a, if some X11
libs other than libX11.a are linked? Last time I chose libXt.a. It
turned out I was wrong. This time I chose libXmu.a. I hope I am right.
Please correct me if I am wrong. It is very important.

H.J.


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: RTS/CTS flow control on serial ports
Keywords: RTS CTS Flow control
Message-ID: <[email protected]>
Date: 10 May 92 10:14:54 GMT
References:
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 18
Nntp-Posting-Host: romeo.cs.colorado.edu

In article [email protected] (Terr
y Dawson) writes:
>I am having trouble getting RTS/CTS flow control to work correctly with Linux.
>I have a device that I must use RTS/CTS flow control (I want it to be
>transparent, and its a slow packet device), and I have set it to use hardware
>handshaking, and have a properly built null-modem cable, but am unsure of how
>to configure the serial device to behave accordingly.
>
>I have used:
>getty /dev/ttys0
>stty 2400 sane >/dev/ttys0
>stty crtscts >/dev/ttys0
>
>and all seems to work just fine, except the flow control.
>
>What am I doing wrong ?
>

Linux pre-.96 or before does not support hard flow control.


[next article]
From: [email protected] (Magnus Alvestad)

Subject: Re: X and supported VGA cards.
Message-ID:
Date: 10 May 92 13:10:49 GMT
References: <[email protected]> <[email protected]
edu>
Sender: [email protected] (Bergen University Newsaccount)
Reply-To: [email protected]
Organization: AlveSoft
Lines: 10
In-Reply-To: [email protected]'s message of Sun, 10 May 1992 02:46:16 GMT

I am coordinating the effort on making more graphics card drivers for
X386 available. If you are working on a card or want to know who are,
mail me at [email protected]

As far as I know no one is working on any TIGA boards.

-Magnus
--
| Magnus Alvestad | I am a prolog program. I get all my opinions
| [email protected] | from a small organization in Birmingham.


[next article]
From: [email protected] (Simon Marlow)

Subject: Re: 0.96 Test
Message-ID: <[email protected]>
Date: 10 May 92 14:12:16 GMT
References:
Organization: Glasgow University Computing Science Dept.
Lines: 20

[email protected] (Court Demas) writes:


> I know that the testimage for v0.96 on tsx-11 was just a 'test' and
>all, but I thought it might be worth mentioning that 'ps' doesn't seem
>to work on my system with it (type 'ps -aux' and get a kmem error or
>something). Also, I get a bunch of info-stuff printed out after I sync
>the disk - I don't know what it's trying to tell me though..


>-court

The ps error is because the places where ps gets its information from
will have changed with the new kernel. Normally you would do a `ps U`
to update the ps resource file, but you need the System.map file
created during kernel compilation for this. Best bet is to wait till
the source (for the kernel) becomes available and compile it yourself,
or let somebody else do it ๐Ÿ™‚

Cheers, Simon


[next article]
From: [email protected] (Theodore Ts'o)

Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <[email protected]>
Date: 10 May 92 16:14:57 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 42


From: [email protected] (Britt Park)
Date: 9 May 92 16:54:04 GMT

As to the requirement that complete sources to a binary on an
archive site must be located on the same archive site; I don't
believe this as any basis either, at least not as the GNU license is
currently phrased:

"You may copy and distibute the Program ... in object code or
executable form under the terms of Paragraphs 1 and 2 above provided
that you also do one of the following:
... Accompany it with the information you received as to where the
corresponding source code may be obtained."

That's GPL version 1. GPL version 2 is not so clear; the same section
in GPL v. 2 states:

Accompany it with the information you received as to the offer
to distribute corresponding source code. (This alternative is
allowed only for noncommercial distribution and only if you
received the program in object code or executable form with such
an offer, in accord with Subsection b above.)

Grr.... I have sent mail to [email protected], asking him to reconsider
the matter. In the meantime, I suggest that people who want to put
stuff under the GPL carefully read both the version 1 and version 2, and
consider which one you want to place your program under, if any.

Certainly I believe it's better for archive sites to make available
complete sources rather than just diffs. This is not, however, always
practicable. It costs a great deal in resources to maintain an
archive site, and GNU software, as we all know, weighs in on the hefty
side. I'd prefer to see archive sites with binaries only rather than
archive sites which fold because they don't have the space to hold
all the sources.

This is my concern also. I'm not sure what can be done, however.
(Except for stating that you wish your program to be placed under GPL
version 2.)

- Ted


[next article]
From: [email protected] (Rob Coleman)

Subject: Bus Mice
Message-ID: <[email protected]>
Date: 10 May 92 16:45:38 GMT
Organization: Georgia Institute of Technology
Lines: 7


Through all this talk about running X, I haven't heard anything whether
or not it likes a bus mouse. All of the above posts have spoken of
serial mice - is it possible to write a driver for bus mice?
--
Rob Coleman ][ [email protected] ][ "I think I'm going bald!"
Comp. Sci. ][ Georgia Institute of Technology ][ -Rush/COS


[next article]
From: [email protected] (Jim Burns)

Subject: Re: Need SCSI install help (a lot!)
Message-ID: <[email protected]>
Date: 10 May 92 18:34:07 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 36

in article <[email protected]>, [email protected] (Denis
e Tree) says:

> 2. So then I boot with vmunix(pre0.96) and remove some stuff from
> the 95a root floppy so I can uncompress tar and do "tar xvf
> /dev/at1" on the disk that I rawrote the scsi.fdisk.tar to...

Yes, 0.95a tar does have a problem, tho' I used it sucessfully on enuf
rawritten tar floppies to boot strap mtools on to the system (and then I
found out about the mcc-interim stuff - much easier! - except of course it
doesn't support scsi yet (probably will w/0.96), so I had to reverse the
order of mounting and edit the install scripts).

> 3. ^^^^^^^^^^^^ (Thinking I'm pretty clever at this point) I
> try out fdisk and get anly "can't open /dev/hd0" or some-
> thing to that effect. That's it...no /dev/hd0....nothing else.

At least the scsi fdisk tells you that much. The 0.95a fdisk tells you
NOTHING, and gives you nothing to go on. Simply 'ln' your /dev/hda from the
root image to /dev/hd0 (I think the scsi fdisk is using the old device
naming scheme), and run scsi fdisk again and find out it will probably like
wise complain about not finding /dev/hd5 (if I remember correctly) - ln it
to /dev/hdb, and finally can't find /dev/sd0 (contrary to Drew's naming
scheme) is handled by created that as mknod /dev/sd0 b 8 0 (the last # is a
multiple of 8 if there are other scsi controllers). Double check your dos
pfdisk work w/the 0.95a pfdisk (tho' don't be surprised if you get garbage
after a reboot, even if you set the (G)eometry - all is ok if you get good
numbers from scsi fdisk) and you're ready to mkfs and install. Don't forget
to redo the mknods for the scsi partitions (and the lp devices) on the hard
disk - the mkdev script doesn't know about them - and their chmoded perm-
issions. Now you're ready to get bootlin/boot.sys, mcc-interim, and other
ftp goodies!
--
BURNS,JIM (returned student)
Georgia Institute of Technology, 30178 Georgia Tech Station,
Atlanta Georgia, 30332 | Internet: [email protected]
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt0178a


[next article]
From: [email protected] (Jim Burns)

Subject: Elvis-1.5.tar.Z corrupt?
Keywords: word order reversed?
Message-ID: <[email protected]>
Date: 10 May 92 19:02:31 GMT
Organization: Georgia Institute of Technology
Lines: 22

What's the deal on the elvis distribution from banjo.concert.net:pub/Linux/
Incoming? It won't uncompress (yes I used BINARY). The 1st few lines of an
od dump are:

0000000 376 001 037 235 220 e 330 330 I 3 347 005 200 203 \b 023
01fe 9d1f 6590 d8d8 3349 05e7 8380 1308
0000020 * \ 310 260 241 303 207 020 # J D \b 242 342 215 032
5c2a b0c8 c3a1 1087 4a23 0844 e2a2 1a8d
0000040 5 @ \0 250 310 021 206 F 216 025 = n 004 I 322 #
4035 a800 11c8 4686 158e 6e3d 4904 23d2

The 1st two words look reversed to me from a normal .Z. Is the whole file
word-reversed? Is there another source of this file? Thanx.

