Dec 122017
 
Linux news from the Internet.
File INTLX025.ZIP from The Programmer’s Corner in
Category UNIX Files
Linux news from the Internet.
File Name File Size Zip Size Zip Type
INTLX025.TXT 187684 65146 deflated

Download File INTLX025.ZIP Here

Contents of the INTLX025.TXT file


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

Subject: Need testers for FIFO patches
Message-ID: <[email protected]>
Sender: [email protected]
Nntp-Posting-Host: theory.tc.cornell.edu
Organization: Cornell University, Ithaca New York, USA
Date: Mon, 1 Jun 1992 23:16:10 GMT
Lines: 32

I have a set of patches to the 0.96 and/or 0.96a kernels which add
FIFO special files to the kernel, and 'fix' the existing pipe code to
perform the atomic writes required by POSIX. Additional changes have
been made to allow non-root users to create FIFO special files with the
mknod() call, or the program by the same name.

The modified system seems pretty stable on my machine, and I am now asking
for people willing to try my FIFOs out on their own systems. I would, of
course, prefer people who have some application for FIFOs so I can get some
feedback on the standards adhearance and general usability of my FIFOs.

Once I can find three people who have applications requiring FIFOs who can
get the applications to run w/o significant modifications, I will post
the patches and send Linus a copy for possible inclusion in the next
release.

NOTE: The FIFOs implement, to the best of my knowledge, the POSIX-required
behavior. The brain-dead O_NDELAYisms of SysV are not implemented, so
applications written for SysV which open the FIFOs with O_NDELAY will
probably not work.
The ability to open a FIFO with O_RDWR (to force read() to block when there
is no process writing the FIFO) is supported, so servers/daemons which
use a FIFO as a request queue will work.

My thanks go to Brandon Allbery and Leslie Mikesell for bringing the
atomic write requirement to my attention, and to Bruce Evans for making
the need for O_RDWR support clear to me.
Thanks also go to Jim Burns who has agreed to be my first guinea pig.
--
Paul H. Hargrove
[email protected]
"A witty saying proves nothing." --Voltaire


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

Subject: Re: Need testers for FIFO patches
Message-ID: <[email protected]>
Sender: [email protected]
Nntp-Posting-Host: theory.tc.cornell.edu
Organization: Cornell University, Ithaca New York, USA
References: <[email protected]>
Date: Mon, 1 Jun 1992 23:21:20 GMT
Lines: 10

In article <[email protected]> I wrote:
>I have a set of patches to the 0.96 and/or 0.96a kernels which add
^^^^^^
Clarification:
This means the same set of patches works on either (since none of the
relevent files were changed in the transition to 0.96a)
--
Paul H. Hargrove
[email protected]
"A witty saying proves nothing." --Voltaire


[next article]
From: [email protected] (Geoffrey Furnish)

Subject: Re: emacs says: Error in init file
Message-ID: <[email protected]>
Date: 1 Jun 92 21:08:27 GMT
References: <[email protected]>
Sender: [email protected]
Organization: The University of Texas at Austin, Austin TX
Lines: 44

In article <[email protected]> I wrote:

>For some reason when I start emacs, it says "Error in init file". I only
>noticed this yesterday, and I can't recall whether this is new behavior
>or whether it has been doing this for a long time. Anyway, there is no
>.emacs file in my ~ directory, or anywhere else, so I don't understand
>what is causing this.

>If I run emacs from root this problem is gone.

>I have checked to make sure that all the .el and .elc files in
>/usr/local/emacs/lisp are all world readable. Same for all the
>executables, and such. I can't find anything which explains this.
>The fact that the problem goes away from the root account seems to
>speak of a permissions problem, but I've got almost every file on my
>whole system set for world read. Note that aside from this inital
>diagnostic, it seems to work okay.

>Has anyone seen this problem before, and if so, how did you fix it?

I have solved this problem now, and thought I would post a quick
note for whoever was interested.

The problem was that I had set the environment variable LOGNAME to
my last name (furnish) rather than the name of the account I was
using (user--from the mcc interim distribution). Why did I do this?

Two reasons. First, I am using the RCS port to Linux (thanks to whoever
did it) and also CVS (compiles trivially easy if anyone cares). These
require the environment variable LOGNAME to be set. Secondly, I was
using the user account instead of anything else because the last time
I went and customized my system giving accounts to every member of my
family etc, the disk got totalled a few days later. it was my fault, but
I just never got up enough interest in heavily customizing the system
ever since. So I was using the "user" account, but assigned LOGNAME
to my own last name.

For some strange reason, this caused emacs to report "Error in init file".
This was particularly vexing since I did not _have_ an init file.

In any event, the problem is resolved, and if any one cares to use RCS and
CVS, you should be advised of this silliness.

Geoff Furnish


[next article]
From: [email protected] (David Giller)

Subject: Re: mtools in linux
Message-ID: <[email protected]>
Keywords: linux mtools
Organization: Occidental College, Los Angeles, CA 90041
References:
Date: Mon, 1 Jun 1992 23:05:09 GMT
Lines: 21

[email protected] (G. Chen) wrote:
>Can anyone tell me what do the numbers in /etc/mtools mean? I seem
>to have problems using mtools for my C: driver, I keep getting
>"FAT Problem" error. I can use mtools for the B: driver though.
>I haven't tried the A: driver.
>
>The line corresponding to the c: is something like:
>C /dev/hda1 12 0 0 0 0 (It's either 3 or 4 zeros).

The zeros are for drive geometry... you don't need them on a hard drive.
Leave them at zero.

Try setting the 12 to a 16. That is they type of FAT the partition has.

-Dave

--
David Giller, Box 134 | Q: How many Oregonians does it take to screw in a light
Occidental College | bulb? A: Three. One to replace the bulb, and two to
1600 Campus Road | fend off all the Californians trying to share the
Los Angeles, CA 90041 | experience. [email protected]


[next article]
From: [email protected] (Jeffrey L Newbern)

Subject: upgrading questions???
Message-ID: <[email protected]>
Keywords: upgrade, mcc-interim, gcc 2.11
Sender: [email protected] (News system)
Nntp-Posting-Host: w20-575-89.mit.edu
Organization: Massachusetts Institute of Technology
Date: Tue, 2 Jun 1992 00:52:20 GMT
Lines: 13

hi,

i'm using the mcc-interim distribution right now, and i want to try to get x
windows going. the specs say that i need version 96. i think i have properly
upgraded to the 96a release, but i'm not sure. my real problem is upgrading to
gcc 2.11. i can't seem to find the right files to satisfy x. someone else had
mentioned this earlier. is there anywhere that i can get the entire 2.11 releas
e
in a few .tar.Z files? also, could someone tell me the best way to upgrade to
the 96a release, i want to be sure i have everything correct.

thanks much,
jeff newbern



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

Subject: Re: using Linux 0.96a and MCC interim 0.95c+
Message-ID: <[email protected]>
Sender: [email protected]
Organization: Computer Science Department, Stanford University.
References: <[email protected]> <[email protected]>
Date: Tue, 2 Jun 1992 01:15:48 GMT
Lines: 12

In article <[email protected]> [email protected] (Jim Burns) writes:
>> this a limit of the partition table?). Are extended partitions a
>> means to get around this?
>
>They will be when someone gets them working under linux.

fdisk-0.92 (I believe included with the MCC-interim release) supports
extended partitions. I have 6 partitions on my 200MB disk:
10 MB DOS, 10MB swap, 20MB root, and 3 user partitions
(because of the 64MB partition limit). It works great.
--Per Bothner
Cygnus Support [email protected] [email protected]


[next article]
From: [email protected] (Dixie Flatline )

Subject: swap space
Message-ID:
Date: 2 Jun 92 01:59:08 GMT
Sender: [email protected] (Net Noise owner)
Organization: University of Illinois at Urbana
Lines: 12

Help, linuxers!

Can anybody tell me how to remove swap space (file, not partition)?
I'm running out of space...

Thanks
Pete Ehlke
--
The | "Dave, I know that you and Frank have been
Dixie | planning to disconnect me. That is something I
Flatline | cannot allow"
[email protected] | -HAL 9000


[next article]
From: [email protected] (Tim Lines)

Subject: Logitech Bus Mouse
Message-ID: <[email protected]>
Date: Sun, 31 May 1992 14:14:45 -0800
Sender: [email protected]
Lines: 7

I have read various messages on the availability of drivers for the logitech bus
mouse. This leads me to a couple questions:
1. Where can I ftp the patches from?
2. Will these patches be incorporated into future versions of Linux?
#include

* Origin: Programmers Information Exchange (Quick 1:343/27.23)



[next article]
From: [email protected] (The Southern Star)

Subject: Backbone
Message-ID: <[email protected]>
Date: Sat, 30 May 1992 01:26:53 -0800
Sender: [email protected]
Lines: 8

Hello. Your echo is now on the Backbone! I'm sending this message to help get
some of the links established. If you have any private feeds, now is the time t
o switch them to the Backbone.

Thanks,

John

* Origin: The Southern Star - 5 Lines - V32b - 504-885-5928 - (1:396/1)




[next article]
From: [email protected]

Subject: SERIOUS SECURITY PROBLEM (I think)
Message-ID: <[email protected]>
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Date: Tue, 2 Jun 1992 02:06:55 GMT
Lines: 13

I was just logged in as a regular user, and there was a file in my home
directory that belongs to root, and I did an rm to it
rm -f test.c
and this is what I saw:
rm: remove 'test.c', overriding mode 0644?
to which I answered yes, and when i did an ls, the file was gone!!
This didn't happen under .95, .95a, .12, .95c++, but it is happening
under .96 (don't have the .96a yet)
I think this is a security risk, but I am not an expert. ๐Ÿ˜‰
I just wanted everyone to be aware that this can happen, and maybe
it shouldn't?
Jim
[email protected]


[next article]
From: [email protected] (Hongjiu Lu)

Subject: FAQ about X386
Message-ID: <[email protected]>
Date: 1 Jun 92 21:28:37 GMT
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 242


You may want to know this. Please save it if you want to use X386 under Linux.
BTW, only X11R5 (X386 1.2) is supported by Linux.

H.J.
-------------------

From serval!news.u.washington.edu!ogicse!uwm.edu!cs.utexas.edu!utgpu!news-server
.ecf!steve Mon Jun 1 14:17:06 1992
Article: 21523 of comp.unix.sysv386
Path: serval!news.u.washington.edu!ogicse!uwm.edu!cs.utexas.edu!utgpu!news-serve
r.ecf!steve
From: [email protected] (Steve Kotsopoulos)
Newsgroups: comp.unix.sysv386
Subject: Frequently Asked Questions About X386 (with answers)
Message-ID: <[email protected]>
Date: 1 Jun 92 16:12:46 GMT
Article-I.D.: ecf.1992Jun1.161246.8764
Expires: Wed, 1 Jul 1992 03:00:00 GMT
Organization: University of Toronto, Engineering Computing Facility
Lines: 222

[Last changed: 3 Apr 1992]

If you have anything to add or change on the FAQ just let me know.
Send changes to [email protected], please put 'X386' somewhere
in the subject line so that my mail filter will put it in the correct
mail folder.

Frequently Asked Questions About X386 (with answers)
----------------------------------------------------

1. What is X386?

X386 is the port of the X11 server to System V/386 that was
done by Thomas Roell ([email protected]).
It supports a wide variety of VGA boards.

There are 2 major versions: X386 1.1 is based on X11R4,
X386 1.2 is included in MIT's X11R5 distribution (ie. you
don't need to patch it into the MIT source anymore).

2a. Where can I get X386 1.2 (X11R5)?

The X386 1.2 sources are available at any site that distributes
the X11R5 source (too numerous to list here, but includes
export.lcs.mit.edu)

The version 1.2 binaries are available at:
ftp.physics.su.oz.au in Esix_4/x11r5/Binaries
nic.funet.fi in pub/unix/386ix/svr4/x11r5
ftp.win.tue.nl in pub/X11R5-esix

2b. Where can I get X386 1.1 (X11R4)?

X386 1.1 is available both as a binary release or as source
patches (apply patches to MIT's X11R4 source tree).
The fonts are available separately, since they are common
for all operating systems.

The official ftp site is flop.informatik.tu-muenchen.de
(131.159.8.35), /pub/i386/X11R4
808424 bytes X386.1.1b.tar.Z source patches
3566993 bytes X386core.SVR4.tar.Z binaries for SVR4
7348521 bytes X11R4core.tar.Z 1.1a binaries
395521 bytes X386.1.1b.isc.Z 1.1b server binary
3307520 bytes X11R4fonts.tar fonts

It is also available on the following machines:

- piggy.ucsb.edu (128.111.72.50), /pub/X386.
808424 bytes X386.1.1b.tar.Z source patches
3566993 bytes X386core.SVR4.tar.Z binaries for SVR4
3307520 bytes X11R4fonts.tar fonts
Please use after 6pm Pacific, as these are large files.

- ftp.win.tue.nl:/pub/X11R4_386

- NSTAR Public Access Unix 219-289-0287
{[email protected], ..!uunet!news.rn.com!larry}

3. What operating systems does it run under?
X386 will run on following Unix operating systems:
- Dell SVR4
- Interactive 386/ix, 2.0.2 or later
- ESIX
- SCO Unix (release 3.2v2.0)
- AT&T SVR4 and SVR3.2
- Microport SVR4
- MST SVR4

4. What VGA boards are supported?

ET3000 (for ex. GENOA 5300/5400)
ET4000 (Tricom, STB PWR Graph, Sigma Legend, etc.)
GVGA (Genoa 6400)
PVGA1A (Paradise VGA Professional)
WD90C00 (Paradise VGA 1024)

5. Why won't my xterm run properly?

If your kernel is not built with the consem module, you should define
CONSEM=no in you enviornment. Otherwise xterm won't run.
csh users should use 'setenv CONSEM no'

6. I can't compile X programs. Am I missing a library?

you must include the BSD (Sockets) API by including -linet in your link
list. Also, for anything that uses the /dev/ptmx 'clone' devices,
use -lpt.
SCO Unix uses -lsocket for the socket library.

In summary, add '-linet -lpt'
after the -lX11. (It won't work if it is not AFTER the -lX11)

Or use instead -linet '/usr/lib/X11/X386/etc/inetemul.o' which
includes some dummy functions, which always will return an error,
but are *very* good for linking.

7. Everything works fine if I use xinit, but I can't use xdm.
It gives me an error that it can't open the mouse.
Why would the server work fine under xinit, but not xdm?
How do I fix it? (I am using the Xque* drivers under SVR4)

This is caused by a bug with the Xque drivers in SVR4.
Dell's SVR4 has fixed this problem (which exists only
with X386 1.1, not X386 1.2).

[email protected] (David E Wexelblat) is using the following
xdm (X11R4pl18) patch successfully:

*** Save/util.c Tue Nov 12 20:24:17 1991
--- util.c Tue Nov 12 20:25:37 1991
***************
*** 165,175 ****
--- 165,177 ----

CleanUpChild ()
{
+ #ifndef SYSV386
#ifdef SYSV
setpgrp ();
#else
setpgrp (0, getpid ());
sigsetmask (0);
+ #endif
#endif
#ifdef SIGCHLD
(void) signal (SIGCHLD, SIG_DFL);

8. How do I find out what clock frequencies are on my VGA board?

X386 1.1 includes a DOS program, /usr/lib/X11/X386/etc/clock.exe, that
you can run (under DOS of course!). It will report all the clock
frequencies that it can find. You can then use this information
when setting up your Xconfig file.
X386 1.2 will automatically find this information when it boots.

David E Wexelblat ([email protected]) maintains a database of known
clock settings for VGA cards and monitor settings. This database is
available from him (for the latest copy), and is kept on
export.lcs.mit.edu in ~/contrib/X386.modeDB.Z, which is updated
occasionally. Obtain a copy of this database. It just might have the
settings you need. If you create new settings, please send them to
David for inclusion in the database.

8a. OK. I have my VGA clocks, and my monitor specifications. How do I
generate mode lines for Xconfig?

First check out the mode database described for question 8. It has lots
of monitor timings contributed by X386 users. It also has the VESA
standard definitions. If the database does not have definitions for
your monitor, try the VESA timings.

If neither of these options work, the CONFIG file with the X386 contains
a tutorial on how to come up with these timings. It may be helpful to
start with setting that almost work, and use this description to get
them right. When you do, send the information to David Wexelblat for
inclusion in the database.

9. How do I switch session under SCO? The normal sequence does not work.

Try ALT-CTL-F{n}

10. I can't get my bus mouse to work!

I had a lot of trouble too! So far, the only way I could get a
bus mouse to work (on SVR4) was to use the Xque drivers and set
up my /usr/lib/mousetab for a bus mouse 'vt00 bmse'.
(This involved re-installing the mouse driver for bus mouse)

People running ISC have reported that using an Xconfig of
'Busmouse "/dev/logi"' works too.

11. Any tips on compiling X11R5?

1. You will need lots of disk space (around 150Meg for the MIT core).

2. If using the standard C compiler, make sure that you don't have
/usr/ucb before /bin in your path.

3. Follow the advice of Steve Hite ([email protected]):
"DON'T use the C shell on Esix 4.0.3a. The X library is HUGE and
csh complains about too long of an argument list. The Korn shell
will work just dandy."