BTW - in my previous post about elle(1), I forgot to mention that its
elle.h has a filename length limit of 14 chars. This will probably have to
be changed when we get new filesystems.
--
BURNS,JIM (returned student)
Georgia Institute of Technology, 30178 Georgia Tech Station,
Atlanta Georgia, 30332 | Internet: [email protected]
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt0178a


[next article]
From: [email protected] (Jim Burns)

Subject: Re: RTS/CTS flow control on serial ports
Message-ID: <[email protected]>
Date: 10 May 92 19:03:19 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 10

in article <[email protected]>, [email protected] (Dr
ew Eckhardt) says:

> Linux pre-.96 or before does not support hard flow control.

Will it ever?
--
BURNS,JIM (returned student)
Georgia Institute of Technology, 30178 Georgia Tech Station,
Atlanta Georgia, 30332 | Internet: [email protected]
uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!gt0178a


[next article]
From: [email protected] (H.J. Lu)

Subject: This is for Orest.
Message-ID: <[email protected]>
Date: 10 May 92 17:23:32 GMT
Sender: [email protected] (USENET News System)
Distribution: usa
Organization: Washington State University
Lines: 93
Originator: [email protected]

Sorry, I have to get Orest.

> >Hi, Orest,
> >
> >I tried to send you mails several times. I am not sure they reached you
> >or not. This is my last resort.
>
> yeah, they're still working out the bugs in my link to the mail machine
> and the mailer. i think things are beginning to stabilize a bit. sorry
> about the frustrating situation.
>
> >
> >I think I have fixed the problem with the shared libs for X11. Please
> >get the latest gcc 2.1. The shared libs for X11 now are in it. It is in
> >the usual places and still uses old LSL prefix.
>
> ok. i'd like to distribute the next release with shared libs, makes a
> great difference in the sizes of the executables
>
> >
> >BTW, which X11 lib will be always linked, beside libX11.a, if some X11
> >libs other than libX11.a are linked? Last time I chose libXt.a. It
> >turned out I was wrong. This time I chose libXmu.a. I hope I am right.
> >Please correct me if I am wrong. It is very important.
>
> thats a tough question. i'd really like to have shared libs for all the
> x11 libs, but that may be too much. if so, just having the largest libs
> shared would be great.
>

I tried to build one big shared lib for all Xii libs. But there are too
many .o files. Ld complained about it.

> when i pulled your (old) x11 shared libs stuff, i found the problem was
> you had stubs for all the libs, but only libX11 and libXt were provided in
> shared lib form. you mentioned that libXt_v0_1 had all the rest of the
> libs, but i can't verify that because my nm can't read it (only the
> libXt.a, which contains the stubs, and that only has Xt functions).
>

They are all stripped.

> these are the current sizes of the libs:
>
> # ll /u2/x11/usr/lib/
> total 994
> drwxr-xr-x 9 root root 272 May 10 01:31 X11/
> -rw-r--r-- 1 root root 421424 Apr 26 15:24 libX11.a
> -rw-r--r-- 1 root root 5454 Apr 26 15:26 libXau.a
> -rw-r--r-- 1 root root 217302 Apr 26 15:29 libXaw.a
> -rw-r--r-- 1 root root 11764 Apr 26 15:31 libXdmcp.a
> -rw-r--r-- 1 root root 21736 Apr 26 15:37 libXext.a
> -rw-r--r-- 1 root root 72160 Apr 26 15:31 libXmu.a
> -rw-r--r-- 1 root root 256966 Apr 26 15:32 libXt.a
> -rw-r--r-- 1 root root 5406 Apr 26 15:34 liboldX.a
>
> so having libX11 and libXt shared are big wins, and libXaw would be a
> logical addition (many apps use it as a default lib). even with just
> libX11 and libXt, the basic clients are very small.

With the new one, libX11.a is in libX11_v0_1. libXaw.a, libXext.a,
libXmu.a and libXt.a are in libXmu_v0_1. But you have to at least have
-lXmu to load the libXmu_v0_1. That is why I asked that question. From
what I saw, libXmu.a is always linked if more than libX11.a is needed.

>
> a problem i had when i played around with an early version of mkshared
> was that it wanted to gobble up the lib in one piece, and i didn't have
> the memory to be able to create a libX11 (too many .o's). i'll pull down
> your latest version, but i was wondering if it has the ability to do the
> job in pieces (even if i have to specify the range of .o's by hand, with

I will try to play with it to see if I can build one big image for X11.

> a little help to piece together the addresses). if that were the case,
> then i can take over providing shared libs, which would free you up.
> otherwise, i can just post the static libs and have you make them for
> me, then pull them down, build the clients and finish the distribution.

I am not so sure about that. Make shared libs really needs HUGE amount
of RAM. I am doing it on a UNIX machine which has more than 12MB RAM.
If you can provide me the static libs, it won't take much time once
mkshared is stabilized.

>
> >
> >H.J.
>
> zorst
> [[email protected]]
>

H.J.


[next article]
From: [email protected] (H.J. Lu)

Subject: Re: Bus Mice
Message-ID: <[email protected]>
Date: 10 May 92 17:26:17 GMT
References: <[email protected]>
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 14

In article <[email protected]> [email protected] (Rob Coleman) write
s:
>
> Through all this talk about running X, I haven't heard anything whether
>or not it likes a bus mouse. All of the above posts have spoken of
>serial mice - is it possible to write a driver for bus mice?

Yes. Get a PD bus mouse driver from SYSV/386 ftp sites. I think they
have Logitech and ATI bus mouse drivers.

>--
>Rob Coleman ][ [email protected] ][ "I think I'm going bald!"
>Comp. Sci. ][ Georgia Institute of Technology ][ -Rush/COS

H.J.


[next article]
From: [email protected] (Denise Tree)

Subject: SCSI partition problem?
Message-ID: <[email protected]>
Date: 10 May 92 19:35:48 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected] (Denise Tree)
Organization: The Internet
Lines: 20

Hello Linuxers,
I still have a problem trying to get Linux to see my scsi drive
a maxtor xt3280 and my controller , a seagate st-01 Bios ver. 3.2

I supposre I have a problem with the partition table on the drive because
the 0.96 vmunix boot recognizes the drive and controller but says:

"cannot read partion table of sd 0"
The scsi fdisk just says it cannot open /dev/sd0 and so I cannot use pfdisk
to repartition.
The partitions were created with the pfdisktc compiled for turbo C in DOS.
Everything works fine on the DOS side though when I put in the new con-
troller the drive would not boot again until I fdisked and sys'ed it (kept say-
ing "missing os" even though it booted with my old Future Domain card).
Any feedback from scsi Linuxers (or non-scsi too) would be /GREATLY/
appreciated as I have not heard from anyone yet (yea yeah I know everybodys
playing with their mice and pixels and plotting how to get a big monitor!!)
..................
Denise Tree
[email protected]


[next article]
From: [email protected] (Greg Lee)

Subject: Re: Some questions...(long)
Message-ID: <[email protected]>
Date: 10 May 92 20:22:03 GMT
References: <[email protected]> >
Sender: [email protected] (News Service)
Organization: University of Hawaii
Lines: 16
Nntp-Posting-Host: uhunix.uhcc.hawaii.edu

In article [email protected] (Michael Pereckas)
writes:
}
}"It is impossible to have Linux filesystems on floppy disks and keep
}them from being badly corrupted."
}
}This is with kernels earlier than 0.96 (which is not yet released).
}If you dissagree, what do you know that I don't?

I know to fsck the floppy device after inserting a disk and
before mounting it. This prevents complaints about inode in use
on my system (I don't know why). I don't know whether this will
help you. I have used floppies a lot (from v. 0.12 to pre-.96),
and I've never had one corrupted.

--
Greg Lee


[next article]
From: [email protected] (Greg Lee)

Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <[email protected]>
Date: 10 May 92 20:29:16 GMT
References: <[email protected]> <[email protected]
cs.jhu.edu> <[email protected]> <[email protected]
om>
Sender: [email protected] (News Service)
Organization: University of Hawaii
Lines: 16
Nntp-Posting-Host: uhunix.uhcc.hawaii.edu

In article <[email protected]> [email protected] (Al Clark) writes:
}In article <[email protected]> [email protected]
U (Britt Park) writes:
}>...
}>modifications. As to the requirement that complete sources to a binary on an
}>archive site must be located on the same archive site; I don't believe this
}>has any basis either, ...
}...
}Not being a lawyer, I have no idea whether or not the above has any validity.
}However, I would suggest that people be very careful about following it.
}...