4. Look at mit/server/ddx/x386/README and follow any instructions
for your particular platform.

5. Change the line in server/Imakefile that reads:
SYSLIBS = -lm /usr/ucblib/libucb.a
to read:
SYSLIBS = -lm -lc /usr/ucblib/libucb.a

This ensures that the directory reading routines from libc.a are
used intead of the ones in libucb.a (BSD version of readdir).

You might want to extract ndbm.o, bcopy.o and bcmp.o from
libucb.a (ar x) into a private library instead (or use GNUdbm).


6. There is an unofficial bug fix for the polygon drawing routine,
it is available for ftp from:

ftp.win.tue.nl:/pub/X11R5/fixes/unofficial-X386fix
ftp.physics.su.oz.au:/Esix_4/x11r5/Unofficial_x386_fill_patch.Z

7. You may want to disable building the PEX stuff, since it
runs miserably slow on most machines.

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

Please send me anything else you want added to this FAQ.
(especially if you had a problem that someone else was able to help you with)

Please do not send me questions that are not answered in the FAQ. Instead,
post your question to the net, and send me the answer when you get it.

Thanks to all the people who already sent me corrections or additions.
--
Steve Kotsopoulos mail: [email protected]
Systems Analyst bitnet: [email protected]
Engineering Computing Facility uucp: uunet!utai!ecf!steve
University of Toronto phone: (416) 978-5898


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

Subject: Re: /lib/libm_v2_v11c
Organization: Wimsey
Date: Tue, 02 Jun 1992 02:37:27 GMT
Message-ID: <[email protected]>
References:
Lines: 2

run inst2.x in /usr/gcc/install or wherever it is on your system - I had
this problem this weekend...


[next article]
From: [email protected] (David Giller)

Subject: Re: SERIOUS SECURITY PROBLEM (I think)
Message-ID: <[email protected]>
Organization: Occidental College, Los Angeles, CA 90041
References: <[email protected]>
Date: Tue, 2 Jun 1992 02:26:18 GMT
Lines: 30

[email protected] wrote:
>I was just logged in as a regular user, and there was a file in my home
>directory that belongs to root, and I did an rm to it
> rm -f test.c
>and this is what I saw:
> rm: remove 'test.c', overriding mode 0644?
>to which I answered yes, and when i did an ls, the file was gone!!
>This didn't happen under .95, .95a, .12, .95c++, but it is happening
>under .96 (don't have the .96a yet)
>I think this is a security risk, but I am not an expert. ๐Ÿ˜‰
>I just wanted everyone to be aware that this can happen, and maybe
>it shouldn't?

No, this is an exception to the normal rules of permissions. Actually,
I guess it IS a normal rule, but it's not part of the basic one, 'if you
don't own it, you can't have it.'

If there is a file in a directory owned by you, you have rights to it, no
matter who owns it. I forget just to what extent the permissions are, but
you have at least the right to delete it.

Notice that you couldn't delete something in the /bin directory, for example.

-Dave

--
David Giller, Box 134 | Q: How many Oregonians does it take to screw in a light
Occidental College | bulb? A: Three. One to replace the bulb, and two to
1600 Campus Road | fend off all the Californians trying to share the
Los Angeles, CA 90041 | experience. [email protected]


[next article]
From: [email protected] (David Giller)

Subject: BSD lisence and copyright: Kernel hacking/merging
Message-ID: <[email protected]>
Organization: Occidental College, Los Angeles, CA 90041
References: <[email protected]> <[email protected]
.oxy.edu>
Date: Tue, 2 Jun 1992 02:34:56 GMT
Lines: 34

I have a couple of questions for the Legal-aware out there.

I am going away from my internet access for the summer (sniff). I am taking
about 40Mb of various sources and info, however, and I will be taking a
CD-ROM of X11R5 and all the GNU sources with me, so don't anyone worry
about my boredom, I've got about 650Mb of stuff to keep me busy.

However, one of the things I'd like to do is play with merging some of the
BSD and Linux sources. Not really merging per se... I want to take some of
the working pieces of BSD and graft them into Linux.

What I'm wondering is, what are the legal ramifications if I want to give
out what I've accomplished? For example, I want to port the CD-ROM
driver to Linux. It is under the BSD Lisence.

Can I give out copies of the finished project? If it were good enough,
could Linus include it in, say, an MCC release of Linux? Would Linux
have to start displaying Regents-Of-UC-Berkeley copyright notices?

I'd like to add to Linux, but I don't want to cause the Linux lisence
to change in the process. And I don't have the knowledge to write a CD-ROM
driver myself (though a GREAT deal of it will be pretty close!)

Another idea I'd like to play with (god forbid) is installable device drivers.
I know, I know, it's almost like a DOSism compared to the UNIX way of doing
things. BUT, it's a convenient DOSism, and it could really help make
Linux configuration easier.

-Dave
--
David Giller, Box 134 | Q: How many Oregonians does it take to screw in a light
Occidental College | bulb? A: Three. One to replace the bulb, and two to
1600 Campus Road | fend off all the Californians trying to share the
Los Angeles, CA 90041 | experience. [email protected]


[next article]
From: [email protected] (David Giller)

Subject: Linux and the 'standard' filesystem
Message-ID: <[email protected]>
Organization: Occidental College, Los Angeles, CA 90041
References: <[email protected]> <[email protected]
.oxy.edu>
Date: Tue, 2 Jun 1992 02:42:21 GMT
Lines: 28

I've been playing around with Linux, and am about to take a dive into the
kernel in a big way. So I;ve got a few questions about where Linux is
headed, so I don't step on anyone's toes in the process (or waste my time).

Right now, Minix is _the_ filesystem for Linux. However, with VFS on the
horizon, it seems we will be able to hook in any filesystem we want.
There's full source for UFS available with 386BSD (see my other post)
and others have been working on a 'bigfs' to alleviate the minix filesystem's
most grievious problems.

However, is there any consideration being given to planning a move to a
different 'standard' filesystem for Linux? Minix is just inadequate for
many (most?) needs. For example, I'm working on a connectionist network
simulator, and it NEEDS long filenames (well, it's damn inconvenient without
them). Now, I'm trying to edit the sources, but we're talking about 7-8Mb
of code here. It's much easier to just rewrite the filesystem! Patching
the Minix fs for longer names is possible, but that's a kludge.

I digress. Well, are there any discussions? Pardon me for not searching
on my own, but I've got to leave in a week and a half, and I'm in finals
right now.

-Dave
--
David Giller, Box 134 | Q: How many Oregonians does it take to screw in a light
Occidental College | bulb? A: Three. One to replace the bulb, and two to
1600 Campus Road | fend off all the Californians trying to share the
Los Angeles, CA 90041 | experience. [email protected]


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

Subject: Re: FAQ about X386
Organization: Wimsey
Date: Tue, 02 Jun 1992 03:10:52 GMT
Message-ID: <[email protected]>
References: <[email protected]>
Lines: 2

Is something wrong with banjo.concert.net? I can't seem to connect to
it these days... Is it not at 192.101.26.5?


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

Subject: tcsetpgrp() bug?
Message-ID: <[email protected]>
Date: 2 Jun 92 03:49:32 GMT
Organization: Georgia Institute of Technology
Lines: 9

Is there a problem w/the tcsetpgrp(0, mypid) call that screen3.2 makes when
it creates a new window? It fails for certain ptys after initially working
for them, and only a reboot clears it. Happens for root and non-root. Errno
= 25 (invalid ioctl). Thanx.
--
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] (Theodore Ts'o)

Subject: Re: SERIOUS SECURITY PROBLEM (I think)
Message-ID: <[email protected]>
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Date: Tue, 2 Jun 1992 03:53:19 GMT
Lines: 25

From: [email protected]
Reply-To: [email protected]
Date: Tue, 2 Jun 1992 02:06:55 GMT

I was just logged in as a regular user, and there was a file in my home
directory that belongs to root, and I did an rm to it
rm -f test.c
and this is what I saw:
rm: remove 'test.c', overriding mode 0644?
to which I answered yes, and when i did an ls, the file was gone!!

Standard Unix semantics: If you have write access to the containing
directory, you can delete any file in the directory. BSD systems have
an enhancement where if the sticky bit is set on a directory, only the
owner of the file can delete it. This is why BSD-derived systems have
/tmp set to mode 1777.

Adding this functionality to Linux would not be difficult. A few lines
of code to fs/namei.c.

I think this is a security risk, but I am not an expert. ๐Ÿ˜‰

Nope, no security risk; at least not from this behavior.

- Ted


[next article]

Subject: Stereo sound DAC(Was Re: Help for X11)
Message-ID:
From: [email protected] (Erik Green)
Date: 1 Jun 92 22:57:31
Distribution: comp.os.linux
Organization: Long polymers in double-helix formation
Nntp-Posting-Host: att2.cs.mankato.msus.edu
Lines: 51


For those of you that missed or skipped it, I found a circuit for a stereo
sound player that works off of the parallel port enclosed with an Amiga .MOD
file player for DOS. I thought that it'd be good for Linux because:

A: It's really cheap to build
B: It works on the parallel port, so we don't need to write a new kernel
driver for it.
C: It's a nifty hardware hack, to go with all of the software hacks already
in Linux.

Well, since so many people have mailed me asking for more information,
I thought I'd post the sources for the DAC chips needed for the circuit.

To get the circuit itself, you can get the file MP219B.ZIP from Simtel or
any of its mirrors in the directory msdos/sound. This is an archive
containing Modplay 2.19b, a great .MOD file player for DOS. The circuit
(actually, several circuits) are included in HARDWARE.DOC(I didn't post
this because I wasn't sure I could distribute it apart from the rest of the
archive). The one you want is the Stereo-on-1 board. As one person wrote
me, real-time applications may not run well in a multitasking environment,
but I think it'll work well for at least FM sound synthesis and single
channel samples,if not for full 4-channel stereo sound.

The chips listed for the circuit are 2 Plessey ZN428 8-bit DACs. The docs
say any D/A with input latches should work, as long as it latches the data
bus when pin 4 is low. If you want an exact match, you can order from:

JDR Microdevices
2233 Samaritan Drive
San Jose, CA, USA, 95124
(800)-538-5000 (Orders)
(800)-538-5002 (Tech support)

Theirs is a good catalog for _anyone_ to have, IMHO.

Part numbers DAC-0800($1.99 US) or DAC-0808($2.49 US) should work, with
the DAC-0808 being recommended as the closer match. You can also order
the other small components and the cables you need from them. You may
need to get some things elsewhere. If you hate soldering, you can
assemble the circuit on a solderless breadboard, price about $2 to $6
US. If it works well for me I may etch a board for it.


-Erik

--
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] (Jim Burns)

Subject: Re: How to create X11 applications with shared libraries ?
Message-ID: <[email protected]>
Date: 2 Jun 92 04:22:11 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 13

in article <[email protected]>, [email protected] (David Engel) says:

> Close, it's called ldd. According to the SunOS manual page, it stands for
> list dynamic dependencies.

Unfortunately, to use it, you have to add a supplied routine to your lib
(libc, I think), and it will only work for anything compiled w/the new
routine, which lets out all the mcc-interim stuff.
--
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: Re: /lib/libm_v2_v11c (offical GCC ftp site for Linux)
Message-ID: <[email protected]>
Date: 2 Jun 92 01:35:30 GMT
References:
Sender: [email protected] (USENET News System)
Organization: Washington State University
Lines: 191

In article [email protected] (J
ay Pfaffman) writes:
>I'm trying to get X1.0 running. I downloaded Xbase1.0 & Xuser1.0 from
>banjo and I seem to be missing /lib/libm_v2_11c. I can't find it in
>any of the X or GCC files. I think I've tried (at least) the obvious
>ones, but I imagine I'm mistaken.
>
>Where is /lib/libm_v2_v11c?

Read the version number v2_v11c as 2.11c.

>
>thanks.

In case somebody doesn't know. The official gcc ftp site for Linux is
banjo.concert.net under /pub/Linux/GCC. For now, gcc 2.11c is
officially released. There are also a few testing releases. To use the
the testing gcc, you MUST join the GCC channel.

Before you install the gcc, please read all the release notes up to
the version you are installing.


H.J.
----------instruction-------


SAVE THIS TO A FILE!!!!!!

Linux-Activists list is now maintained by Mail-Net

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

Fast way to subscribe or unsubscribe

Send mail to:

[email protected]

Subscribe:

Your mail HEADER must contain the line or
the first line of your mail body muse be:

Mn-Admin: join normal

Unsubscribe:

Your mail HEADER must contain the line or
the first line of your mail body muse be:

Mn-Admin: leave normal

Problems?

Mail to:

[email protected]

The current list admin can be reached through that
address.

-----
arl

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

Why this system?

I noticed people were asking questions like "who was porting
SCSI driver to Linux?" or "what is going on with X?". This
system tries to make contacts easier and possibly we avoid
duplicate work.

It saves time to admin mailing lists.

Mail-Net gives for the user possibility to create his own
mailing lists (here called channels).

Linux-Activists mailing list:

The former mailing list is just one channel on the system.
The default channel is called "normal".

Send your linux-activists mail to:

[email protected]

and remember to have at your mail header or
the first line of your mail body

Mn-Key: normal

How to use mail-net software:

You send mail to the server defining what you want. Now
I do have only commands sent within mail header lines or
the first line of your mail body.

The header lines are those lines which contain fields
like "To:" or "From:".

I hope, that the admin and mail goes to the same address,
if it won't work - I have to make other e-mail address
for requests.

Commands:

MN-Admin: create channel_name

The system creates a channel for your discussion
interests, like

MN-Admin: create SCSI

would create channel SCSI (where you could speak
about SCSI w/o taking "looser mail"/"line noise").

Join is automatically done for the channel creator.

MN-Admin: delete channel_name

๐Ÿ˜‰ Channel deleting is not needed now ...

MN-Admin: join channel_name

Join to some discussion channel. The channel must
exist prior joining ...

MN-Admin: leave channel_name

You leave the channel.

Like leaving the main channel i.e. Linux-Activists list

MN-Admin: leave normal

you can still use other channels ... naturally.

Admin commands can be sent to the mailing list
([email protected]) or to the
request address ([email protected]).

Mailing to the Linux Activists list

The e-mail address still is:

[email protected]

You need to have definer MN-Key: channel_name if you would
like you mail to go to some channel [, else it will go to the
default "normal" channel.] You need to have key for the
default channel too, that is:

MN-Key: normal

This filters out the multiple error messages I managed to make
having sender as the whole list (and all the mailers sent
and are sending error messages to the whole list).

Need to send mail to the SCSI channel:

MN-Key: SCSI

Notice

"X-MN-Key:" etc definers are supported, Mn-Key, and Mn-key
are valid definers.


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

Going-to-be [priority level]

info service to find out channels, their topics,
users and mail traffic generally. [1]

address change handling not implemented yet ..
going to be real soon now. [2]

multiple commands in one mail not implemented (rsn). [3]

multiple parameters for a command not implemented
yet (rsn). [3]

Ideas?

Send your ideas to:

[email protected] // if there's any problems with
// this software I try to correct
// them instantly.


[next article]
From: [email protected] (Daniel Jimenez)

Subject: "X won't work"-type newbie question
Message-ID: <[email protected]>
Date: 2 Jun 92 04:29:44 GMT
Organization: University of Texas at San Antonio
Lines: 22

I got the binaries for X off of banjo and installed them as per the
instructions. I (hope I) made the correct modifications to Xconfig
for my mouse and VGA card (1200bps microsoft mouse, Paradise VGA card).
I'm trying to see if I can get X to work within just three megabytes
of RAM (yes, I know it would be horribly slow, even if it did work,
bit I'm getting more memory soon...). Six megabytes of swap space.

With the 'bootimage' boot disk provided with the X stuff, I try
startx and the server dies with "No screens found."

With the 0.96a boot disk, the server just dies and says "giving up."

If I comment out the part about the mouse in Xconfig, I can get the
X server to just hang.

Does anyone know what I'm doing wrong? I saw someone earlier had the
"no screens" problem, but I haven't seen any responses. Is there
a FAQ for X386?

Thanks.
--
Daniel Jimenez


[next article]
From: [email protected] (Anthony Rumble)

Subject: All asking about Oak VGAand X386..
Message-ID:
Keywords: oak, vga, X386, X11
Sender: [email protected]
Nntp-Posting-Host: extro.ucc.su.oz.au
Organization: Sydney University Computing Service, Sydney, NSW, Australia
Date: Tue, 2 Jun 1992 02:01:50 GMT
Lines: 9

ok.. basically.. If you have an Oak.. and you would like
to run X386... You cant.. There is as yet
NO support for Oak cards in X386..

So either you have to fork out for a different VGA

card. or Add Oak support to X386.. I couldn't do this
so someone else will have to..

Anthony Rumble


[next article]
From: [email protected] (Greg Naber)

Subject: Re: Backbone
Message-ID:
Date: Mon, 01 Jun 92 18:37:41 PST
References: <[email protected]>
Reply-To: [email protected] (Greg Naber)
Organization: S'Qually Holler's, Renton WA
Lines: 39

[email protected] (The Southern Star) writes:

> Hello. Your echo is now on the Backbone! I'm sending this message to
> help get some of the links established. If you have any private feeds,
> now is the time to switch them to the Backbone.