Since we are so heavily indebted to fsf, it would make sense to think first
about what is courteous and much later about what we could get away with.

--
Greg Lee


[next article]
From: [email protected] (Scott Dunn)

Subject: Re: X-Windows Story
Message-ID: <[email protected]>
Date: 10 May 92 20:27:21 GMT
References: <[email protected]> <[email protected]
.wcc.govt.nz>
Sender: [email protected] (News Administrator)
Organization: Dept of Computer Science, Heriot-Watt University, Scotland
Lines: 58

In article <[email protected]>, [email protected]
nz (Michael Hamilton) writes:
|> In article <[email protected]> [email protected] |
> writes:

|> I don't have an ET4000.

|> I'm a bit impatient, so I tried X386 on my existing card, an Hewlett
|> Packard Headland Video 7, of some kind, with 256 KB - I presume this
|> is not a supported card, anyone got any info on this brand of card, as
|> I've got nothing technical? My monitor is a Philips 7 CM 3209. I
|> found that if I selected et4000, 640x400, and lied about having 512
|> KB, I could get a ghosting and wrapping image. I've also found that
|> the following gives a crisp image, but with wrapping and alternating 1
|> inch blue an white striping horizontally across the screen:
|>
|> WARNING - during playing, my monitor sometimes received some pretty
|> weird signals, and I often switched it off just in case it was being
|> over stressed. I don't fully understand H/W, so I didn't want to risk
|> it.


Wow, I'm jealous. I *do* have an ET4000 (Pro Designer II) with 1 meg
and an NEC 3D and I can't get beyond :

Fatal server error :
KDSETMODE KD_GRAPHICS failed
giving up.
xinit : Interrupted system call (errno 4) unable to connect to X server
xinit : No such process (errno 3) : Server error


I have read everything I can find, used the settings in vga.dba.
No luck.

The relevant current Xconfig bits :

Logitech "/dev/ttys2"
BaudRate 1200
SampleRate 150

#
# The graphics drivers
#
vga256
virtual 1024 800
viewPort 0 0
videoram 1024
modes "800x600" "900x690"
# clocks 25 28 40 36
clocks 25 28 57 72 78 68 50 62

ModeDB
"800x600" 45 800 912 974 998 600 600 609 631
"900x690" 60 900 1024 1080 1112 690 690 695 720


[In a quiet,forlorn little voice :- Please help ๐Ÿ™ ]

Scott.


[next article]
From: [email protected] (Tim Peoples)

Subject: Re: Bus Mice
Message-ID: <[email protected]>
Date: 10 May 92 21:05:48 GMT
References: <[email protected]> <[email protected]>
Sender: [email protected] (NetNews Administrator)
Reply-To: [email protected]
Organization: University of Arkansas
Lines: 25
Nntp-Posting-Host: engr.uark.edu
Organiztion: University of Arkansas, Dept. of Computer Systems Engineering

[email protected] (H.J. Lu) writes:

>In article <[email protected]> [email protected] (Rob Coleman) writ
es:
>>
>> Through all this talk about running X, I haven't heard anything whether
>>or not it likes a bus mouse. All of the above posts have spoken of
>>serial mice - is it possible to write a driver for bus mice?
>
>Yes. Get a PD bus mouse driver from SYSV/386 ftp sites. I think they
>have Logitech and ATI bus mouse drivers.
>

Does anyone, off hand, KNOW the address to one of these sites?


--

+--------------------------------------+-------------------------------------+
| Tim Peoples | The time has come the user said, |
| [email protected] | to talk of many things; |
| Dept. of Computer Systems Engineering| of simms and sockets and symaphores,|
| University of Arkansas, Fayetteville | of processes and pings.... |
+--------------------------------------+-------------------------------------+
| I need no disclaimer; nobody listens to what I have to say anyway!! |
+----------------------------------------------------------------------------+


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <[email protected]>
Date: 10 May 92 22:14:51 GMT
References: <[email protected]> <[email protected]
om> <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 24
Nntp-Posting-Host: romeo.cs.colorado.edu

In article <[email protected]> [email protected] (
Greg Lee) writes:
>In article <[email protected]> [email protected] (Al Clark) writes:
>}In article <[email protected]> [email protected]
DU (Britt Park) writes:
>}>...
>}>modifications. As to the requirement that complete sources to a binary on an
>}>archive site must be located on the same archive site; I don't believe this
>}>has any basis either, ...
>}...
>}Not being a lawyer, I have no idea whether or not the above has any validity.
>}However, I would suggest that people be very careful about following it.
>}...
>
>Since we are so heavily indebted to fsf, it would make sense to think first
>about what is courteous and much later about what we could get away with.
>
>--

Please see section 3, subsections b and c of the GPL, version 2. Read it
carefully. People haven't been following it to the letter, but the implications
are that if whoever does the port sends a written note, promissing to make
the complete sources available in a machine readable format for
a period of three years, fora cost not to exceed that of the actual physical
act of distribution, to an FTP site, FTP sites may maintain binaries only
with an electronic version of the note stating where sources may be


[next article]
From: [email protected] (Drew Eckhardt)

Subject: Re: Need SCSI install help (a lot!)
Message-ID: <[email protected]>
Date: 10 May 92 22:22:30 GMT
References: <[email protected]> <[email protected]>
Sender: [email protected] (The Daily Planet)
Organization: University of Colorado at Boulder
Lines: 48
Nntp-Posting-Host: romeo.cs.colorado.edu

In article <[email protected]> [email protected] (Jim Burns) writes:
>in article <[email protected]>, [email protected] (Deni
se Tree) says:
>
>> 2. So then I boot with vmunix(pre0.96) and remove some stuff from
>> the 95a root floppy so I can uncompress tar and do "tar xvf
>> /dev/at1" on the disk that I rawrote the scsi.fdisk.tar to...
>
>Yes, 0.95a tar does have a problem, tho' I used it sucessfully on enuf
>rawritten tar floppies to boot strap mtools on to the system (and then I
>found out about the mcc-interim stuff - much easier! - except of course it
>doesn't support scsi yet (probably will w/0.96), so I had to reverse the
>order of mounting and edit the install scripts).
>
>> 3. ^^^^^^^^^^^^ (Thinking I'm pretty clever at this point) I
>> try out fdisk and get anly "can't open /dev/hd0" or some-
>> thing to that effect. That's it...no /dev/hd0....nothing else.
>
>At least the scsi fdisk tells you that much. The 0.95a fdisk tells you
>NOTHING, and gives you nothing to go on. Simply 'ln' your /dev/hda from the
>root image to /dev/hd0 (I think the scsi fdisk is using the old device
>naming scheme), and run scsi fdisk again and find out it will probably like
>wise complain about not finding /dev/hd5 (if I remember correctly) - ln it
>to /dev/hdb, and finally can't find /dev/sd0 (contrary to Drew's naming
>scheme) is handled by created that as mknod /dev/sd0 b 8 0 (the last # is a
>multiple of 8 if there are other scsi controllers). Double check your dos


Not quite correct : all present SCSI busses are scanned at bootup, with
minors 0-15 assigned to the first found, 16-31 to the second, etc.
It's a multiple of 16, or bitshift left by one nibble for the minor number.

After the extended partition code stabalized, I was going to splice that
in to the sd.c driver, and just haven't gotten around to it yet -
hence 16 minors reserved per disk.

>pfdisk work w/the 0.95a pfdisk (tho' don't be surprised if you get garbage
>after a reboot, even if you set the (G)eometry - all is ok if you get good
>numbers from scsi fdisk) and you're ready to mkfs and install. Don't forget
>to redo the mknods for the scsi partitions (and the lp devices) on the hard
>disk - the mkdev script doesn't know about them - and their chmoded perm-
>issions. Now you're ready to get bootlin/boot.sys, mcc-interim, and other
>ftp goodies!

As far a the SCSI fdisk, eventually I expect that the naming convention
will be /dev/, with a digit, and
a letter, like "normal" Unix systems. When this happens, I will provide
a fdisk that uses the new "sane" partitioning scheme. Until then,
DOS works, etc.


[next article]
From: [email protected] (Wen-Chun Ni)

Subject: xterm problem
Message-ID: <[email protected]>
Date: 10 May 92 22:53:40 GMT
Sender: [email protected]
Organization: Brown University Department of Computer Science
Lines: 15



I got X11 run on my system. It's very exciting! Everything seems
to work fine except the xterm problem. When I use vi in xterm,
it seems to "over-scroll;" so the cursor doesn't seem to be correct
anyway.I don't know whether this is a bug of elvis or xterm.


On my 385/25 system, the performance is good enough. Thank you Orest!
I'd better go for a large drive to compile something myself.
Hello Maxtor, did you hear me? ๐Ÿ˜‰



Wen-Chun Ni


[next article]
From: [email protected] (Magnus Alvestad)

Subject: X and Screen cards
Message-ID:
Date: 11 May 92 04:53:17 GMT
Sender: [email protected] (Bergen University Newsaccount)
Reply-To: [email protected]
Organization: AlveSoft
Lines: 9

Could everyone that has tried an ET4000 screen card with X386 please
*mail* me their results, so I can compile a list of cards that work.
Remember to include exact card spesifications, including memory. I
will, of course, summarize.

-Magnus
--
| Magnus Alvestad | I am a prolog program. I get all my opinions
| [email protected] | from a small organization in Birmingham.


[next article]
From: [email protected] (Steffen Finger)

Subject: make compatible to bsd-make ?
Keywords: make bsd linux libs
Message-ID: <[email protected]>
Date: 6 May 92 10:15:50 GMT
Organization: Student Pool, CSD, University of Erlangen, Germany
Lines: 36

Hi everybody,
I tried to port a selfwritten library for picl-emulation, developed on a sun-spa
rc,
to linux and found make does not understand my makefile.
Which make is guilty ? gmake on linux or make on sunos ?

Please have a look on this:

[details deleted ...]

OBJ = $(SRC:.c=.o)
LIB = lib/lib$(MODE).a

.SUFFIXES: .c .a
.PRECIOUS: $(LIB)
.SILENT:

.c.a:
echo "Compiling $<"
$(CC) $(INCLUDE) $(CCFLAG) -c $< -o $*.o
@ar r [email protected] $*.o
@rm -f $*.o
ranlib [email protected]

$(LIB): $(LIB)($(OBJ))


Any ideas ?

Thanks Steffen


---
| I T ___ | Steffen Finger University of Erlangen-Nbg Germany |
| J U S T /\__\ | Internet: [email protected] |
| W O R K S ! \/__/ | NeXT Mail: [email protected] |
+------------------------------------------------------------------------------+


[next article]
From: [email protected] (H.J. Lu)

Subject: Re: Bus Mice
Message-ID: <[email protected]>
Date: 10 May 92 21:51:39 GMT
References: <[email protected]> <[email protected]>
<[email protected]>
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 25

In article <[email protected]> [email protected] writes:
>[email protected] (H.J. Lu) writes:
>
>>In article <[email protected]> [email protected] (Rob Coleman) wri
tes:
>>>
>>> Through all this talk about running X, I haven't heard anything whether
>>>or not it likes a bus mouse. All of the above posts have spoken of
>>>serial mice - is it possible to write a driver for bus mice?
>>
>>Yes. Get a PD bus mouse driver from SYSV/386 ftp sites. I think they
>>have Logitech and ATI bus mouse drivers.
>>
>
> Does anyone, off hand, KNOW the address to one of these sites?
>
>
>--

Host gatekeeper.dec.com

Location: /.2/micro/sysv-386
DIRECTORY dr-xr-xr-x 512 Nov 13 16:27 mouse
DIRECTORY dr-xr-xr-x 512 Nov 13 16:27 ati-mouse

H.J.


[next article]
From: [email protected] (Terry Dawson)

Subject: Re: RTS/CTS flow control on serial ports
Keywords: RTS CTS Flow control
Message-ID:
Date: 10 May 92 22:18:01 GMT
References: <[email protected]
do.edu>
Sender: [email protected]
Organization: Sydney University Computing Service, Sydney, NSW, Australia
Lines: 29
Nntp-Posting-Host: extro.ucc.su.oz.au

[email protected] (Drew Eckhardt) writes:

>>I have used:
>>getty /dev/ttys0
>>stty 2400 sane >/dev/ttys0
>>stty crtscts >/dev/ttys0
>>
>>and all seems to work just fine, except the flow control.
>>
>>What am I doing wrong ?
>>

>Linux pre-.96 or before does not support hard flow control.

It doesn't ? ๐Ÿ™‚ I guess that'll explain it then ๐Ÿ™‚
Will 0.96(proper) support it ?

I tried setting my machine up so I can dial into it at 2400, and the thing
is dropping characters there as well .. interestingly, it only seems to happen
on stuff that linux transmits, not receives. Certainly not what I would have
expected.

Thanks.
Terry


--
Terry Dawson, [email protected], vk2ktj%[email protected]
+61 2 925 1556 (voice), +61 2 922 5973 (fax). __\*/__ 0^Ooooo _____


[next article]
From: [email protected] (Erik J. Arthur)

Subject: pcomm95c
Message-ID: <[email protected]>
Date: 11 May 92 00:50:31 GMT
Sender: [email protected] (Usenet News Administration)
Organization: University of Minnesota
Lines: 20
Nntp-Posting-Host: dexter.psych.umn.edu

Hi,

I think this has probably been asked before, but here goes again. When I connec
t to
a machine via pcomm, I can't see the line that I am typing. After I hit enter,
the
line shows up. Would someone please tell me what I am doing wrong. I have trie
d
flow control, and altering how pcomm interprets incoming and outgoing CRs. I am
using .95c+ and an internal modem on ttys4.

Thanks in advance,

Erik

ps. Great job on linux

-----------------------------------------
Erik Arthur [email protected]
-----------------------------------------
Phone : (612) 626 - 7521
(612) 626 - 1063
-----------------------------------------


[next article]
From: [email protected] (Jonathan Marsden)

Subject: Re: Suggestion: Multiformat compress
Message-ID: <[email protected]>
Date: 9 May 92 10:32:08 GMT
Sender: [email protected] (newsout1.26)
Organization: FidoNet node 12:2000/6300 - FEBnet Test, Val MM Philippines
Lines: 27

[email protected] (Eugenio Sanchez) said:

> So, why not create a general 'compress' that can be fed with *any*
> compac-ted file? Say, if I want to compress my files using the Zip format, I
> woulduse:

> compress -zip files...

> compress -arj files... <- Does anybody have sources 4 this?

Source for ARJ DEcompress is available. For ARJ compress, you pay $$. Contact
the author, [email protected], for details. He's friendly.

> Obviously, no switches mean that you're using the .Z format (but we can
> intro-duce a -Z switch...) ๐Ÿ™‚

> The problem is, to recognize the file format at
> uncompress time.

The job of the file program, no? There is a program for DOS and OS/2 called POL
YXARC that recognizes signatures in the compressed files and runs appropriate de
archivers. It is free with C source, and the set of signatures and decompressor
s is set in a c
onfig file. If you are serious about doing this work, start from that, hack fil
e to handle all desired signatures, and a shell script is all you'd need. WHy a
ll the talk of a huge program? BTW, the author of POLYXARC is Jeffrey Nonken of
FidoNet 1
15. File on BBSs is often PXDOS_21.LZH.

Jonathan


--
Jonathan Marsden - via Global Mission Network node 12:2000/6304
UUCP: {elroy}!wciu!febc!6300!jm
INTERNET: [email protected]


[next article]
From: [email protected] (Denise Tree)

Subject: SCSI problems YET STILL
Message-ID: <[email protected]>
Date: 11 May 92 02:06:56 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected] (Denise Tree)
Organization: The Internet
Lines: 28


JIm Burns writes:
"ln /dev/hda /dev/hd0 <...> ln /dev/hdb /dev/hd5 <...> and then
mknod /dev/sd0 b 8 0 "
A thousand thanksyous JIm but I still got trouble. Now when I rn scsi.fdisk
I get :
scsidisk I/O error
dev 0800, block 0
scsidisk I/O error
dev 0800, block 0
scsidisk I/O error
dev 0800 block 0

When LInux boots it complains of non-standard sector size (I think)... and
when I look at the boot sector with Nortons things don't look too good; for
instance in the field for sys ID is a bunch of high bit characters and such.
I am confused about why I get no complaint from DOS though if the boot sector
or partition table is messed up.