> * Origin: The Southern Star - 5 Lines - V32b - 504-885-5928 - (1:396/1)

For those who have no idea what this is doing in comp.os.linux, this is
one of three fidonet 'backbone' operators. The newsgroup of comp.os.linux
is being gated to the fidonet world. Something that I have been doing
for the last month and the 'backbone' is the Fidonet way of passing
newsgroups (echo's).

Fidonet is a conglomeration of over 16,000 systems world wide, it is a
store and forward type network, using ma bell as the medium for sending data.

One of the 'rules' for Fidonet to get an echo on the backbone, is that it
must have a moderator. I have assigned this position to myself;-) So if there
are any complaints from the newsgroup side, please address them to:
[email protected]

(not sure if my return address above will get to me, as
that address is for gating and testing purposes)

For those on the fidonet side, I have set-up here as a 'clone' of
tsx-11.mit.edu and banjo.concert.net. The files you see metioned will be
available here via f'req or 9600+ user D/L. 1-206-235-0189, As soon as I can
get them and make them available.

The latest FAQ is available as 'magic name" of FAQ



---
S'Qually Holler's BBS Renton, Wa. 1-206-235-0270 USR HST D/S
1:343/[email protected]
1:343/[email protected]
[email protected]


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

Subject: Re: breakout lists/discussion channels
In-Reply-To: [email protected]'s message of 2 Jun 92 02:07:41 GMT
Message-ID: <[email protected]>
Sender: [email protected]
Organization: St. Olaf College; Northfield, MN USA
References: <[email protected]> cs.hut.fi>
Date: Tue, 2 Jun 1992 04:45:03 GMT
Lines: 26


In article [email protected] (Ari Lemmke) wri
tes:
From: [email protected] (Michael K. Johnson) Date: 27 May 92 03:47
:37 GMT:
: The main lists for linux are the [email protected] lists
^^^^^
: -- They are really one list with several "channels", or at least that
: is the way the guy who wrote the software (Ari Lemke,
: [email protected]) thinks of it.

comp.os.linux is the MAIN information source for Linux,
I have NOT said anything else - I wouldn't even write
the damn RFD for comp.os.linux if I had other opinion ...
(you'd possible still thinking if it's worth of having
comp.os.linux ๐Ÿ˜‰ wanna have old alt.os.linux articles?)

Excuse me, I said, "lists." comp.os.linux is a /newsgroup/. I have
gotten more flak about this one than I can beleive, from people who
have not paid attention to what I wrote. A mailing list is /not/ the
same thing as a newsgroup. I /never/ said that
[email protected] was the "MAIN information source for
Linux," as I am not that crazy. Come on people, read what I write,
not what you think I write. Off my case, all of you, please.

'Nuff said. flame mode off. Thanks for your patience.

michaelkjohnson


[next article]
From: [email protected] (Michael O'Reilly)

Subject: Re: "X won't work"-type newbie question
Message-ID: <[email protected]>
Sender: [email protected] (USENET News System)
Nntp-Posting-Host: tartarus.uwa.edu.au
Organization: University of Western Australia
References: <[email protected]>
Date: Tue, 2 Jun 1992 05:11:04 GMT
Lines: 15

[email protected] (Daniel Jimenez) writes:
:
: With the 'bootimage' boot disk provided with the X stuff, I try
: startx and the server dies with "No screens found."

I presume you have made /dev/tty0. If you have then, your video card is
very likely not supported. I THINK the message 'No screens found' is
printed by X if it doesn't recognize your chipset.. At least this is the
message it gives when I tried it by accident on a Trident card.

: Does anyone know what I'm doing wrong? I saw someone earlier had the
: "no screens" problem, but I haven't seen any responses. Is there
: a FAQ for X386?

: Daniel Jimenez


[next article]
From: [email protected] (the MOST SIGNIFICANT bit)


Subject: a problem with /dev/kmem
Message-ID: <[email protected]>
Sender: [email protected] (netnews admin account)
Organization: Nyx, Public Access Unix at U. of Denver Math/CS dept.
Date: Tue, 2 Jun 92 05:25:30 GMT
Lines: 21

I've looked in the FAQ, and didn't find anything, so...

Last night I did something quite brilliant, I accidently nuked
/dev/kmem and so far haven't found any documentation that lists
the major/minor numbers. So, could anyone tell me what to do to
re-make it?

Also on the subject of devices, how do I get linux to see the
computer's ethernet card? Assuming I can make a /dev/eth, will
ka9q work?

I'm using the 0.96a kernel on a 386-25, after starting with the
MCC interim distribution.

Thanks for any help.
------
[email protected] <-- all mail gets here eventually.
[email protected]
[email protected]




[next article]
From: [email protected] (Dan Miner)

Subject: Linux crashing. Help needed.
Summary: Pcomm and zmodem problems.
Message-ID: <[email protected]>
Date: 2 Jun 92 05:26:45 GMT
Sender: [email protected]
Organization: University of Missouri - Rolla
Lines: 24


Here is the scoop. I have finally upgraded to
Linux 0.96a and have not had too many problems. The MOST
annoying one is this:
When downloading with Pcomm (I think 0.95a version) and
rz (zmodem receive, got from a netter), if rz aborts because of
too many errors or what I get this:
Pcomm doesn't respond but I can type junk in that session.

In ALL other "windows", I can't type anything.

If I press the scoll lock and get the debug info, it whole
thing stops. I can't even go to any virtual consoles after that.

Any ideas?

Dan
[email protected]

--
-------------------------------------------------------------------------
Dan Miner email:
[email protected] (prefered)
Computer Science [email protected]


[next article]
From: [email protected] (Hubert Bahr)

Subject: ESDI disk ? anybody
Message-ID: <[email protected]>
Sender: [email protected]
Organization: engineering, University of Central Florida, Orlando
References: <[email protected]>
Date: Tue, 2 Jun 1992 05:25:02 GMT
Lines: 15


Does anybody use or have tried to use an ESDI disk under Linux.

I've heard that there are PC controllers that appear to dos as
an ST506 or MFM controller. Can anybody verify. If this is true
will Linux treat this controller the same way. I would appreciate
recomondations if available. I've been offered the use of a Maxtor
3380E old but 320 Meg with resonable access times. If it works it
would make recompilation of X11, GCC, EMACS etc. much easier and possible
online duplication of sources for easy generation of Diffs.

Thanks for any input

Hubert A. Bahr [email protected] 132.170.200.67



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

Subject: Re: SERIOUS SECURITY PROBLEM (I think)
Message-ID: <[email protected]>
Sender: [email protected] (The Daily Planet)
Nntp-Posting-Host: ophelia.cs.colorado.edu
Organization: University of Colorado at Boulder
References: <[email protected]>
Date: Tue, 2 Jun 1992 05:27:16 GMT
Lines: 23

In article <[email protected]> [email protected] writes:
> From: [email protected]
> Reply-To: [email protected]
> Date: Tue, 2 Jun 1992 02:06:55 GMT
>
> I was just logged in as a regular user, and there was a file in my home
> directory that belongs to root, and I did an rm to it
> rm -f test.c
> and this is what I saw:
> rm: remove 'test.c', overriding mode 0644?
> to which I answered yes, and when i did an ls, the file was gone!!
>
>Standard Unix semantics: If you have write access to the containing
>directory, you can delete any file in the directory. BSD systems have
>an enhancement where if the sticky bit is set on a directory, only the
>owner of the file can delete it. This is why BSD-derived systems have
>/tmp set to mode 1777.
>
>Adding this functionality to Linux would not be difficult. A few lines
>of code to fs/namei.c.

This is the behavior that Linux follows. However, jgifford's
homedirectory probably wasn't set sticky or world writeable.


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

Subject: SCSI FAQ
Message-ID: <[email protected]>
Sender: [email protected] (The Daily Planet)
Nntp-Posting-Host: hazelrah.cs.colorado.edu
Organization: University of Colorado at Boulder
Date: Tue, 2 Jun 1992 05:31:21 GMT
Lines: 239


Q : What SCSI hosts are supported?

A : The Adaptec 1542 (The EISA bus master 17xx may work with
minor changes), Seagate ST01/ST02, (These are
repackaged Future Domain 88x series boards.
If you have an 8 bit future domain controller,
you may be able to modify the seagate driver
to detect your card to. If any one does this,
please send diffs.), and Ultrastor 14F are supported in the
stock distribution.

Also, patches are available for the WD7000 FAAST, and Future Domain
16 bit controllers.

At this point, the Adaptec 152x series, and 17xx series
DO NOT WORK.

Q : Which disks ?

A : Anything that works with your host adapter electronically, and
is new enough to support a December 1985 draft of SCSI-I.

Sysquest removable drives are known to work. However, DO NOT
remove a mounted cartridge. This will trash your filesystem.

CDROM should also work, but we don't support the High Sierra
filesystem.

Q : What about SCSI tapes ?

A : The stubs are sort of there. The detection stubs are there,
there is a raw SCSI ioctl which forms the basis for much of the
tape driver (it goes between user / kernel space, where as
the block driver code operates entirely in kernel space),
all that remains is completion of enough ioctls
to support an mt command. It's 1-2 days of coding
and debugging once someone gets around to it.

Q : How do I get SCSI information?

A : Subscribe to the SCSI channel of the linux-activists mailing list.

mail [email protected]

And put

X-MN-Admin: join SCSI

in the header.

Mailing [email protected]
will get you a list of commands.

Q : How do I partition the disk?

A : The linux partitioning programs don't work, as you might have noticed.
Partition the disk under DOS (pfdisk compiled for DOS will work),
or if you are a masochistic hacker type and familiar with your SCSI
BIOS's remapping scheme, use dd and a binary capable editor =8^)

The problem with partitioning SCSI disks and Linux is that Linux
talks directly to the SCSI interface. Each disk is viewed
as the SCSI host sees it : N blocks, numbered from 0 to N-1,
all error free. There is no portable way to get disk
geometry.

However, DOS doesn't like things like this, it demmands that
BIOS present it with a normal Cylinder / Head / Sector
coordinates. So, BIOS does, and it comes up with some fabrication
that fits what DOS wants to see. You don't want to disagree with
what BIOS thinks when you write the partition table.

Q : Where is the latest version maintained?

A : headrest.woz.colorado.edu:/pub/linux

Since SCSI made it into .96, this is mostly alpha/beta
test for new things. Right now,
an interrupt driven Seagate driver is what's there.

It works for me, but your mileage may vary.
Join the list before you grab anything.

Q : what about extended partitions?

A : not yet. Let's just say that I almost have a very
large drive up on a SCSI->SMD bridge, and
that it's a priority.

Q : I get a message saying READ CAPACITY FAILED.
What does this mean?

A : One of two things. First thing is that there is a bug in
the midlevel error handling code. When an error
condition is returned, things aren't retried. Some
SCSI disks get unhappy at the sight of the
initialization routines, and will return an
error condition for the first "real"
command that they run. INQUIRE doesn't count,
since that will allways return successfully,
irregardless of pending error conditions,
unless your drive is Kaput.

You could also have a flakey disk, but this is
unlikely.

Q : What are the major / minor numbers for SCSI drives?

A : Right now, it is theoretically possible to have
21 SCSI devices in the system. With SCSI-
MFM / ESDI / SMD bridges this can be even
more drives, as each id may have several drives on it
each set up as a LUN of that id. Needless to say,
if we have allowed for 16 partitions per SCSI drive,
we're pushing the limits on minor numbers. Therefore,
a yucky dynamic numbering scheme is used.

Block device major 8 is used for SCSI drives. I
personally like /dev/sd[n][p], where n is the drive
number, and p partition letter - but this is
entirely personal preference.

Minors are assigned in increments of 16 to SCSI disks
as they are found, scaning from host 0, ID 0 to host
n, ID 7, excluding the host ID. Most hosts
use ID 7 for themselves.

A minor where minor mod 16 = 0 is the whole drive,
where minor mod 16 is between 1 and 4, that partition,
and extended partitions are not yet implemented.

Example : I have four SCSI disks, set up as follows
Seagate ST02, ID=0
Seagate ST02, ID = 5
Ultrastor 14, ID = 0
Adaptec 1542, ID = 0

The first disk on the seagate at ID 0 will become
minors 0-15 inclusive, the second at ID5 16-31
inclusive, the disk on the Ultrastor 32-47, on the
Adaptec 48-63.

Q : I can't use more than 2 SCSI disks.

A: Change MAX_SD in sd.h, MAX_SCSI_DEVICE in scsi.h, and recompile the kernel.

Q : How do I reduce kernel bloat and eliminate the drivers I
don't want?

A : Simply #undef CONFIG_DISTRIBUTION in
include/linux/config.h, and define the
macros for the SCSI hosts you want
enabled.

Q : What about bugs?

A : It works on MY hardware. It works on Tommy Thorn's
Adaptec system, Dave Gentzel's Ultrastor, Thomas
Wuensche's WD7000, Rick Faith's Future Domain,
and quite a few other systems.

Bugs that are there will mostly be very hardware
specific, and nasty to track down. SCSI should
be basically error free - consequently, the
error code has not been heavily tested, and
there are known bugs in it.

If you have found a bug, please mail it
to the mailing list with specifics of
your hardware. Other people may have
the same problem, a solution, etc.

Chances are I will provide you with a debugging
version of the kernel, which will dump out important
SCSI information and let me see what exactly is
going on.

Q : What are the known bugs?

A : The seagate driver should fail on i486, and i486sx machines. This is because

the seagate uses memory mapped I/O, and if it gets the chance, the
i486 will go for the internal cache, bypassing the memory mapped
registers of the Seagate SCSI host.

386 cache machines will not have problems IF the Seagate's address
space (typically C8000 - CAFFFF) is marked "non cacheable."

This can be set in the XCMOS of most machines.

The error routines are imperfect, and this causes READ CAPACITY
to fail in sd_init.

Many of the drivers cannot abort a command that is in progress - if some
thing
goes haywire, all drives off that controller will hang.

Q : Why can't I swap to a SCSI disk?
Q : Why can't I mount a SCSI disk as root?

A : This was due do a race condition that has since been fixed (I think).

It only manifested itself in drivers using the command()
interface (see bellow), but this interface isn't used very much
as it was there mostly for develeopment.

Q : Why does Linux "hang" when the disk is accessed?

A : There are two possible interfaces between mid and lowlevel SCSI routines.
These are the queue_command() and command() interfaces.

The difference between the two is that queue_command()
provides a way to do generic, interrupt driven SCSI,
whereas command() is not interrupt driven.

The latest Seagate driver (in testing), WD7000 and Future Domain 16 bit
use the queue_command() interface, Adaptec has it disabled, someone shou
ld
recompile it with it enabled, and I don't know about Dave's latest Ultra
store
driver.

Q : I get SCSI timeouts.

A : If you are using the latest interrupt driven seagate driver,
you will have to jumper the card for IRQ5. The factory settings are
for MSLOSS, and have interrupts disabled. Interrupts are controlled
by the W3 (ST01) or JP3 (ST02) jumper, which should have pins FG
shorted. See your manual.

Q : What future developments are planned?

A : The following changes will occur (listed from highest to lowest priority) :

- Support for extended partitions will be added
- Support for LUN != 0 will be added
- Support for tapes will be added

And, best of all there will be a new interface to the SCSI routines that
will facilitate
scatter/gather, allow multiple outstanding commands (1 per LUN), and all
sorts of
other goodies.



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

Subject: Re: using Linux 0.96a and MCC interim 0.95c+
Message-ID: <[email protected]>
Date: 2 Jun 92 05:37:53 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 17

in article <[email protected]>, [email protected]
rd.EDU (Per M. Bothner) says:

> fdisk-0.92 (I believe included with the MCC-interim release) supports
> extended partitions. I have 6 partitions on my 200MB disk:
> 10 MB DOS, 10MB swap, 20MB root, and 3 user partitions
> (because of the 64MB partition limit). It works great.

I know that there are versions of fdisk that support extended partitions,
but I never saw an announcement that the kernel itself was reliable on
them. Also, the latest thing I saw from Drew was that they were not
supported for scsi. Has anything changed? What device minors do you use?
Thanx.
--
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] (Dan Miner)

Subject: Upgrading from GCC 1.40 to GCC 2.11c, problems
Message-ID: <[email protected]>
Date: 2 Jun 92 05:40:33 GMT
Sender: [email protected]
Organization: University of Missouri - Rolla
Lines: 32


I just finally got around to upgrading my system.
From 0.95a to 0.96a, and GCC 1.40 (0.95a) to GCC 2.11c (0.96a).

I'm having some strange problems...

1) When first install, it would not run at all.
I tried the simple old hello.c and ld complainted about
something not being there. I check everything I could think it
might be looking for. Nothing "wrong".