A while back someone mentioned that there was a reference available for
hard drive geometries on the net... could that person please repeat the
name of that file so I could call Archie with it? I may have messed stuff
up when I pfdisked my new partitions.

Not having Linux running is VERY DEPRESSING...!
...................
Denise Tree
[email protected]


[next article]
From: [email protected] (Udo Christ)

Subject: Linux and XWindows
Message-ID: <[email protected]>
Date: 7 May 92 12:03:49 GMT
Sender: [email protected] (USENET News account)
Organization: University of Wuerzburg, Germany
Lines: 8

Could anyone please mail me which version of X i need to get it running
on my computer ( Linux 0.95c ) and where i can get the whole X stuff
including probably patches .


please mail [email protected]

Thanks a lot


[next article]
From: [email protected] (Budi Rahardjo)

Subject: PS/2 mouse and X386
Keywords: x386 mouse
Message-ID: <[email protected]>
Date: 11 May 92 04:11:01 GMT
Sender: [email protected]
Organization: Electrical Engineering, U of Manitoba, Winnipeg, Manitoba, Canada
Lines: 15
Nntp-Posting-Host: barkeep.ee.umanitoba.ca

Finally after playing with the vga256 entry I got the screen right.
I am using Gateway 2000 386SX/16 with 10 Meg RAM, 30 Meg Linux Partition.
It turned out the VESA 800x600 entry in the vga.database file works fine.
Unfortunately I have PS/2 compatible mouse, and twm expect me to put xclock.
Is my mouse considered a bus mouse ?
I have downloaded mouse.{c,h} from gatekeeper.dec.com:/.2/micro/sysv-386/mouse
but it's for Logitech mice. Anybody has a driver for PS/2 type mouse ?

Thanks.... Linux is realy realy realy great....

-- budi
--
Budi Rahardjo

Electrical Engineering - University of Manitoba - Canada


[next article]
From: [email protected] (James Henrickson)

Subject: Re: X386, Shared Libs, and OS/2 2.0 Boot Manager, etc.
Keywords: X386 OS/2 2.0 BOot Manager
Message-ID: <[email protected]>
Date: 10 May 92 01:05:36 GMT
References: <[email protected]>
Organization: State University of New York -- Institute of Technology
Lines: 32



[email protected] (David W. Summers) writes:

>(1) I was just getting ready to try X386 for the first time today and somehow
> managed to zap my /lib/libXX.YY.ZZ (Shared library file), I THINK.
>
>I think this is what happened because I was doing something (probably going
>to fast in a speed zone ๐Ÿ™‚ in the /lib directory and all of a sudden a
>lot of my programs like 'cat', 'ln', 'mv', 'cp', etc., started bombing
>when I tried to use them. I assume I zapped the shared library.

I have a problem, too, but not that bad. Most of my stuff works, but 'mv'
doesn't want to move directories. It keeps complaining that it can't find
'movdir'. As far as I have found, 'mv' is the only problem. It worked fine
until I started installing X386. Instead of moving directories, I went through
the tedious process of making directories and then moving the files. Weird.

-------------------------------------------------------------------------------

Now to an X question: how do I select a window for keyboard input? I've tried
nearly every key/mouse button combination I could think of. I have a
Microsoft two-button serial mouse. Could that be part of the problem? Can
anyone recommend a three button mouse that will work with X386?



--
James L. Henrickson, Software Engineer * "machine learning for
Critical Technologies Inc. * autonomous vehicles, digital
311 Turner St Suite 303 * signal processing & parallel
Utica, NY 13501 [email protected] * computing"


[next article]
From: [email protected] (SYSTEM OPERATOR)

Subject: Info on Linux
Message-ID:
Date: 11 May 92 04:15:41 GMT
Organization: TCS Constulting Services, Peachtree City, GA
Lines: 8

I'm sorry to ask this possibly repeated question:

a: What is Linux (impression so far: OS for IBM/clones in shareware)
b: How would I get Linux myself?
c: Is their a FAQ? If so is there a mailserver I can get it from?
d: Is this Unix compat. or any other OS compat?

Thanx for any info?


[next article]
From: [email protected] (Wen-Chun Ni)

Subject: Xterm, elvis, etc
Message-ID: <[email protected]>
Date: 11 May 92 05:53:15 GMT
Sender: [email protected]
Organization: Brown University Department of Computer Science
Lines: 16

After playing with the Linux X11, I found that elvis cannot work under
xterm even (row, column)= (80,24). Surprisingly, the other editor joe
works just find. Because of the not-yet-fixed bug in xterm, the window
change will not be reflected.

On the other hand, kermit and pcomm behave a little weird. Since kermit
is supposed to work perfectly, I tried to figure out the problem. From
what I've observed, there is some little bug in the ttyp? driver such
that raw control under kermit can't work right.



Wen-Chun Ni

P.S. Today is my happiest day recently, since I got X11 run under my machine
and it just looks like the Sparc at school.


[next article]
From: [email protected] (Peter MacDonald)

Subject: More ET4000: a bug?
Message-ID: <[email protected]>
Date: 11 May 92 05:08:49 GMT
Sender: [email protected]
Organization: University of Victoria, Victoria, BC, CANADA
Lines: 19
Nntp-Posting-Host: sanjuan.uvic.ca


I have found that two ET4000 cards (Aamazing and VGA 2theMax) which both work
fine under DELL Unix at work, do not work under Linux at home. And I am seeing
some funny behaviour on the clock reporting. The Aamazing has a utility that
sets modes (dmode) and tells you the Video Clock associated. It reports a
set of 8 modes. Cooincedentally, the modes reported by X386 when it starts up
are "exactly half". Ie:

Aamazing Utility: 50 56 72 45 60 65 40 47

becomes

X386: 25 28 36 23 30 32 20 24

And no matter what I do, I always get multiple images (4 usually) on the screen.
Could this be a bug in the X386 server?


PS: Solved my mouse problem by switching to a different MS mouse.


[next article]
From: [email protected] (Homer Wong)

Subject: .login/.profile/.cshrc problems
Keywords: tcsh 6
Message-ID:
Date: 11 May 92 06:42:17 GMT
Sender: [email protected] (USENET News System)
Organization: Prentice Centre, University of Queensland
Lines: 13

I have tcsh 6.xx installed properly, but for some reason it isn't executing
.login or .cshrc (I even tried .profile but no luck) on login. I think
I had it executing at one stage, but I installed the shadow password system
and now it doesn't. ๐Ÿ™

The home dir is getting set properly - when the user logs in, they are sitting
in their proper home dir, with a '> ' prompt, but that's it.

Any help would be appreciated. (sorry if this is a FAQ problem)
Even a sample of passwd/shadow files from someone who has tcsh going properly
would be nice.

Cheers


[next article]
From: [email protected] (Matthias Lautner)

Subject: dos emulator
Message-ID: <[email protected]>
Date: 11 May 92 07:51:04 GMT
Sender: [email protected] (Matthias Lautner)
Organization: University of Wuerzburg, Germany
Lines: 30

Hi!

About the state of my dos emulator:

+ running in virtual 8086 mode
+ boots dos from disk or hard disk (dos partition)
+ can read, write, dir the LINUX file system (as drive D)
+ can also boot from LINUX file system
- has problems to run programs on LINUX file system (working on)
- doesn't emulate direct screen access
- problems with detecting the format of floppies and the change of a floppy
- to access dos file system on hard disks /dev/hda must be readable (writeable)
- video int 0x10 only partial implemented

If most of the problems are solved I'm going to post the kernel patches and
the emulator to the net.

About the implementation:

Emulating the bios calls (screen(10), configuration(11), memory test(12),
disk io(13), keyboard(16), load system(19)) is enough to boot dos from floppy.
But dos is polling the keyboard, so all dos system calls using the keyboard
needed emulation too, to use blocking keyboard reads.
To access the LINUX file system, I emulate most of the dos system calls, if
they use the special drive devoted for LINUX.