Well, after a while I stumbled across the 0.96ainc.tar.Z.
I install it over the includes in /usr/include. (More history for
step 1. I had gotten 0.96inc.tar.Z and did what release 2.11a said.
I felt that the headers in 0.96a really should be in there, not 0.96.
So, I copied the headers on top of the 0.96. But that didn't fix anything)

Back to reality....
This fixed many things. One is that hello.c would compile. ๐Ÿ™‚
But other things would not. Common errors are:
No such file or directory (coming from ld.. Like before)
Parse error in stdio.h (I was compiling reboot.c I got from here)

I looked at stdio.h and I can't see anything wrong..?? What is going on?

Enough moaning.. ๐Ÿ™‚
Dan
--
-------------------------------------------------------------------------
Dan Miner email:
[email protected] (prefered)
Computer Science [email protected]


[next article]
From: [email protected] (Dan Miner)

Subject: Flushing problem with Linux?
Message-ID: <[email protected]>
Date: 2 Jun 92 05:46:58 GMT
References: <[email protected]>
Sender: [email protected]
Organization: University of Missouri - Rolla
Lines: 23


Got a weird one for ya.
You start something and do a contol-c to stop it. What happens?
Sometimes nothing and all is right in the universe. Or...
I guess the moon is where it shouldn't be, but sometimes when I
start typing it won't show up on the screen til I hit return.
Why?

To fix this is usually logout and login again.. If
really tough, try to kill the base shell from another console.
(I don't have a working ps yet.. I need the system file and I
don't dare compile the kernal until I get the bugs out...)
I have not applied the patches level 1 I find after the fact..

Ideas? Linus?

Thanks
Dan
--
-------------------------------------------------------------------------
Dan Miner email:
[email protected] (prefered)
Computer Science [email protected]


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

Subject: Re: SERIOUS SECURITY PROBLEM (I think)
Message-ID: <[email protected]>
Date: 2 Jun 92 05:51:02 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 19

in article <[email protected]>, [email protected] says:

> I was just logged in as a regular user, and there was a file in my home
> directory that belongs to root, and I did an rm to it

> rm: remove 'test.c', overriding mode 0644?
> to which I answered yes, and when i did an ls, the file was gone!!

This is standard unix behaviour, so long as you have write perms on the
directory itself. I would be surprised that 0.95c+ didn't do it also.

What IS a security problem is you can chown a file w/o clearing the suid
bit. Fortunately, at least, chown doesn't let an ordinary user change
ownership to root!
--
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: FAQ about X386
Message-ID: <[email protected]>
Date: 2 Jun 92 06:00:35 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 11

in article <[email protected]>, [email protected] (Bill He
nning) says:

> Is something wrong with banjo.concert.net? I can't seem to connect to
> it these days... Is it not at 192.101.26.5?

My ypmatch gives 192.101.21.6.
--
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] (Linus Benedict Torvalds)

Subject: Re: Is it possible to disable ctrl-alt-del while running linux?
Message-ID: <[email protected]>
Date: 2 Jun 92 05:51:43 GMT
References: <[email protected]
md.edu>
Distribution: comp
Organization: University of Helsinki
Lines: 52

In article <[email protected]> [email protected] (R Michael McMaho
n) writes:
>Yes. From kernel/sys.c:
>
>/*
> * this indicates wether you can reboot with ctrl-alt-del: the deault is yes
> */
>static int C_A_D = 1;
>
>Mike [email protected]

Indeed. Here is a snippet of code that should enable/disable ctrl-alt-del:

----------
#define __LIBRARY__
#include
#include

static _syscall3(int,reboot,int,magic,int,magigtoo,int,flag)

int main(int argc,char ** argv)
{
if (argc == 2)
if (!strcmp("on",argv[1])) {
reboot(0xfee1dead,672274793,1);
return 0;
} else if (!strcmp("off",argv[1])) {
reboot(0xfee1dead,672274793,0);
return 0;
} else if (!strcmp("now",argv[1])) {
for (argc=0 ; argc < 5 ; argc++) {
sync();
sleep(5);
}
reboot(0xfee1dead,672274793,0x01234567);
return 0;
}
fprintf(stderr,"usage: reboot [on|off|now]\n");
return 1;
}
----------

The program can only be run as root, or nothing will happen. Usage is:

reboot on - accept ctrl-alt-del (default)
reboot off - ctrl-alt-del is ignored
reboot now - reboot after syncing and waiting 25 seconds (so that you can ^C it)

And as with most code I write, none of the above has ever been tested: I
only guarantee that this is how I intended the reboot system call to
work.

Linus


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

Subject: Re: SERIOUS SECURITY PROBLEM (I think)
Message-ID: <[email protected]>
Date: 2 Jun 92 06:09:47 GMT
References: <[email protected]>
Organization: University of Helsinki
Lines: 17

In article <[email protected]> [email protected] writes:
>
>Standard Unix semantics: If you have write access to the containing
>directory, you can delete any file in the directory. BSD systems have
>an enhancement where if the sticky bit is set on a directory, only the
>owner of the file can delete it. This is why BSD-derived systems have
>/tmp set to mode 1777.
>
>Adding this functionality to Linux would not be difficult. A few lines
>of code to fs/namei.c.

It's already in there: I think 0.96 and earlier didn't handle the
rename() system call correctly (ie you could move a file even if it
wasn't yours), but 0.96a should have that fixed too. Other than that,
sticky directories have worked since 0.12, I believe.

Linus


[next article]
From: [email protected] (Anucha Pitak)

Subject: Re: "X won't work"-type newbie question
Message-ID: <[email protected]>
Organization: Intergraph Corporation, Huntsville, AL.
References: <[email protected]> <[email protected]
iwa.uwa.edu.au>
Date: Tue, 2 Jun 1992 14:42:38 GMT
Lines: 27

In article <[email protected]>, [email protected]
u (Michael O'Reilly) writes:
> [email protected] (Daniel Jimenez) writes:
> :
> : With the 'bootimage' boot disk provided with the X stuff, I try
> : startx and the server dies with "No screens found."
>
> I presume you have made /dev/tty0. If you have then, your video card is
> very likely not supported. I THINK the message 'No screens found' is
> printed by X if it doesn't recognize your chipset.. At least this is the
> message it gives when I tried it by accident on a Trident card.
>

I also have the server error message "no screens found".
My chipset is WD90C00 which is in the x386 FAQ.

From x386 FAQ.
4. What VGA boards are supported?

ET3000 (for ex. GENOA 5300/5400)
ET4000 (Tricom, STB PWR Graph, Sigma Legend, etc.)
GVGA (Genoa 6400)
PVGA1A (Paradise VGA Professional)
WD90C00 (Paradise VGA 1024)

Does x386 under linux supoort WD90C00 ?

Thanks.


[next article]
From: [email protected]@hq.af.mil (Rick Weldon)

Subject: 386BSD
Message-ID: <[email protected]>
Date: 2 Jun 92 15:16:22 GMT
Sender: [email protected]
Reply-To: [email protected]
Organization: 7TH Communications Group
Lines: 12

I seem to remember seeing a posting saying to keep a look
out for 386BSD updates in this group. Am I totally
lost? Anyway-- Does anyone know the latest on the
Bill Jolitz 386BSD release and what group I need to
follow for the latest? That is if there is such a group.

Maybe there haven't been any postings? 8-(

--
Rick Weldon I-NET Inc. (Pentagon, 7TH Com Group)
E-mail: [email protected]
Phone: 703-697-4654


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

Subject: console.el
Summary: put this in /usr/local/emacs/lisp/term/console.el
Keywords: emacs
Message-ID:
Date: 2 Jun 92 15:35:15 GMT
Sender: [email protected] (Net Noise owner)
Reply-To: [email protected] (Ben Cox)
Organization: Ancient Illuminated Bavarian Sears
Lines: 72
Originator: [email protected]

Here is an emacs-lisp file that defines some of the keys on the
keyboard to do reasonable things. I got the first part of this file
from the FAQ, and added some of my own definitions, too.

If you put this in /usr/local/emacs/lisp/term as $TERM.el, where $TERM
is your console type (mine is con80x50.el), it will be automatically
loaded whenever emacs is started.

Here are the definitions:

cursor keys = standard, insert = toggle overwrite mode, delete = ^D,
home/end = standard, num_5 on keypad when not in numlock =
set-mark-command, F1 = help, F2/3/4 = cut/copy/paste, F5 =
auto-fill-mode, F6 = toggle-read-only, F7 = not-modified, F8 =
kill-buffer, F9 = switch-to-buffer, F10 = find-file, F11 =
save-buffer, F12 = save-buffers-kill-emacs

I hope you like these. There is nothing magical about my function key
bindings (though F1-4 are pretty standard); feel free to change them.

--
Ben Cox
[email protected]
--------------------------------------------------console.el
;;; handle the arrow keys
(global-unset-key "\e[")
(setq esc-c-map(make-keymap))
(fset 'esc-c-prefix esc-c-map)
(define-key global-map "\e[" 'esc-c-prefix)
(define-key global-map "\e[B" 'next-line)
(define-key global-map "\e[A" 'previous-line)
(define-key global-map "\e[C" 'forward-char)
(define-key global-map "\e[D" 'backward-char)
(define-key global-map "\e[H" 'beginning-of-line)
(define-key global-map "\e[Y" 'end-of-line)
(define-key global-map "\e[G" 'set-mark-command)

(setq fn-key-map(make-keymap))
(fset 'fn-key-prefix fn-key-map)
(define-key global-map "\e[[" 'fn-key-prefix)
(define-key global-map "\e[[A" 'help-for-help)
(define-key global-map "\e[[B" 'kill-region)
(define-key global-map "\e[[C" 'copy-region-as-kill)
(define-key global-map "\e[[D" 'yank)
(define-key global-map "\e[[E" 'auto-fill-mode)
(define-key global-map "\e[[F" 'toggle-read-only)
(define-key global-map "\e[[G" 'not-modified)
(define-key global-map "\e[[H" 'kill-buffer)
(define-key global-map "\e[[I" 'switch-to-buffer)
(define-key global-map "\e[[J" 'find-file)
(define-key global-map "\e[[K" 'save-buffer)
(define-key global-map "\e[[L" 'save-buffers-kill-emacs)

(setq pgup-map(make-sparse-keymap))
(fset 'pgup-prefix pgup-map)
(define-key global-map "\e[5" 'pgup-prefix)
(define-key global-map "\e[5~" 'scroll-down)

(setq pgdn-map(make-sparse-keymap))
(fset 'pgdn-prefix pgdn-map)
(define-key global-map "\e[6" 'pgdn-prefix)
(define-key global-map "\e[6~" 'scroll-up)

(setq insrt-map(make-sparse-keymap))
(fset 'insrt-prefix insrt-map)
(define-key global-map "\e[2" 'insrt-prefix)
(define-key global-map "\e[2~" 'overwrite-mode)

(setq delt-map(make-sparse-keymap))
(fset 'delt-prefix delt-map)
(define-key global-map "\e[3" 'delt-prefix)
(define-key global-map "\e[3~" 'delete-char)


[next article]
From: [email protected] (Rogier Wolff)

Subject: Weird serial line behaviour
Message-ID: <[email protected]>
Date: 2 Jun 92 07:49:13 GMT
Sender: [email protected] (UseNet News System)
Organization: Delft University of Technology, Dept. of Electrical Engineering
Lines: 13
Nntp-Posting-Host: zen.et.tudelft.nl

While trying to get Linux to run X I connected my old computer as a terminal
to my Linux box. However the sessions that run on this terminal have some
very weird buffering going on: I only get to execute commands when I press
return twice. Also bash doesn't get hold of the characters that I type
until I press return for the first time.

Anybody know if this is a bug or did I forget to do a relevant stty command?

Roger
--
If the opposite of "pro" is "con", what is the opposite of "progress"?
(stolen from [email protected], who got it from [email protected])
EMail: [email protected] ** Tel +31-15-783644 or +31-15-142371


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

Subject: Re: SERIOUS SECURITY PROBLEM (I think)
Message-ID: <[email protected]>
Date: 2 Jun 92 07:28:15 GMT
References: <[email protected]> <[email protected]
.oxy.edu>
Sender: [email protected] (USENET News System)
Organization: Brandeis University
Lines: 42

In article <[email protected]> [email protected] (Da
vid Giller) writes:

[stuff from jgifford deleted :-)]

>
>If there is a file in a directory owned by you, you have rights to it, no
>matter who owns it. I forget just to what extent the permissions are, but
>you have at least the right to delete it.

As far as I know, when it comes to directories, the permissions are as
follows:
r - you may `read' the directory, meaning, you can do an `ls' on it and see
the files in it.
w - you can write to the directory, meaning you can create or delete files
in it. (which is what happened in this case)
x - you can `pass thru' the directory, meaning you can `cd' to it or anything
below it (providing the permissions on the object that follow allow to be
`cd'ed to.

So, in this case, though you don't own the file you own the directory where it
exists. Hence, you may decide whether the file may exist in the directory or
not. Notice, btw, that `x' is the most important bit of the three as you must
atleast have it turned on for you to BE in the directory -- after you are there
you can read (ls) or write (create/delete).

>Notice that you couldn't delete something in the /bin directory, for example.

Write, you don't own it and the write permissions for you is turned off so
you can't delete it.

>
>-Dave
>

Hope this helps, (and is right ๐Ÿ™‚
Oren.

--

Oren Kastner E-mail address:

Brandeis University [email protected]


[next article]
From: [email protected] (Charles Hedrick)

Subject: success with MGR on EGA
Message-ID:
Date: 2 Jun 92 09:09:31 GMT
Organization: Rutgers Univ., New Brunswick, N.J.
Lines: 42

I have a 386sx with an EGA screen and no (usable) mouse. I thought
I'd see if could get MGR to work. Turns out it works fine on an EGA.
However you have to specify 640x350. I.e. run it as

mgr -S 640x350

It won't restore the registers when it exists, so you might as well
reboot when you're done. (Apparently it reads the registered at
startup and puts back the old values on exit. This works fine for a
VGA, but most of the EGA registers aren't readable.)

I had been getting really odd behavior with the pty's. I found out
why. Somehow I had gotten the definitions of the pty's and ttyp's
reversed. This caused most of the trouble. However I do still have
the problem of needing to do something odd to get bash to run in the
window. I followed somebody's suggestion and run getty on pty0 and
pty1. I still have to fix up a few things, e.g.

stty onlcr

to get output to start at the beginning of a line. But basically it
works fine. progman.mgr has an example termcap entry, which you'll
want to add to /etc/termcap, in order to allow your editors, etc., to
run.

Unfortunately I'm afraid I have to say that at the moment I prefer
using the terminal emulator built into the kernel. It's a lot faster,
and I can use color. With an EGA, really there aren't enough bits to
want more than one window on the screen. It would be sort of nice to
have a clock and a few other things, but not nice enough to give up
having a color mode line in emacs, color in my prompts, etc. Also, I
like black letters on white. While MGR has the ability to do inverse
video, it's nearly unreadable. I think it's using intense white,
whereas what you really want is normal intensity. (Unfortunately, it
may turn out that color is too slow to be useful on a 386sx.
Scrolling is already too slow even with monochrome.)

I'll be interested to follow future developments with MGR. Once
colors are working and some more of the tools are working, I'll give
it another try. It does look like it might be more appropriate than X
for small systems. In fact I've even wondered if it might make sense
to try and put it in the kernel as the normal terminal emulator.


[next article]
From: [email protected] (Dan Miner)

Subject: Floating point problem (Where did it go? ๐Ÿ™‚
Message-ID: <[email protected]>
Date: 2 Jun 92 09:23:44 GMT
Sender: [email protected]
Organization: University of Missouri - Rolla
Lines: 36


Here is a VERY strange problem. I am playing with
vgalib and expanding it since X requires several megs (memory
and disk, I have 2 Meg RAM and 40 Meg partition.. w/o mouse).
So X is out. Vgalib gives us poor people (or students) a
very simple graphical access.

Anyone who has done any interactive graphics knows that
you can't live without floating point. Well, I am making a font
"system". Since I can't find code to help me with decoding the
vast vector fonts out there, I am making one myself. I need
a couple of scaling factors. Floating point.

Here is a blurp from my code...

double scalar;
[...]
scalar = 480 / 640;

You would expect a value between 0 and 1. You
print it.. What do I get? 0.00000
What is going on here?

I have had alot of little problems with 2.11c. Someone
thinks that tar is the fault. I think I had replaced for untarring
the files. Has anyone had similar problem?

(I have no 387...)

Thanks.
Dan
--
-------------------------------------------------------------------------
Dan Miner email:
[email protected] (prefered)
Computer Science [email protected]


[next article]
From: Mike Vore W3CCV

Subject: your problems
Message-ID: <[email protected]>
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Date: Tue, 2 Jun 1992 15:39:32 GMT
Lines: 49


|From: Brad Taplin
|Subject: Boot problem? Also, minimum disk requirements?
|Date: Tue, 2 Jun 1992 05:50:37 -0500
|
|Hello. I downloaded these files: install.doc
|and then unzipped them on our rawrite.doc
|Gateway 386sx16 SVGA drive d: rawrite.exe
|and then used rawrite to make bootimag.zip
|a 360K bootable disk from the rootimag.zip
|bootimage. No complaints yet,

But did you read the install.doc file - You need HD disks for both
the root and boot images!! They must be the same density!!

|so I rebooted with the new disk, and watched as the system echo'd
|"Loading........................................................"
|The system then did a CR/LF and stopped. Period. KB buffer then
|fills with keystrokes and beeps to get me to stop trying. Some
|thoughts are that maybe bootimag.zip isn't what I need, or maybe
|rawrite doesn't like creating a 360k boot disk on a 1.2 mb drive,

I hate to say it but RTFM


|
|NEXT, I'd like to program in a rather stripped-down Linux but am
|concerned that maybe the file system, plus basic necessities, gcc
|and some kind of debugger, and either vi or an anemic emacs, may
|not fit my ten meg available drive partition. My fiancee and I
|must use the other 30 megs of this disk for DOS stuff. IFF I do
|fall in love with Linux there's a chance I'll eventually spring
|for a larger drive dedicated to *x?x, but that's a long way off.
|Any suggestions on how I might make best use of 10 mb and where
|I might find the best compact software? I assume it's at tsx-11.
|--

I have 0.95 on about 7 megs - a lot of 'stuff' that needs cleaning out.
So a bare minimum will probably fit, enough to convince your feancee to
let you spring for a good size disk - or even a bigger and better
machine.. Good luck!!


/*---------------------------------+--------------------------------*/
Michael Vore W3CCV | [email protected]
System Administrator | [email protected]
Federal Information Exchange | voice (301)975-0103
/*---------------------------------+--------------------------------*/



[next article]
From: Brad Taplin

Subject: Boot problem? Also, minimum disk requirements?
Message-ID: <[email protected]>
Summary: 11 meg available and probably no more. Workable?
Sender: [email protected]
Organization: Indiana University
Date: Tue, 2 Jun 1992 05:50:37 -0500
Lines: 32

Hello. I downloaded these files: install.doc
and then unzipped them on our rawrite.doc
Gateway 386sx16 SVGA drive d: rawrite.exe
and then used rawrite to make bootimag.zip
a 360K bootable disk from the rootimag.zip
bootimage. No complaints yet,
so I rebooted with the new disk, and watched as the system echo'd
"Loading........................................................"
The system then did a CR/LF and stopped. Period. KB buffer then
fills with keystrokes and beeps to get me to stop trying. Some
thoughts are that maybe bootimag.zip isn't what I need, or maybe
rawrite doesn't like creating a 360k boot disk on a 1.2 mb drive,
or maybe my system just doesn't like Linux (4mb RAM from Gateway),
or maybe I just screwed something up. ALSO, when I pkunzipped the
two files on the Epson XT I use to download stuff, all was well,
but when I pkunzipped these on the 386sx, bootimag unzipped fine
but rootimage "failed the CRC check" so I redid everything with
the same result. I use DOS 5.0 backup/restore to move files that
won't fit the Epson's 360k disks. That's likely not the problem.

NEXT, I'd like to program in a rather stripped-down Linux but am
concerned that maybe the file system, plus basic necessities, gcc
and some kind of debugger, and either vi or an anemic emacs, may
not fit my ten meg available drive partition. My fiancee and I
must use the other 30 megs of this disk for DOS stuff. IFF I do
fall in love with Linux there's a chance I'll eventually spring
for a larger drive dedicated to *x?x, but that's a long way off.
Any suggestions on how I might make best use of 10 mb and where
I might find the best compact software? I assume it's at tsx-11.
--
[email protected] or simply [email protected] will
appreciate your intelligent suggestions, anecdotes, comments & offers.


[next article]
From: [email protected] (Dan Miner)

Subject: Man pages, programmers style ?
Message-ID: <[email protected]>
Date: 2 Jun 92 11:23:44 GMT
Sender: [email protected]
Organization: University of Missouri - Rolla
Lines: 15


I know that someone out there is/has doing man pages
for utils, etc. What about man pages for the programmer?
Pages for portions of the kernal and functions. I'd like to know
about the filesystem or the newer networking stuff. I don't know
where to look for this info. Is anyone doing this?
If there is enough interest I might be persuaded into the job..
????

Dan
--
-------------------------------------------------------------------------
Dan Miner email:
[email protected] (prefered)
Computer Science [email protected]


[next article]
From: [email protected] (Brendan B. Boerner)

Subject: Re: using Linux 0.96a and MCC interim 0.95c+
Message-ID: <[email protected]>
Date: 2 Jun 92 13:53:21 GMT
References: <[email protected]> <[email protected]>
Sender: [email protected]
Organization: Novell, Inc. --Austin
Lines: 29
Nntp-Posting-Host: ca

In article <[email protected]> [email protected] (Jim Burns) writes:
>in article <[email protected]>, [email protected] (Brendan
B. Boerner) says:
>
>> Should I first boot and install the interim 0.95c+ and then install
>> 0.96a in order to fix this problem?
>
>That's one way. The main problem is you need to copy /mnt/lib/lib92.04.06
>to your /lib, and chmod it world executable (a+x). Then the mcc stuff will
>work. You can then edit the install scripts on the disks to reflect that

Thanks for this info. Since I hadn't gotten very far this weekend,
I just started by installing the 0.95c+ MCC-Interim stuff, and then
0.96a. Works like a charm.

For those who missed my original article, I had booted 0.96a and was
trying to run stuff from the MCC-Interim disks and getting "Memory
Fault" messages. I hadn't installed the MCC Boot disk so I didn't
have the lib92.04.06 library.

Another thing - when I load Kermit5A, and type in command, say,
"set baud 2400", only the first s shows up until I hit return. Has
anyone else seen this?

Thanks,
Brendan
--
Brendan B. Boerner Phone: 512/346-8380
Internet: [email protected] MHS: [email protected]
Please use ^^^^^^^^^^^^^^^^^ if replying by mail.


[next article]
From: [email protected] (Chris Flatters)

Subject: Re: Floating point problem (Where did it go? :
Message-ID: <[email protected]>
Sender: [email protected]
Reply-To: [email protected]
Organization: NRAO
References: <[email protected]>
Date: Tue, 2 Jun 1992 15:44:03 GMT
Lines: 19

In article [email protected], [email protected] (Dan Miner) write
s:
>Here is a blurp from my code...
>
>double scalar;
>[...]
> scalar = 480 / 640;
>
> You would expect a value between 0 and 1. You
>print it.. What do I get? 0.00000
>What is going on here?

480 / 640 is an integer expression since both of its operands are integers.
It yields the result 0, which is then cast to double before assignment to
`scalar'. You should have written

scalar = 480.0 / 640.0

Chris Flatters
[email protected]


[next article]
From: [email protected] (David Mulcihy 283-4279)

Subject: Header Files
Message-ID: <[email protected]>
Sender: [email protected] (USENET News System)
Organization: McDonnell Douglas Space Systems Company
Date: Tue, 2 Jun 1992 12:55:18 GMT
Lines: 6

I tried compiling ps, shadow, and admutils. All of these want header files
which I do not have. One of them wants termio.h. termio.h includes termios.h.
I do not have termios.h. Also, one of them wants Linux specific headers.
Any help appreciated. Maybe I just did not grab enough files.
David Mulcihy
[email protected]


[next article]
From: [email protected]

Subject: Sorry for SERIOUS.....
Message-ID: <[email protected]>
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Date: Tue, 2 Jun 1992 14:00:13 GMT
Lines: 34

As usual, I open my mouth before thinking. ๐Ÿ˜‰
Actually, I sent the message to the wrong alias, I meant it to go to about 5
or 6 people, before sending it to the digest and making a fool out of myself.
๐Ÿ˜‰
I have received many corrections both by digest and e-mail, and I must say
thanks everyone for your patience(I didn't get ONE flame out of anyone) ๐Ÿ™‚
I had temporarily forgotten the ONE thing I learned in my UNIX class in
college before they tought me the rn command and I found that rrn worked
instead. After that, my studies were never the same. ๐Ÿ™

Alright, now for the lesson I learned tonight:
LOGIN AS A GENERAL USER for day-to-day work, root can rm ANYTHING ๐Ÿ™
(including /bin and /usr/bin)
Thanks {insert favorite diety here} daily for that backup diskette you haven't
gotten around to re-formatting yet.
ALWAYS keep an up-to-date backup of important stuff(I have a 20 meg partition
with .tar.Z files I am saving, haven't put on floppy yet)
If you have a buggy program(like tar and compress)and replace them with good
ones, BACK THEM UP!!
I have wiped my tar, compress, and bash(using ash, yech)(I restored everything
else)
What happened was this:
I was compiling elm, when it barfed and said :/bin/chmod no such file,
So I got the BRILLIANT(??) idea to put EVERYTHING in /bin, and make a symbolic
link to /usr/bin(I already had a few links from bin to usr/bin)
I did a mv * /bin, and poof, things are a little dead.
I managed to mount a floppy and cp it's /bin to /bin and then ln -s * /usr/bin
This weekend, I am going to back everything up to another machine's harddrive,
and re-partition and install everything.
My partitioning scheme gives win3.1 a fit.
Oh, well....
later
Jim Gifford
[email protected]


[next article]
From: [email protected] (Norbert Bladt)
Newsgroups: comp.sys.ibm.pc.hardware,comp.unix.bsd,comp.os.linux,comp.os.os2,asc
om.bsdi,comp.periphs.scsi

Subject: Bustek-742 EISA controller (compatible with Adaptec 1742 ?)
Keywords: OS/2 386BSD UNIX-V.4
Message-ID: <[email protected]>
Date: 2 Jun 92 06:50:48 GMT
Sender: [email protected] (News Administration)
Organization: Ascom Autelca AG, CH-3053 Guemligen, Switzerland
Lines: 28

Hi there,

I saw an interesting offer for an EISA system which includes a 330 MB SCSI disk
and the Bustek EISA-SCSI controller (742). Now, that there is a Adaptec-1742
I would really like to get this one. However, this would add to the cost
of the system. Now, my question is whether these controllers are compatible.
I know that the Bustek and the Adaptec-1740 will work with OS/2 2.0. What I
don't know is will these work with Dell or Interactive UNIX and 386BSD-UNIX
and especially the Adaptec-1742 with OS/2 2.0. In other words, is it worth
the trouble and cost of changing the Bustek-742 to an Adaptec-1742 ?

Please, post or e-mail your experience. Because others may be interested in
this, perhaps posting is not a bad idea. But, if this question was asked
several times in the last weeks (which is entirely possible), just mail me
the last summary, etc.

Thanks,

Norbert.

--
Norbert Bladt, Ascom Autelca AG, Worbstr. 201, CH-3073 Guemligen, Switzerland
Phone: +41 31 999 65 52 FAX: +41 31 999 65 44
Mail: [email protected] UUCP: ..!uunet!mcsun!chsun!hslrswi!aut!nbladt
--
Norbert Bladt, Ascom Autelca AG, Worbstr. 201, CH-3073 Guemligen, Switzerland
Phone: +41 31 52 95 52 FAX: +41 31 52 95 44
Mail: [email protected] UUCP: ..!uunet!mcsun!chsun!hslrswi!aut!nbladt


[next article]
From: [email protected] (Brendan B. Boerner)

Subject: Re: Can't mount rawritten disk
Keywords: rawrite, mount, error 16
Message-ID: <[email protected]>
Date: 2 Jun 92 14:03:08 GMT
References: <[email protected]>
Sender: [email protected]
Organization: Novell, Inc. --Austin
Lines: 20
Nntp-Posting-Host: ca

In article <[email protected]> [email protected]
du.au (Mike McBain) writes:
>I am attempting to transfer some linux files from DOS to floppy.
>I use rawrite to write the file (turkey.tar). When I then try
>to mount the floppy using the command mount /dev/at0 /mnt I get
>the following error message
>
> magic match failed
> mount: error 16

The floppy contains a tar image, not a filesystem image. Use

tar tvf /dev/at0 to read the archive and

tar xvf /dev/at0 to expand it onto you disk.

Brendan
--
Brendan B. Boerner Phone: 512/346-8380
Internet: [email protected] MHS: [email protected]
Please use ^^^^^^^^^^^^^^^^^ if replying by mail.


[next article]
From: [email protected] (Brendan B. Boerner)

Subject: Re: I can't boot Linux: why?
Message-ID: <[email protected]>
Date: 2 Jun 92 14:13:45 GMT
References: <[email protected]>
Sender: [email protected]
Organization: Novell, Inc. --Austin
Lines: 21
Nntp-Posting-Host: ca

In article <[email protected]> [email protected] (Du
perval Laurent) writes:
>That's when i decided to get the mcc- version (what's mcc stand for anyway?).

Manchester Computing Center (or perhaps Centre :-))

>P.P.S: I just remembered, when I rawrite rootimag or root-us:
> 1) It formats to 7 sectors instead of 13 (or 14). Is that normal?
> 2) I get an address mark not found error. What's that?

The last time I got an address mark not found error, it was because the
image was too big to fit on the disk (because I had forgotten to
download the file in binary form so it expanded). I doubt if you could
boot off of such an image so it is probably because you are writing a
hopefully good image to too small of a disk. Are you sure you
formatted this disk as either a 1.2MB 5.25" or a 1.4MB 3.5"?

Brendan
--
Brendan B. Boerner Phone: 512/346-8380
Internet: [email protected] MHS: [email protected]
Please use ^^^^^^^^^^^^^^^^^ if replying by mail.


[next article]
From: muenz[email protected] (Stefan Muenzel)

Subject: Re: TAR problems / fix && GCC parse problems.
Message-ID:
Sender: [email protected] (News Operator)
Organization: TAT (Theoretical Astrophysics Tuebingen), U. of Tuebingen, FRG
References: <[email protected]>
Date: Tue, 2 Jun 1992 09:25:53 GMT
Lines: 37

[email protected] (Mike Me) writes:
>1st I was having some serious problem untar-ing some files. I was able
>to "list" the files, but the untarring gave a memmory fault.

>I uploaded the files back to our unix system, untarred them (fine),
>and re-tarred them using the "N" option. This option tars according to
>the new POSIX option. I downloaded the files ,and it untarred fine.
>The file BTW was 0.96inc.tar.Z from headrest.woz.colorado.edu.

>I believe that people are using several tar versions, and that the
>linux tar can not deal with it. (unlike the tar on our HP).

>Maybee this can give some people a clue as to how to adapt the linux
>tar version.

>=====================================================================

>Next:

> When I try compiling something with gcc211c I get a parse
>error "(". The error happens in STDIO.h!!!!!! WHY?!?!?!

>I then tried to remove this from the c file, and the same error
>ocurred in the next include file.....

>I need help here guys,

> Thanx,
> MikeE

>PS I have 16 MB scsi drive, and no swap space.

He's not the only one with that sort of problems, so please HELP.

PS: My system: 386/40, no 387
4MB ram 8MB swap
60 MB partition (at-bus)


[next article]
From: [email protected] (Alex C. Liu)

Subject: Patch that fixes notorious MGR bug
Date: 2 Jun 1992 10:02:39 -0700
Organization: None to Speak of
Lines: 38
Sender: [email protected] (Alex Liu)
Distribution: world
Message-ID:
NNTP-Posting-Host: aludra.usc.edu
Keywords: mgr patch

Hi,
I would like to apologize to all those MGR users that were
bothered by the MGR bug that would make the windows act erratically
and wouldn't even let BASH work. Here is a patch to fix that bug.
(It will also enable job control and command line editing on TCSH)

I will be uploading a new version of MGR to banjo this weekend.
This will include a fix for the startrek effect. (THanks
[email protected]) A fix for bash bug. And a utility to turn on
128K mode on TVGA cards. Hopefully color and better mouse support
will come in the future. If you have any patches that you would like
me to include in the next release of MGR please do so as soon as
possible.

(It is embarrasing how simple was the fix for the bash problem)

*** getshell.c-old Sat May 23 19:15:02 1992
--- getshell.c Tue Jun 2 09:52:33 1992
***************
*** 142,147 ****
--- 142,150 ----

#ifdef SYSV
setpgrp();
+ #ifdef LINUX
+ setsid();
+ #endif
#endif
if ((fd=getatty())<0) {
perror("Slave side of p-tty won't open");



--
_____________________________________________________________________________
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] (Daniel Jimenez)

Subject: Re: "X won't work"-type newbie question
Message-ID: <[email protected]>
Date: 2 Jun 92 14:05:59 GMT
References: <[email protected]> <[email protected]
iwa.uwa.edu.au>
Organization: University of Texas at San Antonio
Lines: 20

In article <[email protected]> [email protected]
(Michael O'Reilly) writes:
>I presume you have made /dev/tty0.

Yes.
> If you have then, your video card is
>very likely not supported. I THINK the message 'No screens found' is
>printed by X if it doesn't recognize your chipset.. At least this is the
>message it gives when I tried it by accident on a Trident card.

I suspected this too, so I just looked on the chip on my VGA card.
It says:
PARADISE'88
PVGA1A-JK
00-02 8905

The man page on X386 seems to imply that this card is supported.
By the way, does anyone know where you can get clock.exe without
having to untar megabytes of additional stuff?
--
Daniel Jimenez


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

Subject: Zsh 2.20
Message-ID:
Date: 1 Jun 92 03:55:04 GMT
Organization: Currently Lacking Organization
Lines: 24
X-NewsSoftware: Amiga GRn V1.29 4/25/92 (Beta) by Mike Schwartz

I was doing some work trying to see if I could get Zsh 2.20 up and running
under Linux.