Matthias
--
*****************************************************************************
Matthias Lautner Carl-Vornberger-Str. 21, 8701 Randersacker, Germany
[email protected]
*****************************************************************************


[next article]
From: [email protected] (Linus Benedict Torvalds)

Subject: Re: Sources (IMPORTANT to managers of LINUX ftp sites)
Message-ID: <[email protected]>
Date: 11 May 92 08:33:44 GMT
References: <[email protected]> <[email protected]
om> <[email protected]>
Organization: University of Helsinki
Lines: 24

In article <[email protected]> [email protected] (
Greg Lee) writes:
>
>Since we are so heavily indebted to fsf, it would make sense to think first
>about what is courteous and much later about what we could get away with.

Indeed. I hope people try to follow the /intent/ of the GPL first, and
then start worrying about the legal implications. I don't feel too
strongly about the legalese: it's there just because it's required to
uphold that intent in a court of law, and wouldn't even be needed
otherwise. I'd suggest not reading the GPL searching for loopholes -
but on the other hand not being unnecessarily strict about it either.
Use a bit of common sense in it all (until somebody starts threatening
with legal action: but in that case you have probably not been following
the GPL even in intent).

There are probably quite a few binaries that should be removed from the
linux archives: especially the older ones that needed a bit more porting
than they need now with the better compiler/library, and that might not
have source (I think my original gcc-1.40 and bash-1.05 should probably
be removed: I've had to delete my sources to get X running.. Nobody
uses them any more anyway). It might even result in some general
cleanup of old binaries...

Linus


[next article]
From: [email protected] (Alex C. Liu)

Subject: GROFF
Message-ID:
Date: 11 May 92 09:35:32 GMT
Sender: [email protected] (Alex Liu)
Organization: None to Speak of
Lines: 11
NNTP-Posting-Host: aludra.usc.edu

Can somebody post a new version of GROFF? The one in TSX-11 is for
0.12 thus can't be executed on the newer kernels (0.95c?) I am unable
to compile GROFF because I lack the diskspace and the G++ libraries.



--
_____________________________________________________________________________
Alejandro Liu |EMail: [email protected] |All mispellings are intentional
3131 Mc Clintock #373F |Voice: 213-745-2431 |Anything mentioned here is not
Los Angeles, CA 90007 | |necessarily true.


[next article]
From: [email protected] (Stephen R. van den Berg)

Subject: Re: Password encryption bogus?
Message-ID: <[email protected]>
Date: 11 May 92 09:42:41 GMT
References: <[email protected]> <[email protected]
k.rwth-aachen.de> <[email protected]>
Sender: [email protected] (Newsfiles Owner)
Organization: RBI - RWTH Aachen
Lines: 39
Originator: [email protected]
Nntp-Posting-Host: hathi

Drew Eckhardt writes:
>Stephen R. van den Berg writes:
>>[email protected] writes:
>>>Karl R. Buck writes:

>>I was wondering, why not use MD5 to do the password encryption? It

>If it's not the DES based crypt algorithm, it is non standard.

It's a standard all right, just not the old DES standard (which has outlived
its usefulness).

>This breaks many things. Forexample, programs like xlock that
>use getpwent(3) and crypt(3) will fail. I can't use a global
>password file. Bad.

Not so, like you say, it uses getpwent(3) and crypt(3), so provide a
substitute for crypt(3) that uses the new algorithm (getpwent does not need
any changes, it can still be used as is), and presto, all applications use
the new and improved password encryption algorithm.

>>supposedly guarantees uninvertability, and it is freely available; what's
>>more, you can easily persuade it to accept more than eight characters.

>UFC is also freely available, export legal, and standard. We don't
>want more than 8 characters for compatability reasons.

I don't see why having more than 8 characters would break many things?
I'm not saying that the *encrypted* string change its length or presentation
in any passwd files, I'm just saying that the encryption routine should
use a better method than DES (like MD5) and should be able to accept more
than 8 characters.

Using this new algorithm defeats the use of the popular crack program.
--
Sincerely, [email protected]
Stephen R. van den Berg (AKA BuGless). [email protected]

"My name is Psmith, the P is not pronounced."


[next article]
From: [email protected] (R.J.Newton)

Subject: Unsupported SVGA cards under X
Message-ID: <[email protected]>
Date: 10 May 92 11:26:00 GMT
References: lsinki.FI> <[email protected]>
Sender: [email protected]
Organization: Cripps Computing Centre, University of Nottingham
Lines: 11

Looks like I might be joining the ranks of those with a useless SVGA card as
far as X is concerned. It's a new card based on the high-performance WD90c30
(Western Digital) chipset. After spending all day yesterday getting all
manner of bent screens and out-of-sync pictures I'm about to give up.

I read in an earlier article that the WD90c00 and WD90c11 are compatible with
the PVGA mode in X386. Can anyone please tell me if the same is true of my
(expensive and I don't want to throw it away) WD90c30 ? I'd dearly love to
see X on my 486-33 !

-Thanks !


[next article]
From: [email protected] (The Master of Symbolic Links)

Subject: Re: X and supported VGA cards.
Message-ID: <[email protected]>
Date: 11 May 92 09:03:23 GMT
References: <[email protected]> <[email protected]
edu>

Sender: [email protected] (USENET Newssystem)
Organization: Inst. fuer Informatik, Technische Univ. Muenchen, Germany
Lines: 26
In-Reply-To: [email protected]'s message of 10 May 92 13:10:49 GMT

>I am coordinating the effort on making more graphics card drivers for
>X386 available. If you are working on a card or want to know who are,
>mail me at [email protected]

And a cc: to me would be nice, too.

>As far as I know no one is working on any TIGA boards.

I simply would forget about this. TIGA might be an independend standard
under MS-DOF, but this is only because for every board you'll get a
customized tigacd.exe. Under UNIX you had to talk to directly to the
hardward itself. And this is quite different for every graphics board.
And besides this, TI doesn't give away these docs for free, nor are you
allowed to pass this information to third parties for free. Besides these
facts are the 340x0 slower than the cheaper 8514/A clones.

- Thomas


--
-------------------------------------------------------------------------------
e-mail: [email protected]

immer ?
nein, nicht immer ...
... aber immer oefter !


[next article]
From: [email protected] (Mika Pekka Liljeberg)

Subject: Re: X, ET4000 and mice
Message-ID:
Date: 11 May 92 10:21:13 GMT
References: <[email protected]> <[email protected]
.com>
<[email protected]>
Sender: [email protected] (Uutis Ankka)
Organization: Department of Computer Science, University of Helsinki, Finland
Lines: 38
In-Reply-To: [email protected]'s message of 10 May 92 09: 22:56 GMT

In article <[email protected]> [email protected]
.EDU (Chin Fang) wrote:

[ interesting stuff deleted ]

> In short, this is a classical constrained optimization problem. One way
> to solve it is to use Danzig's simplex method [a] or a simplex method
> with BFGS update nonlinear programming [b] afterwards for polishing (geez,
> what a fussy guy :). However, the big trouble I had (and still having!) is
> that there is no way I can 100% sure about the constraint values that *my
> program* gets. That is, it's not that hard to write a C code,
> probing the hardware, and obtaining constraints discussed above. But
> without extensive testing and validation, the timing suggested by such a code
> would be almost untrust-worthy as a random guess as soon as it is used on
> a unknown hardware. Thus it would be really imprudent to release it.

Perhaps the solution is not to write a program that does the whole job
for you, but to write a program that _facilitates_ the process of
configuring the display. The program could start with a less than optimal
mode that is very likely to work. The user would then be provided with
a set of controls to tune the parameters, ie. increase/decrease frame
length/beam retrace timing/clock etc. The user would also have the option
of saying: "this one works". This would let the program collect data about
monitor timings.

I'm probably missing a lot here, but what I'm driving at is a sort of
a learning program that let's the user tune things (or maybe adjusts
the parameters all by itself) and asks: "Does this work?".

Compared to the current Linux X386 configuration cycle (edit, startx,
reboot... edit, startx, reboot... etc.) this sure would make things
a helluva lot easier.

Mika

--
Mika Liljeberg Email: [email protected]
Helsinki University [email protected]
Dept. of Computer Science


[next article]
From: [email protected] (The Master of Symbolic Links)

Subject: Re: X386: R4 or R5?
Message-ID: <[email protected]>
Date: 11 May 92 09:11:41 GMT
References: <[email protected]>
Sender: [email protected] (USENET Newssystem)
Organization: Inst. fuer Informatik, Technische Univ. Muenchen, Germany
Lines: 16
In-Reply-To: [email protected]'s message of Sat, 09 May 1992 20:17:27 GMT

>Lastly, Thomas, if I want to post my driver, which is for 1.1b, to say,
>alt.sources, can I do it (re: copyrights, etc.)? Must I post diffs
>(I'd prefer to post the whole thing: driver.c and ban

There were major changes. Especially the 1.2 driver do now autodetecting
of the HW and try to find out which dot-clocks are on the VGA (superseedes
clock.exe completely). Also there were many internal changes.

- Thomas
--
-------------------------------------------------------------------------------
e-mail: [email protected]

immer ?
nein, nicht immer ...
... aber immer oefter !


[next article]
From: [email protected] (The Master of Symbolic Links)

Subject: Re: X, ET4000 and mice
Message-ID: <[email protected]>
Date: 11 May 92 09:28:02 GMT
References: <[email protected]> <[email protected]
.com>
<[email protected]>
Sender: [email protected] (USENET Newssystem)
Organization: Inst. fuer Informatik, Technische Univ. Muenchen, Germany
Lines: 22
In-Reply-To: [email protected]'s message of 10 May 92 09:22:56 GMT

>David Wexelblat at AT&T Bell Laboratories started collecting timing info
>for various hardware a while back, primarily in comp.unix.sysv386. Early
>on I naively thought that soon this database would grow very significantly
>and thus I would have a good base to train a neural network code [c],
>which, after trained using data in this database, would provide a good guess
>(more precisely, interpolation, as neural network <=> glorified data
> interpolator). But after months, even the latest version of X386 timing
>data base is not as extensive as I would like to see. So another regret...

Actually I also thought about using neuronal nets to predict a better display.
But actually it failed miserably. One reason was that both the monitor and
the SVGA show a non-linear behaviour regarding timings. So often the net
suggested timings that were not possible on the hardware.

- Thomas
--
-------------------------------------------------------------------------------
e-mail: [email protected]

immer ?
nein, nicht immer ...
... aber immer oefter !


[next article]
From: [email protected] (The Master of Symbolic Links)

Subject: Re: More ET4000: a bug?
Message-ID: <[email protected]>
Date: 11 May 92 09:30:41 GMT
References: <[email protected]>
Sender: [email protected] (USENET Newssystem)
Organization: Inst. fuer Informatik, Technische Univ. Muenchen, Germany
Lines: 27
In-Reply-To: [email protected]'s message of 11 May 92 05:08:49 GMT

> I have found that two ET4000 cards (Aamazing and VGA 2theMax) which both wor
k
> fine under DELL Unix at work, do not work under Linux at home. And I am see
ing
> some funny behaviour on the clock reporting. The Aamazing has a utility tha
t
> sets modes (dmode) and tells you the Video Clock associated. It reports a
> set of 8 modes. Cooincedentally, the modes reported by X386 when it starts
up
> are "exactly half". Ie:
>
> Aamazing Utility: 50 56 72 45 60 65 40 47
>
> becomes
>
> X386: 25 28 36 23 30 32 20 24
>
> And no matter what I do, I always get multiple images (4 usually) on the scr
een.
> Could this be a bug in the X386 server?

Yeah, this is a known bug in X386 1.2 on the MIT tape. I fixed it for the
driver Dell sells ...

- Thomas
--
-------------------------------------------------------------------------------
e-mail: [email protected]

immer ?
nein, nicht immer ...
... aber immer oefter !


[next article]
From: [email protected] (Matt Mosley)

Subject: elm2.3
Message-ID:
Date: 11 May 92 00:22:07 GMT
Organization: Starnet BBS
Lines: 13

I've been trying to compile the Elm 2.3 kit, and keep getting errors
about "unreferenced functions" in curses and several other .o files.
Anyone else get this, and know how to fix it?

Also, for anyone interested, the binmail for Minix seems to work under
Linux. I ftp'ed a simple /bin/mail that was designed for Minix, and
compiled it under Linux quite easily.

--Matt

---
Matt Mosley - starcom!mattm / [email protected]
Starcom support system administrator


[next article]
From: [email protected] (Stephen Harris)

Subject: Re: Shared Libs (now fixed! Thanks!)
Keywords: Shared Libs
Message-ID: <[email protected]>
Date: 9 May 92 21:30:47 GMT
References: <[email protected]>
Organization: Spuddy's Public Usenet Domain
Lines: 18

In article <[email protected]> [email protected] (David W
. Summers) writes:
>Several people suggested that I re-boot with the installation floppy and
>do a 'ln -s /usr/shared/lib/lib92.04.06 /lib'. I did that and my
>problem is now fixed.

Actually, I would have moved the lib92 file to /lib and reversed the symlink:
ln -s /lib/lib92.04.06 /usr/shared/lib

That way if you can't mount the /usr partition for some reason the shared
library would still be available.

Of course, this won't matter to you if /usr is on the root partition ๐Ÿ™‚

--
Stephen Harris
[email protected] ...!uknet!axion!spuddy!sweh

* Meeeeow ! Call Spud the Cat on > +44 203 638780 < for free Usenet access *


[next article]
From: [email protected] (Stephen Harris)

Subject: Re: ESDI drive compatibility
Message-ID: <[email protected]>
Date: 10 May 92 01:37:23 GMT
References: <_ssv#[email protected]> <[email protected]>
Organization: Spuddy's Public Usenet Domain
Lines: 45

In article <[email protected]> [email protected] writes:
> b) Remap my drive so that it had fewer than 1024 cylinders. My drive
> has 1300 cylinders, but since Linux uses the standard DOS
> partition table, I was restricted to 10-bits to represent
> cylinder numbers, which limits me to 1024 cylinders.

No way! I'm currently using a DPT ESDI card (with old firmware) and a
Seagate 4766E drive. The drive is set up as 16 heads, 63 sectors, 1283
cylinders using the DPTFMT command.

I have the first 1023 cylinders for DOS 5.0 (504Mb) in partition
hda1. I set partition hda3 up as a 128Mb extended partition, and in that:
hda5 = 10Mb root
hda6 = 12Mb swap
hda7 = 64Mb usr
hda8 = 42Mb home

All of these partition spread well over the 1024 cylinder limit!

Linux in fact uses the starting sector from the partition table. Since that
is a full 32bit number, it can point to anywhere on the disk.
When Linux boots it prints out the partition table SECTOR numbers properly.

The fdisk program by LeBlanc (which I used to create the above partitions)
moans that the partitions overlap (incorrectly) since it compares by cylinder,
but otherwise sets the information correctly (versions 0.91 and 0.92)

My next problem is when I plug in my SECOND disk (same size) I'll be forced
to have 10x64Mb partitions because of the current FS restrictions ๐Ÿ™

NOTE: This may not be true for Shoelace. The standard BIOS is restricted
to 1024 cylinders, so any program that uses that (eg DOS!) can't access
above that. I dunno how shoelace reads the disk. I don't use it. I find the
MBOOT program from PC-MAG along with bootlin program by Fabien COUTANT
an excellent combination. (MBOOT takes NO memory to work in, and only
modifies the DOS boot record, not the hard disk MASTER boot record).

Sorry for waffling on so long, but having a large disk like this means I
spent a lot of time looking into the above-1024 situation ๐Ÿ™‚

--
Stephen Harris
[email protected] ...!uknet!axion!spuddy!sweh

* Meeeeow ! Call Spud the Cat on > +44 203 638780 < for free Usenet access *


[next article]
From: [email protected] (Magnus Alvestad)

Subject: Re: X, ET4000 and mice
Message-ID:
Date: 11 May 92 16:22:24 GMT
References: <[email protected]> <[email protected]
.com>
<[email protected]>

Sender: [email protected] (Bergen University Newsaccount)
Reply-To: [email protected]
Organization: AlveSoft
Lines: 11
In-Reply-To: [email protected]'s message of 11 May 92 10:21:13 GMT

[ message about a somewhat self-configuring x386 ]

IMHO it would be wiser if the program itself ran in text mode. You
edited the parameters, and then it put up a graphical screen with
those parameters for a short period of time.