A few anomolies:

Why does prototype as follows:

getpgid(void);
setpgid(void);

Zsh wants to supply one arg to getpgid, and two to setpgid.


sigblock and sigpause are not found. From the looks at signals.h, they
are not currently implemented. I would assume this means zsh is a no go?


EWOULDBLOCK is not defined in errno.h. I'm assuming this is due to the
sigblock functions.

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


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

Subject: Elm -- STILL giving Segmentation errors
Message-ID:
Date: 1 Jun 92 03:58:18 GMT
Organization: Currently Lacking Organization
Lines: 23
X-NewsSoftware: Amiga GRn V1.29 4/25/92 (Beta) by Mike Schwartz

It's frustrating. ๐Ÿ™

While everything else I compile works fine, Elm refuses to function properly
for me.

It will read mail and do all the neat things it's supposed to do, but is NOT
sending mail.

I've compiled smail2.5, and if I use it locally feeding it a message on stdin,
it generates mail, and elm can read this. However, when I hit the S)end key
in Elm, I get segmentation errors.

I'm at 0.96a, and GCC2.12b. At least with 0.96a Elm reports that it got the
seg violation signal. I recall seeing that being changed in 0.96=>.

I'm at a loss at this point. Others have said they compiled with no problems,
and it is working just great except for that ONE SMALL bug.


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


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

Subject: /bin and /usr/bin sources
Message-ID:
Date: 1 Jun 92 03:59:59 GMT
Organization: Currently Lacking Organization
Lines: 14
X-NewsSoftware: Amiga GRn V1.29 4/25/92 (Beta) by Mike Schwartz

I am trying to piece together source for all the /bin and /usr/bin stuff. That
way I can re-compile with newer versions of GCC and not have to keep all the old
libraries lying around.

I know that the site name was floating around before, but I didn't nail it down.
I'll engrave it this time.

Thanks!


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


[next article]
From: [email protected] (Derek Lieber)

Subject: Re: Floating point problem (Where did it go? ๐Ÿ™‚
Message-ID: <[email protected]>
Date: 2 Jun 92 16:01:39 GMT
References: <[email protected]>
Sender: [email protected] (NNTP News Poster)
Reply-To: [email protected] (Derek Lieber)
Organization: IBM T.J. Watson Research Center
Lines: 14
Disclaimer: This posting represents the poster's views, not necessarily those of
IBM
Nntp-Posting-Host: gumby.watson.ibm.com

In article <[email protected]> [email protected] (Dan Miner) writ
es:
>double scalar;
>[...]
> scalar = 480 / 640;
>
> You would expect a value between 0 and 1. You
>print it.. What do I get? 0.00000
>What is going on here?

You're getting integer division. You want floating point. Try:
scalar = 480.0 / 640.0;
--
Derek Lieber
[email protected]


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

Subject: some cdiffs for keyboard.c
Message-ID:
Date: 2 Jun 92 15:22:11 GMT
Sender: [email protected] (Net Noise owner)
Reply-To: [email protected] (Ben Cox)
Organization: Ancient Illuminated Bavarian Sears
Lines: 175
Originator: [email protected]

I was looking at the keyboard.c code last night, and noticed a small
bug (more of an "oversight" than a "bug"): the ". Del" key on the
numpad produces a "," and not a "." in the num-lock mode on US
keyboards. I changed the appropriate line to make it produce "." on
US keyboards.

I patched this, and added a couple of other things as well:

* a new symbol, USE_ALTGR. If this is defined, the right ALT key on
extended keyboards has the ALTGR function (in other words, it does
what the code does now). If you #undef it, it causes the right
ALT key to behave exactly as the left one (let you switch VC's,
lets you use it as a meta key, lets you enter extended chars on
numpad, etc.

* a new symbol, ALLOW_KBD_RESET. If this is defined, CTRL-ALT-DEL
will reboot. If it's #undef'ed, CTRL-ALT-DEL will not reboot.

* a new symbol, SHOW_SCRLED. If this is defined, then the SCROLL
LOCK key will cause the appropriate LED to light. Since this key
doesn't actually toggle a key state (shows process info instead),
I find this annoying so I #undef this symbol. I that case, the
LED never lights.

* a new symbol, CAPS_TO_CTRL. If this is defined, then the CAPS
LOCK key behaves as another CTRL key instead of a CAPS LOCK key.
I never use CAPS LOCK, and I find its location to be more
comfortable as a CTRL key, so I #define it.

Following are the patches. I hope you like them. I make no warranty
on them whatsoever. ๐Ÿ™‚
--------------------------------------------------
*** keyboard.c.old Mon Jun 1 20:55:10 1992
--- keyboard.c Mon Jun 1 22:42:19 1992
***************
*** 13,18 ****
--- 13,42 ----
#include
#include

+ /*
+ * Define this if you want the right ALT key to behave as ALTGR.
+ * Otherwise, both alt keys act the same.
+ */
+ #undef USE_ALTGR
+
+ /*
+ * Define this if you want the SCROLL LOCK key to toggle its LED.
+ * I find this annoying, so I don't set it.
+ */
+ #undef SHOW_SCRLED
+
+ /*
+ * Define this if you want to allow CTRL-ALT-DEL to reset your system.
+ * If you don't have a "reset" button, you probably want this defined.
+ */
+ #define ALLOW_KBD_RESET
+
+ /*
+ * Define this if you want the CAPS LOCK key to act as another CTRL key.
+ * I never use CAPS LOCK anyway, so I like to define this.
+ */
+ #define CAPS_TO_CTRL
+
#define LSHIFT 0x01
#define RSHIFT 0x02
#define LCTRL 0x04
***************
*** 118,126 ****
--- 142,152 ----

static void alt(int sc)
{
+ #ifdef USE_ALTGR
if (ke0)
kmode|=ALTGR;
else
+ #endif
kmode|=ALT;
}

***************
*** 134,142 ****

static void unalt(int sc)
{
if (ke0)
kmode&=(~ALTGR);
! else {
kmode&=(~ALT);
if (npadch != 0) {
put_queue(npadch);
--- 160,171 ----

static void unalt(int sc)
{
+ #ifdef USE_ALTGR
if (ke0)
kmode&=(~ALTGR);
! else
! #endif
! {
kmode&=(~ALT);
if (npadch != 0) {
put_queue(npadch);
***************
*** 199,206 ****
--- 228,237 ----
show_mem();
else
show_state();
+ #ifdef SHOW_SCRLED
kleds^=SCRLED;
set_leds();
+ #endif
}

static void num(int sc)
***************
*** 836,842 ****
}


! #if defined KBD_FR
static unsigned char num_table[] = "789-456+1230.";
#else
static unsigned char num_table[] = "789-456+1230,";
--- 867,873 ----
}


! #if defined(KBD_FR) || defined(KBD_US)
static unsigned char num_table[] = "789-456+1230.";
#else
static unsigned char num_table[] = "789-456+1230,";
***************
*** 879,885 ****
--- 910,918 ----
return;
sc-=0x47;
if (sc == 12 && (kmode&(LCTRL|RCTRL)) && (kmode&(ALT|ALTGR))) {
+ #ifdef ALLOW_KBD_RESET
ctrl_alt_del();

+ #endif
return;
}
if (ke0 == 1) {
***************
*** 1054,1060 ****
--- 1087,1097 ----
do_self,do_self,do_self,do_self, /* 2C-2F z x c v */
do_self,do_self,do_self,do_self, /* 30-33 b n m , */
do_self,slash,rshift,star, /* 34-37 . - rshift * */
+ #ifdef CAPS_TO_CTRL
+ alt,do_self,ctrl,func, /* 38-3B alt sp caps f1 */
+ #else
alt,do_self,caps,func, /* 38-3B alt sp caps f1 */
+ #endif
func,func,func,func, /* 3C-3F f2 f3 f4 f5 */
func,func,func,func, /* 40-43 f6 f7 f8 f9 */
func,num,scroll,cursor, /* 44-47 f10 num scr home */
***************
*** 1086,1092 ****
--- 1123,1133 ----
none,none,none,none, /* AC-AF br br br br */
none,none,none,none, /* B0-B3 br br br br */
none,none,unrshift,none, /* B4-B7 br br unrshift br */
+ #ifdef CAPS_TO_CTRL
+ unalt,none,unctrl,none, /* B8-BB unalt br uncaps br */
+ #else
unalt,none,uncaps,none, /* B8-BB unalt br uncaps br */
+ #endif
none,none,none,none, /* BC-BF br br br br */
none,none,none,none, /* C0-C3 br br br br */
none,none,none,none, /* C4-C7 br br br br */


[next article]
From: [email protected] (Ajay Shah)

Subject: Linux hangs in kermit
Date: 2 Jun 1992 11:45:34 -0700
Organization: University of Southern California, Los Angeles, CA
Lines: 28
Sender: [email protected] (Ajay Shah)
Message-ID:
NNTP-Posting-Host: almaak.usc.edu

I started off by saying "c" in kermit.
He said "set line first". So I set line /dev/ttys1.

Next, when I tried "c", he said "can't open lock file". So I created
directories /usr/spool/{uucp,locks}.

Next, when I tried "c" he froze.

I was logged in as root on two virtual consoles, and there should have
been a getty on two more. Alt-F* worked; I was able to switch from
one screen to another. ScrollLock and Ctrl-ScrollLock worked. Nothing
else worked -- the kermit was frozen, ^C or ^Z did not affect it,
the other login session was frozen, the two gettys would not let me
login.

I waited for 30 seconds hoping update would do his stuff and did a power
out. fsck on both /dev/hda1 and /dev/hda2 does not complain, so I have
no hard disk damage?

Q1: What does one do to make kermit work right?
Q2: A good Unix should not let this happen. I'm no kernel hacker, but
what information should I pass on to the folks who are?

-ans.

--

Ajay Shah, (213)749-8133, [email protected]


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

Subject: Linux .96a, Adaptec AHA 1542B, and 2 scsi disks.
Message-ID: <[email protected]>
Sender: [email protected] (netnews admin account)
Organization: Nyx, Public Access Unix at U. of Denver Math/CS dept.
Date: Tue, 2 Jun 92 19:22:58 GMT
Lines: 19


I have downloaded the bootimage-0.96a and rootimage-0.95a files. I am
able to mount the filesystem on my scsi disk, but I cannot get it to
boot. I got the scsi 'fdisk' program and am able to mkfs and mount the
file system. I did a mknod /dev/sd1 b 8 17, second scsi drive, 1st
partition. The partition is less than 32 meg, as reported from fdisk.
I used the pboot.exe program to set the boot disk, too mount my hd
parition as root. I think that I have entered the hex word for pboot
wrong, what should it be?

I have also read all FAQ's, and README files.

Scott Mace

--
*********************************************************************
* Scott Mace internet: [email protected] *
* uucp : uunet!nyx!smace *
*********************************************************************


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

Subject: linux-0.96a .. keyboard freezes +
Message-ID: <[email protected]>
Date: 2 Jun 92 14:43:10 PDT
Organization: Security Pacific Automation Co., LA, CA
Lines: 17

Hi

Linux-0.96a, made from 0.96 source & the 0.96->0.96a patch, compiled with
gcc2.11c .. boots O.K. runs for ~10-12 seconds .. then the keyboard freezes.

takers ?

attempting to 'gcc -s -lm awk.a -o awk' .. get a bunch of 'undefined symbol
ref'd from text segment' for _pow, _exp, _floor, _ceil, _log, _sqrt ..

the shared libs are set up as no 387 (since I have none ;( ...)

that one is from a 2.11a shared *.a's tar .. using gcc2.11c .. most others
compile O.K. .. need awk to get rc to compile.
--
----> [email protected]
David L.


[next article]
Newsgroups: comp.sys.ibm.pc.hardware,comp.unix.bsd,comp.os.linux,comp.os.os2,asc
om.bsdi,comp.periphs.scsi
From: [email protected] (Paul A Vixie)

Subject: Re: Bustek-742 EISA controller (compatible with Adaptec 1742 ?)
In-Reply-To: [email protected]'s message of Tue, 2 Jun 1992 06:50:48
GMT
Message-ID:
Sender: [email protected] (News)
Organization: DEC Network Software Lab
References: <[email protected]>
Date: 2 Jun 92 15:14:53
Lines: 10

the adaptec 174x has a 154x emulation mode which seems quite compatible
with the various 154x drivers. if the bustek 74x is as similar to the
adaptec 174x as the bustek 54x is to the adaptec 154x, it will have this
emulation mode and you can just use it. perhaps a call to bustek is in
order?
--
Paul Vixie, DEC Network Systems Lab
Palo Alto, California, USA "Ready, Fire, Aim"
decwrl!vixie
vixie!paul alt.pink.bunny.boom.boom.boom moderator


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

Subject: Re: Floating point problem (Where did it go? ๐Ÿ™‚
References: <[email protected]> <[email protected]>
Message-ID:
Originator: [email protected]
Sender: [email protected] (Net Noise owner)
Reply-To: [email protected] (Ben Cox)
Organization: Ancient Illuminated Bavarian Sears
Date: Tue, 2 Jun 1992 23:07:30 GMT
Lines: 14

[email protected] (Pete Chown) writes:

>scalar = 480.0 / 640.0;

>or else put explicit casts of the integers to doubles:

>scalar = ((double) 480) / ((double) 640);

Actually you only need to cast one of these, "480.0 / 640" would do
the trick.

--
Ben Cox
[email protected]


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

Subject: Gcc , a quick question on libraries?
Message-ID: <[email protected]>
Originator: [email protected]
Sender: [email protected]
Organization: George Washington University
Date: Tue, 2 Jun 1992 22:10:43 GMT
Lines: 12

Hey folks I have a program that tries to link to a libm.a library.

My gcc (2.12b) didn't come with one. (I think), so whats so special
about libm.a??

Is it called something else in gcc?

Thanx,
MikeE





[next article]
From: [email protected] (Jeff Blanchet)

Subject: Help with Logitech Mouse using X v1.0 and Linux 0.96a
Message-ID: <[email protected]>
Date: 2 Jun 92 13:03:22 GMT
Organization: SDRC, Cincinnati
Lines: 47


I just recently got X11 v1.0 up and running on my Linux 0.96a system. I was
even able to compile a game (xbattle with gcc2.11c). When I startup X
everything seems to look normal.

Now on to the problem (you knew there would be one:-). My mouse (Logitech
3-button mouse) is working but is VERY hard to control. When I try and move
the cursor around the screen it is VERY jumpy. Also if I move the mouse over
my xterm window, it appears to be sending button clicks when I am not
pushing any of the buttons. It appears to be selecting and pasting text into
my xterm window when I move the mouse over the window. If I position the
cursor in the window and do not move the mouse, I can use the xterm and no
problems exists.

I have seen some notes about trying to put the following line in the xinitrc
file but it has not worked:

stty -istrip > /dev/stty1

Has anyone else seen this before? I have read through all the mail and FAQ's
I have but cannot find a solution. I have the normal entries for my mouse
in the Xconfig file. Here it is incase you want to see it:

Logitech "/dev/ttys1"
BaudRate 1200
SampleRate 150
# Emulate3Buttons

I get the same effect if the SampleRate line is commeted.

I have seen some reference to a mouseinfo.c program. Can someone send this to
me if you think it would help?

Even with the above problems I have tried to run xbattle with success. It
makes it very difficult however when the cursor is jumping around the screen.

Thanks in advance for any help you can give me.

P.S. I found a problem in either the X11 includes or GCC includes. When
compiling xbattle, the typedef wchar_t was undefined. I found if I used
-DX_WCHAR during the compile, everything worked fine. Any ideas? Once I
have more information I will send it to X11 and GCC.


Jeff Blanchet [email protected]
SDRC
Cincinnati Ohio


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

Subject: ``du'' reports double the real disk usage. Hmmm.
Message-ID: <[email protected]>
Sender: [email protected] (USENET News system)
Nntp-Posting-Host: rac2.wam.umd.edu
Organization: University of Maryland at College Park
Date: Wed, 3 Jun 1992 01:15:40 GMT
Lines: 16

I've noticed that ``du'' (and, for that matter, ls -l) report double
the actual disk usage. Is this a bug or are they reporting 512K units?

For example:

#ls -l
total 50
-rw-r--r-- 1 joel joel 25478 Jun 2 13:38 note.wd
#du .
51 ./
#

Shouldn't the totals be 25 and 26?

-Joel
([email protected])


[next article]
From: [email protected] (David Engel)

Subject: Re: ``du'' reports double the real disk usage. Hmmm.
Message-ID: <[email protected]>
Organization: Optical Data Systems, Inc.
X-Newsreader: Tin 1.1 PL3
References: <[email protected]>
Date: Wed, 3 Jun 1992 02:23:55 GMT
Lines: 13

[email protected] (Joel M. Hoffman) writes:
: I've noticed that ``du'' (and, for that matter, ls -l) report double
: the actual disk usage. Is this a bug or are they reporting 512K units?

No bug. Posix requires du to report 512-byte blocks. You can make it
report 1024-byte (1k) blocks with -k. I have it aliases in my .bashrc
like this: alias du='du -k $*'.

David
--
David Engel Optical Data Systems, Inc.
[email protected] 1101 E. Arapaho Road
(214) 234-6400 Richardson, TX 75081


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

Subject: Dexxa mice
Message-ID: <[email protected]>
Date: 2 Jun 92 18:07:04 GMT
Sender: [email protected] (UNT USENet Adminstrator)
Organization: University of North Texas
Lines: 25

Anyone out there have the programming manuals for Dexxa Mice?

I'm interested in finding out why Dexxa mice aren't Microsoft mice compatible
without a driver. *grrr*

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

I guess it's time for me to add my kudos:

Truly amazing, y'all. Linux is fantastic.

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

Oops...another thought:

Is the Interrupt-driven SCSI driver going to make it into the distribution
release for Linux anytime soon?

Does it improve performance very much?

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

I guess that's it for this message. Keep on Linuxing all!

-Kenc


[next article]
Nntp-Posting-Host: gyda.ifi.uio.no
From: [email protected] (Nicolai Langfeldt)

Subject: Re: 386BSD
Message-ID: <[email protected]>
Organization: Folkeaksjonen for (tungt) eple cider p} Inf. stua
References: <[email protected]>
Date: Tue, 2 Jun 1992 17:44:40 GMT
Lines: 6
Originator: [email protected]

In article <[email protected]> [email protected] writes:
[Question on where to find 386BSD news]

The 386BSD folk hang out in comp.unix.bsd

Nicolai


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

Subject: Re: Linux .96a, Adaptec AHA 1542B, and 2 scsi disks.
Message-ID: <[email protected]>
Sender: [email protected] (The Daily Planet)
Nntp-Posting-Host: ophelia.cs.colorado.edu
Organization: University of Colorado at Boulder
References: <[email protected]>
Date: Wed, 3 Jun 1992 04:59:01 GMT
Lines: 18

In article <[email protected]> [email protected] (Sco
tt Mace) writes:
>
>I have downloaded the bootimage-0.96a and rootimage-0.95a files. I am
>able to mount the filesystem on my scsi disk, but I cannot get it to
>boot. I got the scsi 'fdisk' program and am able to mkfs and mount the
>file system. I did a mknod /dev/sd1 b 8 17, second scsi drive, 1st
>partition. The partition is less than 32 meg, as reported from fdisk.

>I used the pboot.exe program to set the boot disk, too mount my hd
>parition as root. I think that I have entered the hex word for pboot
>wrong, what should it be?
>

Are you booting off of floppy or the hard disk? If you boot the
hard disk, you will need to use shoelace and laceup on the system,
and will need to install the kernel on the root partition.

8, 17 is the right number for the device.


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

Subject: Re: using Linux 0.96a and MCC interim 0.95c+
Message-ID: <[email protected]>
Sender: [email protected]
Organization: Computer Science Department, Stanford University.
References: <[email protected]> <[email protected]
EDU>
Date: Wed, 3 Jun 1992 05:37:52 GMT
Lines: 35

In article <[email protected]> [email protected] (Jim Burns) writes:
>I know that there are versions of fdisk that support extended partitions,
>but I never saw an announcement that the kernel itself was reliable on
>them. Also, the latest thing I saw from Drew was that they were not
>supported for scsi. Has anything changed? What device minors do you use?

This is my current set-up. It works pretty well.
This is not a SCSI disk; it is the internal disk drive
of my 486 Notebook.

# fdisk

Command (m for help): p

Disk /dev/hda: 16 heads, 38 sectors, 683 cylinders

Device Boot Begin Start End Blocks Id System
/dev/hda1 1 38 19455 9709 1 DOS 12-bit FAT
/dev/hda2 * 19456 19456 58367 19456 81 Linux/MINIX
/dev/hda3 58368 58368 284543 113088 5 Extended
/dev/hda4 284544 284544 415263 65360 81 Linux/MINIX
/dev/hda5 58370 58370 134367 37999 81 Linux/MINIX
/dev/hda6 134368 134368 153823 9728 82 Linux swap
/dev/hda7 153824 153824 284543 65360 81 Linux/MINIX

Command (m for help): q
# df

inodes inodes inodes blocks blocks blocks mount
device total used free total used free point
--------------------------------------------------------------------
/dev/hda2 6485 337 6148 19456 7869 11587 /
/dev/hda4 21786 2655 19131 65360 43134 22226 /usr
/dev/hda7 21786 1232 20554 65360 22353 43007 /home
/dev/hda5 12666 1958 10708 37999 19506 18493 /misc


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

Subject: Patches to make-3.62
Message-ID: <[email protected]>
Sender: [email protected]
Organization: Computer Science Department, Stanford University.
References: <[email protected]>
Date: Wed, 3 Jun 1992 05:49:10 GMT
Lines: 261


I had some problems with make, so I grabbed make-3.62, and
compiled it. Below are the patches I had to make. Note
that most of the changes are generic changes needed for
Posix signal-handling, and are not Linux specific. (I
hope I got it right ...)

Could someone install this on some suitable archive?

Apply the following patches to GNU Make version 3.62
for something that will work on Linux.

diff -c make-3.62/Makefile make-linux/Makefile
*** make-3.62/Makefile Wed Oct 30 02:04:26 1991
--- make-linux/Makefile Tue Jun 2 22:09:58 1992
***************
*** 19,25 ****
# Makefile for GNU Make
#

! CFLAGS = $(defines) -g
LDFLAGS = -g

# Define these for your system as follows:
--- 19,25 ----
# Makefile for GNU Make
#

! CFLAGS = $(defines) -g -O # -m486
LDFLAGS = -g

# Define these for your system as follows:
***************
*** 38,44 ****
# without complaint but produce losing code,
# so beware.
# NeXT 1.0a uses an old version of GCC, which required -D__inline=inline.
! defines =

# Define these for your system as follows:
# -DUMAX Encore UMAX
--- 38,44 ----
# without complaint but produce losing code,
# so beware.
# NeXT 1.0a uses an old version of GCC, which required -D__inline=inline.
! defines = -DPOSIX

# Define these for your system as follows:
# -DUMAX Encore UMAX
***************
*** 52,58 ****
# Define:
# -DNLIST_NAME_UNION If `struct nlist' has a n_un member.
# -DNLIST_NAME_ARRAY If `n_name' is an array.
! LOAD_AVG =

# If you don't want archive support, comment these out.
ARCHIVES = arscan.o ar.o
--- 52,58 ----
# Define:
# -DNLIST_NAME_UNION If `struct nlist' has a n_un member.
# -DNLIST_NAME_ARRAY If `n_name' is an array.
! LOAD_AVG = -DNO_LDAV

# If you don't want archive support, comment these out.
ARCHIVES = arscan.o ar.o
***************
*** 80,92 ****
# Comment this out if POSIX.2 glob is installed on your system
# (it's in the GNU C Library, so if you're using that, this is
# not needed at all.)
! globdep = glob/libglob.a

# Library containing POSIX.2 `glob' function.
# Comment this line out if it's in the C library (which is the case if you
# are using the GNU C Library), or change it to the appropriate file name
# or -l switch.
! globlib = $(globdep)

# Name under which to install GNU make.
instname = make
--- 80,92 ----
# Comment this out if POSIX.2 glob is installed on your system
# (it's in the GNU C Library, so if you're using that, this is
# not needed at all.)
! # globdep = glob/libglob.a

# Library containing POSIX.2 `glob' function.
# Comment this line out if it's in the C library (which is the case if you
# are using the GNU C Library), or change it to the appropriate file name
# or -l switch.
! # globlib = $(globdep)

# Name under which to install GNU make.
instname = make
diff -c make-3.62/job.c make-linux/job.c
*** make-3.62/job.c Thu Oct 24 21:58:33 1991
--- make-linux/job.c Tue Jun 2 21:26:41 1992
***************
*** 117,122 ****
--- 117,123 ----
extern int setgid (), getgid ();
#endif /* GNU C library. */

+ #ifndef getdtablesize
#ifndef USG
extern int getdtablesize ();
#else
***************
*** 123,128 ****
--- 124,130 ----
#include
#define getdtablesize() NOFILE
#endif
+ #endif

extern void wait_to_start_job ();
extern int start_remote_job_p ();
***************
*** 178,184 ****
--- 180,190 ----
^L
extern void block_remote_children (), unblock_remote_children ();

+ #ifdef POSIX
+ extern sigset_t fatal_signal_mask;
+ #else
extern int fatal_signal_mask;
+ #endif

#ifdef USG
/* Set nonzero in the interval when it's possible that we may see a dead
***************
*** 192,197 ****
--- 198,210 ----
static void
block_signals ()
{
+ #ifdef POSIX
+ sigset_t set, oset;
+ set = fatal_signal_mask;
+ sigaddset(&set, SIGCHLD);
+ /* Block the signals. */
+ (void) sigprocmask (SIG_BLOCK, &set, &oset);
+ #else
#ifdef USG

/* Tell child_handler that it might see children that aren't yet
***************
*** 201,207 ****
/* Ignoring SIGCLD makes wait always return -1.
Using the default action does the right thing. */
(void) SIGNAL (SIGCLD, SIG_DFL);
-
#else /* Not USG. */

/* Block the signals. */
--- 214,219 ----
***************
*** 208,213 ****
--- 220,226 ----
(void) sigblock (fatal_signal_mask | sigmask (SIGCHLD));

#endif
+ #endif

block_remote_children ();
}
***************
*** 216,221 ****
--- 229,241 ----
static void
unblock_signals ()
{
+ #ifdef POSIX
+ sigset_t set, oset;
+ set = fatal_signal_mask;
+ sigaddset(&set, SIGCHLD);
+ /* Unblock the signals. */
+ (void) sigprocmask (SIG_UNBLOCK, &set, &oset);
+ #else /* !POSIX */
#ifdef USG

(void) SIGNAL (SIGCLD, child_handler);
***************
*** 228,233 ****
--- 248,254 ----
/* Unblock the signals. */
(void) sigsetmask (sigblock (0) & ~(fatal_signal_mask | sigmask (SIGCHLD)));

+ #endif
#endif

unblock_remote_children ();
diff -c make-3.62/main.c make-linux/main.c
*** make-3.62/main.c Mon Sep 9 23:36:14 1991
--- make-linux/main.c Tue Jun 2 21:19:07 1992
***************
*** 268,274 ****
--- 268,278 ----
struct file *default_file;
^L
/* Mask of signals that are being caught with fatal_error_signal. */
+ #ifdef POSIX
+ sigset_t fatal_signal_mask;
+ #else
int fatal_signal_mask;
+ #endif

int
main (argc, argv, envp)
***************
*** 299,305 ****
--- 303,316 ----
init_siglist ();
#endif

+ #ifdef POSIX
+ sigemptyset(&fatal_signal_mask);
+ #else
fatal_signal_mask = 0;
+ #ifndef sigaddset
+ #define sigaddset(setp, sig) *(setp) |= sigmask(sig)
+ #endif
+ #endif

#define FATAL_SIG(sig)
\
if (SIGNAL ((sig), (SIGHANDLER) fatal_error_signal) \
***************
*** 306,312 ****
== (SIGHANDLER) SIG_IGN)
\
(void) SIGNAL ((sig), SIG_IGN); \
else
\
! fatal_signal_mask |= sigmask (sig);

FATAL_SIG (SIGHUP);
FATAL_SIG (SIGQUIT);
--- 317,323 ----
== (SIGHANDLER) SIG_IGN)
\
(void) SIGNAL ((sig), SIG_IGN); \
else
\
! sigaddset(&fatal_signal_mask, sig);

FATAL_SIG (SIGHUP);
FATAL_SIG (SIGQUIT);
***************
*** 321,329 ****
--- 332,344 ----
FATAL_SIG (SIGDANGER);
#endif
FATAL_SIG (SIGFPE);
+ #ifdef SIGBUS
FATAL_SIG (SIGBUS);
+ #endif
FATAL_SIG (SIGSEGV);
+ #ifdef SIGSYS
FATAL_SIG (SIGSYS);
+ #endif
FATAL_SIG (SIGTERM);
#ifdef SIGXCPU
FATAL_SIG (SIGXCPU);


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

Subject: Should init create INIT_PROCESS in utmp?
Message-ID: <[email protected]>
Date: 3 Jun 92 01:09:15 GMT
Organization: Chiral Symmetry Breakers
Lines: 11

Just how do most unixes do this? The getty_ps package seems to expect an
entry in utmp before it runs. On the other hand, the standard init (is it
simpleinit.c from poe-IGL?) doesn't touch utmp, and the agetty.c from poe
dumps a utmp entry without checking for an INIT one.

Is it also standard practice for both getty and login to stick an
entry into wtmp? I end up with LOGIN/username pairs.
--
Steve Robbins -- iconoclast at large
"Dyslexics of the world, untie!"
[email protected]


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

Subject: fix for my adduser
Message-ID: <[email protected]>
Sender: [email protected] (USENET News System)
Nntp-Posting-Host: deimos.ucc.umass.edu
Organization: University of Massachusetts at Amherst
Date: Wed, 3 Jun 1992 05:34:15 GMT
X-News-Reader: VMS NEWS 1.20
Lines: 110

it appears that my adduser program didn't survive the trek to shared
libraries, as the gets changes somewhat. i put in some fflushes
that make it work for linux 0.96a and gcc-2.1 (or at least it worked
on my system, i am not sure how up to date my libraries are).

I sent it to tsx, but i am not sure if i did it right, so i am going
to post it here, too