Who makes it? ๐Ÿ™‚

--
| Magnus Alvestad | I am a prolog program. I get all my opinions
| [email protected] | from a small organization in Birmingham.


[next article]
From: [email protected] (Stephen R. van den Berg)

Subject: Re: burst (mail de-digestifier) - anyone?
Message-ID: <[email protected]>
Date: 11 May 92 10:32:23 GMT
References: <[email protected]>
Sender: [email protected] (Newsfiles Owner)
Organization: RBI - RWTH Aachen
Lines: 31
Originator: [email protected]
Nntp-Posting-Host: hathi

V5068U%[email protected] writes:
> Does anyone have (or know of) a PD source for burst? I use a VM mailer
>that gloms everything into one big file and I really need to sort my mail out.

Try the "formail" program from the "procmail" package, it probably is exactly
what you are looking for. (To install it, a simple "make formail" should
suffice).

Feature summary for formail:
+ Can generate auto-reply headers
+ Can convert mail into standard mailbox format (so that you can
process it with standard mail programs)
+ Can split up mailboxes into the individual messages
+ Can split up digests into the individual messages
+ Can split up saved articles into the individual articles
+ Can do simple header munging/extraction

----------------------
A recent version can be picked up at various comp.sources.misc archives.
The latest version can be obtained directly from the ftp-archive at:

ftp.informatik.rwth-aachen.de (137.226.112.31)

as compressed tar file: pub/unix/procmail.tar.Z <128KB
or in compressed shar format: pub/unix/procmail.0?.Z
----------------------
--
Sincerely, [email protected]
Stephen R. van den Berg (AKA BuGless). [email protected]

"My name is Psmith, the P is not pronounced."


[next article]
From: [email protected] (northstar ken)

Subject: Ram disk
Message-ID:
Date: 10 May 92 10:50:29 GMT
Organization: Satellite Telecom, Geldrop Holland
Lines: 5

Hi..

Is it possible to make a RAM disk under Linux?

Please reply through email, to [email protected]


[next article]
From: [email protected] (Mika Pekka Liljeberg)

Subject: Re: X Problems..
Message-ID:
Date: 11 May 92 11:49:37 GMT
References: <[email protected]> <[email protected]
ak.com>
Sender: [email protected] (Uutis Ankka)
Organization: Department of Computer Science, University of Helsinki, Finland
Lines: 56
In-Reply-To: [email protected]'s message of 10 May 92 06: 11:41 GMT

In article <[email protected]> [email protected] (Orest Zbor
owski COMP) wrote:
> [email protected] (Mark Holden) writes:
>>In article , [email protected]
i.FI (Mika Pekka Liljeberg) writes:
>>|> sure I did it right, too. I can easily get a stable picture, but instead
>>|> of a single beautiful one I get two ugly ones, side by side. Both contain
>>|> the visible desktop, only squeezed in half along the X axis. I suspect
>>|> it's the Sierra DAC chips that cause this behaviour. Can anyone confirm
>>
>>I doubt it's the Sierra DAC, the same thing has happened to me with Diamond
>>Speedstar w/o Sierra DAC, when I had the incorrect configuration in Xconfig.
>>With the Diamond Speedstar+ and the Gateway 2000 Crystal Scan monitor, it's
>>a simple process to set up X386, and I'd done it before under Mach anyhow ๐Ÿ™‚
>>The important thing is to get the correct clocks for your card, the ones that
>>are included in the Xconfig probably aren't right for yours. To see what
>>they are (as far as X386 is concerned) take the "clocks" linne out of Xconfig,
>>and run xinit, preferably from a login on one of the serial ports (I used my
>>Linux machine to set up my Mach machine, then reversed the process later ๐Ÿ™‚
>>then make sure you have a line in modedb that corresponds to one of the clocks
>>that X386 shows (probably the last one).
>
> that's right. it's probably up there on the x386 faq! i hoped i was clear
> enough in the readme that you need to examine the xconfig very carefully.
> the clock recognition in linux's x386 isn't stable because i had no clean
> way to disable interrupts while the detection was going on. that's why i
> provided the vga.dbase file with all the clocks. i'm afraid you have to

[ and so on]

Thank you, everyone, for the replies! The amount of support one gets
for Linux sure beats the hell out of many commercial systems!

In my case, the problem was, in fact, with the DACs, as I suspected.
Orest's recent tip about putting 'Vendor "hiclock"' into Xconfig
solved the double image problem immediately. Incidentally, this was not
in the X386 FAQ either. ๐Ÿ˜‰

My mouse still refuses to work in Microsoft mode, but at least disabling
istrip makes the MousesSystems mode, which I prefer anyhow, work ok.

stty -istrip >/dev/ttys1 didn't do the trick, however. This makes
X freeze at startup, for some reason (Linux pre-0.96). Krishna's
little program works all right, though, if you remove the debug
messages or direct them to /dev/null (The Linux pre-0.96 tty driver
can't handle output in graphics mode).

Well, problems aside, X works well enough (Great work Orest!) and the
future looks bright and full of wonders. We live in interesting times. ๐Ÿ˜‰
A big hooray to the man who made it all possible. Hooray, Linus!

Mika


--
Mika Liljeberg Email: [email protected]
Helsinki University [email protected]
Dept. of Computer Science


[next article]
From: [email protected] (Sean Goggin)
Newsgroups: comp.unix.bsd,comp.os.msdos.programmer,comp.os.linux

Subject: Partition ID List - Version 1.5
Message-ID: <[email protected]>
Date: 11 May 92 05:22:03 GMT
Organization: U36 Research
Lines: 51

As per Peter Anvins' request here is version 1.5 of the list of partition codes
If you know of an other entry please email me it at [email protected]

Thank you

Sean

Partition Type Number
no partition 0x00
Dos 12bit FAT Table 0x01
Xenix Root 0x02
Xenix Usr filesystem 0x03
Dos 16bit FAT Table 0x04
Dos 3.3 Extended 0x05
Dos 4.0 large volume 0x06
OS/2 HPFS 0x07
QNX 0x07
Advanced Unix 0x07
AIX file system 0x08
AIX boot partition 0x09
Coherent 0x09
OPUS 0x0A
OPUS 0x10
VENIX 80286 0x40
Novell ? 0x51
Disk Manager(DOS drive utility) 0x51
CPM ? 0x52
Microport SysV/AT 0x52
386/IX 0x63
UNIX System V/386 0x63
MACH MtXinu BSD 4.3 on Mach 0x63
GNU HURD 0x63
Novell 0x64
PC/IX 0x75
Minix 1.1 - 1.4a 0x80
Minix 1.4b - 1.5.10 0x81
Linux 0x81
Linux (Planned Swap partition) 0x82
Amoeba file system 0x93
Amoeba bad block table 0x94
BSDI file system(s), then swap 0xB7
BSDI swap, then file system(s) 0xB8
CCPM Concurrent CPM 0xDB
C.DOS 0xDB
CTOS Convergent Tech OS -Unisys 0xDB
DOS 3.3+ Secondary 0xF2
Bad Track table 0xFF

This list is in the public domain
--
[email protected].U36.Research...


[next article]
From: [email protected] (Magnus Alvestad)

Subject: Re: ESDI drive compatibility
Message-ID:
Date: 11 May 92 17:47:56 GMT
References: <_ssv#[email protected]> <[email protected]> <[email protected]
puddy.uucp>
Sender: [email protected] (Bergen University Newsaccount)
Reply-To: [email protected]
Organization: AlveSoft
Lines: 11
In-Reply-To: [email protected]'s message of 10 May 92 01:37:23 GMT

In article <[email protected]> [email protected] (Stephen Harris)
writes:
My next problem is when I plug in my SECOND disk (same size) I'll be forced
to have 10x64Mb partitions because of the current FS restrictions ๐Ÿ™

Don't worry - the Minix file system is *not* here to stay. I just hope
it will be easy to transfer files from a minix fs to the new one. Will
you do 'mount x x [minix|vfs|msdos\..]' in 0.96..? Linus?

--
| Magnus Alvestad | I am a prolog program. I get all my opinions
| [email protected] | from a small organization in Birmingham.




 December 12, 2017  Add comments

Leave a Reply