-- Craig Hagan

---cut here--
begin 644 adduser.tar.Z
M'YV08M)[email protected][email protected]"$#[email protected];,FY<7$E#[email protected]+-2KJW,FSIT\ !.F$D7,[email protected])PW;^C\?'@T
MZ=*G4*-*G4JUJLZ PL>'",G#QPZ+MY8'3O58\:-'3V"3#N2K<<8*F5MAHVZ,&;0F-DRID489 ,+7AAT:-&F2JDB'LRXL>/'9 \$2,AA,D0P#B$YK)*H
MW"!V"*Y8F;*H#X-%=2 LZH)A2:(N_ ;M^_> 4#F%D'(M/%8$WP&%B>3-B# H
M&($'MX0MPK0P1A<$,W 14!A,G6I-S9_/T#4]X:(J%(87/YY\>7;HTJE;7X1=
MH7/HW!5^AT[<.')A,V M7 2H.X 815!0'00SR+*?)/[% %XB^HAG7W+Y[=>?
M>P$.6."!"8+W3Q/X#-('/K054L0_),Z72!T()&([email protected](B&3<36!S\AMB7508,45_Z"(SCO^D-A(&>QLX<47R0RC4 .6_09 ],H9$!"##0C)97!*%0
ME;MH224J"G7'@"AA4FF)[email protected]":0+ P!]7NFE'G RTX:6;9-S)P!1ENJE$
[email protected]:5. [email protected]@*"4"E?_XM-,&=#'[email protected]&4,I3"I0S,M1$
MJ MYH= 98XPAPQ=CO-$&'[email protected]#P11UFC/$%&6^DH=2MN7ZQ1AEY""6&
MK;CJ.H<8,2 ;[+* :NL&#,X.^U-TN[ZAAEIN!$&&W 89"=DY)9K[KGHIJON
M8%@19- +[6KEPACK-F:61B2IE6];)8$00PPUW&##12FMU))&-^SEKPPS-/Q7
MO1 K5!A1'"VF&%*)1:SQQAPK]((*(,1K$ @ZA"Q0R""X4<8=(+A+5!BIOE&'
M&W3T*Y(*+RCP,0A#R!%&&F> @$089X3AALU()VTSSCJ#K/334"?-],YP#,7K
[email protected](:[!1!QEGE.&""W?(X2L=91P]D-EN!(WTU""WVD8;[email protected]+ M
M"',(110=5=?A06"
M&4>U 0(<;Y310A)',(%[email protected](+3P!^^,BORR0R_-N[A58C_?+M + C]#M
M&%Z3L34/?9.1QALNH.&# L*[0?S7Q_?-]AG-/Q_]],:#@'P><[[email protected]_]
M\,4?/W,:R6=_OO3I>S\'^"_0D0;=[F\?/P_VXV\^],;CELI ([email protected]&$*5R#"
M%XR0!"84 00B>$$9Z#"&%U1M#G.X QE$ , R"'!K1"B"$*IP!."- 0V&40%7
M=H>"%.P >&WXF1M:J( ]*. B)S2,&9"RA1G(H L[2( [email protected]>1G86A95H!0> "
M H)N?2YT() @[email protected]'!H8-_J;AA%D#F.;YEKHAFPN#67>8MN3M$IP)<8AMHH1+#[email protected]"$0B=I&,HS-()+W%AI35<6M[XR,21U;(0YK,>&388AP;.;HJ
M9I ,E,QC$;VHM59:\0UR2&4H:^DR0AK2<(E,Y2(AQTKER6$+,M!()?5(RZT9
MLPQC<%Q7]#9&43YRD[XT93!5R4C#S$%K;& #,I4IRTN*4F4L<]D)\O;-,H23
MF\0T#!NZ589QUF"9LT39/ <(.E=A4HH5O. K/Q>KK0US9G, FLI2V:V:G>$H
M=8 #"^J0!C*4M:)4< A=-F(:-#G/XPAV&XH9#KO2B(#,#[email protected]!\HY$9?&DT/)F)9/,3#QBP79
M6AI$VD>#_'[email protected]%:U(!LKA4HQ\ YS1PA967(T1A7ECZB5%4LBT$M)GO9!J"AH6:%'%T=:N9$C484!"!1Z\Y<&K)PON$.>;/F41Y:1\(][email protected]\!4(
M[J"UHQVQEVCUI5NW*=BXX9*F-6,#4N"0MYG9KY-'%[email protected]=D#K('M< *^'DVKJA5FSO)YTYSN5 X]!:9A$8M),PB5(&APK"_YMH8TP"%[email protected]=()IW>U:/1))+ I29G2 PUX'V;=8=5)69TB#&&0%3SDF8 [[email protected],R*WD1
M9F(2OFRH&0JJ)H>:[7)KJF5MVE;XE>4=;9C]*\,7:D:'[email protected]"UHTG1FB.'XE^J
MCHYE ^:<_20)3P8Z$'&Y_0*W9)7'[email protected]"A.L8I7S.(6N_C%,%YQSKA(V8*B
M( 0R!6T*[email protected];Z!HZ:-%&,"6F=D$/[email protected]
M94EIXFTG&%!77E&M(JFQK&[LTA0H( $V3,!%X, [email protected][email protected]@ANJX,=/*AM)1.!"X&G9C&BH+DA=[PSN8KRM\$-VD[S]JTEDB$/?$/#89U\9I$8UVO?
M1$'RP.O"*0?UN-\,*Z_VZE4^RN$.[-M:01A,W:$&C6E]=F=!*NT10"/:IS'(
M8Y][BP(ZTFW/?69S0]\[email protected] Y=D+,()OK'/=,ZQ"-^Z0[#-<,"'C"!"[email protected]
[email protected]@ CGH^85]%G,9;FR&,[email protected]@[email protected](0F, $( B!4\G0[DM*>&M)("8T
M!R=*W*9[BWV&\AM!0&M+\WJ?96#[email protected]*;7:#NU_6&P,<\H""';Z!!=9SW!I0
[email protected]'!NZT#/[email protected];"*M^[6P,=9AIVD)+
MS\7YO(TS%OE%YMV#$$3\515'"K,_&7*;@TQE>* #Y<[email protected]:O)8#F1ND$4[>
MTZ(6HAY5I=#];'3 C0'I%N=V"];.]K:[_>U9;\/6Z:GL%)A9B"!(<\"%33-B
MPT%65=Q:X-KP!CL $NMUT#H(N)Y5+F_UE74'=\ AZRL4M" &P%9S'[email protected]
M.5>B+0BZN;IN60'[(BXM^KP?3O ^3Q4%\]ZQI^>:9#5N>8JY!;A(TMPOOKL9
MSE1 8BR(QXCM73;F;8C)P
M?:/F?>[email protected]*Q/?$EO.QUYE)2$KR6YY$8YZ7_NL7]NO2R-4T'49-ASWWB]?7
M)$G^5Q+8YQ%$EVMOM#/@-T$I%5VN1CF$%SA:)EXU$T.0EE(6ZP
M%Q!WQWNCTV;$MGYYXU<<-7^^[email protected]:G][email protected]@;ZQW\M. =C8#0EN$$L8 (HV(*\
M9G%-UX ]@((]\%.T)E=_!E%P,(0P<'>[QV=.=W55111:U%=("%C')TG5)3J2
M]3MCAWYP)F='D :&!P+'QC<191"])&QV, <_ );D =\X 9=0#(01'\L2'T>
[email protected]^9MFE[U8(BT8-(\8-LI5(HD'1OL 4PT 4AT ,GD 4GD +O9HB(J(B,F >/
MZ(>=-W9G=GV]1X)?Z 8FV#(518?+MH(#:&DO&(.<9G\E08,V"&M6&O898C.UHG#]HFA*&EH]$S1A$MY4#);\ ($<1 N$S['-HMR(?]QXI_>(O&E(LE881E)P=G!P*(X1=(S&J!7*. >EV&S6
MF(>IF#S=HHWGYXDB0 0>% 9UD%\@8$R#9&S([email protected]+*UE)I( ?0.(*[&&>@. ?L
M%T#\Z(_M]$[$^ )4YP;[email protected];,6(>F&(_1.(_[MXH%>!&\-I%LD) C"[email protected]
M:)+C:'?EZ(4N&4$7F9'?EI)>F(_[V(\U8Y(!N8P%:9(HJ8M]QXL.F3M94L9Q%K$N(RD:(?/Z)$NB&G3&))]2)4 F&[VZ!'7?FZ(FYQ6W'
M]HY,1Y5IEF H8 (-!FSX)5^):[email protected]@ (-YFZ A0MX!QV68AY"6\[email protected] =FL([email protected] ,^ (T\)<@< )A0)@IN7GU192_QY!RUC-E0#Y1
MZ(9 Z09J67\XB8])H#8'"4W2E >4(Y!=P (%F8U4F9.HEC6F.9!5A#(0R9.1
MYDYL8)NHB7'@=)*YB8]&DTJT23G2A3(!EDO.=YRYE9 Y*6=)\')*]EFPV89O
M&(<;>9/W!Y(RJ)6WZ(/E)!*I1W0Y!IL]((F).(2-> )\P <6-XGT:8DNU"\%
MR(D*690BT 1AT%R& XRQ.9O+6)W+AIM) L ,W -+6$X-)RL]
M,[email protected]+6AD *Y&';)@)A0)Z?XX4;FG B<&PZX**Q^*)D\*)S0*,TJFRT)I4$
[email protected]=AJ4&4,YM^)[email protected]'U%-CL#4ZT&[#M&1W,$/B"%)"[email protected]"<27 =4Z=+
M43LO0*##[email protected]^ ):AO40,Q0# JP1(P4!,W(!,+ QBX3
MM>JJI*CLM "[email protected](HX*GMYJNR:JI3T&XM0#NWDT2UDSL[&A8*P*G0
M&JW2.JW4.A5X*@5%$ 1$T 1%4*T^\:>"NA:!ZA9\40,G43"*BC #XR\Q8!M\##>2A:56C&7>C%.$:_XFJ_YAD)C&%%BBC:P54LH- :#TUU3*EFK)0=K4%H>
M-3+[5 =X %@ND ,V$ 9&10<*D(&HI%S\JFI863/2J#F%)%U;,YI6UUMWF35:
M9([email protected] $YPH #,YTCF]G?.U"L'[email protected]:&)UIP0#M\');LT/#=4B\[email protected] H>UEU1C>[email protected]#">)?!F5^)]3CKB(SL^ (\L((^T&[-&3(*
MX)/AN)-:JUP6V2T9&:0=E4JIY66LE;9F$)&R-5=3RX(_JP"2Q4='PQ6CB39*
M6U)(J:4)>DI**SI?9$UTFTL$-3D*4+8YU#9)JS=KJDE9Q%'_A:56VQ5]VYI8
MI$6?"VG?)#-LD$J%]+3-VCMTB$D[2DC=,A20UD^[email protected],0#[email protected]);*0
MEBPML*,*@*W:RJU]5C*[email protected]!ZZD$%=1'**TI/"KWL]475>THN,[G)VGRO
M,[VU%$DY5Z:M8CS SQ=RA4_$S0H5#1N #SM:S1 8#SWPV0N4 >IM,"MU<+[Y&L "/, $7, &?, (G, *O, ,W, ._, 0',$2/,$47,$6?,$8G,$:
MO,$_,[email protected]',(B/,(D7,(F?,(HG,(JO,(LW,(N_,(P',,R/,,T7,,V?,,X
MG,,ZO,,\W,,^_,- ',1"/,1$7,1&?,1(G,1*O,1,W,1._,10',52/,547,56
(?,58G,5:7!4^

end

----------

Craig Hagan
University Computing Center
UMASS/Amherst


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

Subject: Re: /bin and /usr/bin sources
Message-ID: <[email protected]>
Date: 3 Jun 92 06:54:21 GMT
References:
Organization: Georgia Institute of Technology
Lines: 15

in article , [email protected] (Steven M. Palm
) says:

> I am trying to piece together source for all the /bin and /usr/bin stuff. Tha
t

ftp.mcc.ac.uk:dsk/pub/linux/mcc-interim/0.95c+, the links-to-sources dir
has mostly links to gnu utils, like bash-1.12.tar.Z, binutils-1.9.tar.Z,
fileutils-3.2.tar.Z, shellutils-1.6.tar.Z, & textutils-1.3.tar.Z; the
patches dir has a bunch of *.ud.Z files that are diffs to the main gnu
utils that mcc used for their stuff; and the sources dir has misc.tar.Z,
which has fsck.c, mknod.c, etc.
--
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]

Subject: Starting X386
Message-ID: <[email protected]>
Date: 2 Jun 92 19:49:50 GMT
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Lines: 11


We *finally* got X386 working on our paradise board with a taxan multisync
monitor. After spending all day (12+ hours) staring at video timings, we
happened to boot up into the 80-column text mode, and type 'startx'. It worked.
I can't remember reading of anyone having mentioned this before - using the
132-column modes can screw your graphics mode registers.

This wasn't just an isolated incident, it happens all the time. Booting in
low-res text is the only solution.

Simon.


[next article]
From: [email protected] (the MOST SIGNIFICANT bit)

Subject: Help with reading "dotfile" on login
Message-ID: <[email protected]>
Date: 3 Jun 92 06:17:38 GMT
Sender: [email protected]
Followup-To: comp.os.linux
Organization: University of California, Santa Cruz
Lines: 14


I just installed the shadow password suite (from the binary
distribution). Everything appears to be working fine, with one
small (actaually, quite aggravating) exception: whenever I log in,
no startup files for the shell are sourced (neither
.profile/.bashrc nor .login/.tcshrc). This wasn't happening
before I put in the shadow password stuff.

I have the MCC distribution, updated with the 0.96a kernel.

thanks for any help

(and I'm still trying to figure out how to get linux to see the
ethernet card)


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

Subject: Re: Zsh 2.20
Message-ID: <[email protected]>
Date: 3 Jun 92 07:01:18 GMT
References:
Organization: Georgia Institute of Technology
Lines: 20

in article , [email protected] (Steven M. Palm
) says:

> I was doing some work trying to see if I could get Zsh 2.20 up and running
> under Linux.

Paul says he doesn't support sysV type systems because he has no access to
them. Hence, zsh is very bsd. Bash is just fine for me - the globbing is
better than ksh, but not as good as zsh. 2.2 *is* much better than 2.1, tho'
- got it running on sequent dynix & sunos. Bash does have an annoying habit
of not being able to pipe output from a function. If you have

ll () { ls -alF [email protected]; }

then ll|more doesn't work, but (ll)|more does. Other functions behave
similarly.
--
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: 386BSD
Message-ID: <[email protected]>
Date: 3 Jun 92 07:08:50 GMT
References: <[email protected]>
Organization: Georgia Institute of Technology
Lines: 20

in article <[email protected]>, [email protected]@hq.af.mil (Rick Weldon) says:
>
> I seem to remember seeing a posting saying to keep a look
> out for 386BSD updates in this group. Am I totally
> lost? Anyway-- Does anyone know the latest on the
> Bill Jolitz 386BSD release and what group I need to
> follow for the latest? That is if there is such a group.

You're lost ๐Ÿ™‚ Subscribe to comp.unix.bsd. Be prepared to wade thru
discussions of why is Mach not unix, why isn't a microkernel micro, how to
squeeze preformance out of serial drivers, and some stuff on BSDI's BSD/386,
which like 386BSD also descends from the Berkeley Net/2 source tree. Not a
whole lot on 386BSD right now, since there hasn't been a new release in two
months. The only relationship to linux is that both groups are liberally
'stealing' code from the other to add features & fix bugs.
--
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] (John Graves)

Subject: Next Step after Boot & Root...
Message-ID:
Date: 2 Jun 92 16:34:12 GMT
Distribution: usa
Organization: University of New Mexico, Albuquerque
Lines: 14

I've don the rawrite of Boot & Root... It works fine...

What's the next step? The ROOT doesn't contain a term program nor
the ms dos utils. I need to download the files using ms-dos, but how
can I get them into the linux system? is there a way to mount an
ms-dos disk without having the ms-dos utils?

Pleasee use email if you can help... I can't get to the newsgroups very often..

[email protected]

Thanks in advance...

-John


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

Subject: Re: "X won't work"-type newbie question
In-Reply-To: [email protected]'s message of Tue, 2 Jun 1992 14:42:38 GMT
References: <[email protected]>
<[email protected]>
<[email protected]>
Sender: [email protected] (USENET Newssystem)
Organization: Inst. fuer Informatik, Technische Univ. Muenchen, Germany
Date: Wed, 3 Jun 1992 06:58:36 GMT
Message-ID: <[email protected]>
Lines: 61

In article <[email protected]> [email protected] (An
ucha Pitak) writes:

Newsgroups: comp.os.linux
Path: NewsServ!math.fu-berlin.de!news.netmbx.de!Germany.EU.net!mcsun!uunet!in
fonode!pitaka
From: [email protected] (Anucha Pitak)
Organization: Intergraph Corporation, Huntsville, AL.
References: <[email protected]> <1992Jun2.051104.21322
@uniwa.uwa.edu.au>
Date: Tue, 2 Jun 1992 14:42:38 GMT
Lines: 27

In article <[email protected]>, [email protected]
u.au (Michael O'Reilly) writes:
> [email protected] (Daniel Jimenez) writes:
> :
> : With the 'bootimage' boot disk provided with the X stuff, I try
> : startx and the server dies with "No screens found."
>
> I presume you have made /dev/tty0. If you have then, your video card is
> very likely not supported. I THINK the message 'No screens found' is
> printed by X if it doesn't recognize your chipset.. At least this is the
> message it gives when I tried it by accident on a Trident card.
>

I also have the server error message "no screens found".
My chipset is WD90C00 which is in the x386 FAQ.

From x386 FAQ.
4. What VGA boards are supported?

ET3000 (for ex. GENOA 5300/5400)
ET4000 (Tricom, STB PWR Graph, Sigma Legend, etc.)
GVGA (Genoa 6400)
PVGA1A (Paradise VGA Professional)
WD90C00 (Paradise VGA 1024)

Does x386 under linux supoort WD90C00 ?

Thanks.


Well, the problem with these WD90Cxx boards is quite trivial. X386
just follows WD's recommendations that a SVGA board using a WD chipset
is REQUIRED to have a certain id string in it's BIOS. Some vendors
decided not to do this (well, they tried to be like IBM ...). Thus
X386 will not autodetect the chipset.

But on the other hand, nothing prevents you from overriding, the
autodetect mechanism, and explicitely say:

Chipset "PVGA1" (or so)

Again this is a typical RTFM question, which IMHO just wastes network
bandwidth by beeing asked here ...

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

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


[next article]
From: [email protected] (Alex C. Liu)

Subject: New version of MGR in banjo.concert.net
Date: 3 Jun 1992 00:14:17 -0700
Organization: None to Speak of
Lines: 52
Sender: [email protected] (Alex Liu)
Distribution: world
Message-ID:
NNTP-Posting-Host: aludra.usc.edu
Keywords: Bug fix release v0.03.

Hi,
This is more of a bug fix release of mgr. (-V/-v still
reports v0.02 but this is v0.03, I just forgot to bump the rev number)

Anyways, these are the changes from v0.02 to v0.03

Job control/pty/shell tty stuff fixes. (Bash should work now)
Star trek effect fix ([email protected])
new install script to configure mouse and default display
mode.
Turn on 128K TVGA modes utility.

Basically the most important change is the bugfix for the problem of
not being able to use bash under mgr.

Also, I no longer provide with the statically linked "mgr" executable.
You now need to have a working gcc v2.1 or better.

If you are first time MGR user simply get the file

banjo.concert.net:/pub/Linux/MGR/linuxmgr.tar.Z

It contains everything you need to have in order to run MGR. Be sure
you read the README file, since some of the questions in the install
script may seem cryptic.

If you are upgrading from v0.02 to v0.03 and still have the mgr.a and
the driver files get the file

banjo.concert.net:/pub/Linux/MGR/upgrade/upgrade-0.03.tar.Z

extract and execute the upgrade script in the same directory you have
your mgr.a file. Then execute makemgr as root to configure and
create a new mgr executable.

If you want to update your mgr source code, you need:

banjo.concert.net:/pub/Linux/MGR/upgrade/diff-0.03.Z

PS: Distributing files in .a format is great huh? That way you can
release upgrade files instead of full executables.

Please send all bug reports and fixes to me.
If you want better EGA support (i.e. restore text modes) send me
e-mail, I might write a EGA driver but will need somebody to test it.
If you have a new display driver or mouse driver for mgr, please send
it to me.
--
_____________________________________________________________________________
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] (Pete Chown)

Subject: Re: SERIOUS SECURITY PROBLEM (I think)
Summary: Meaning of permissions on directories
Message-ID: <[email protected]>
Date: 2 Jun 92 20:02:30 GMT
References: <[email protected]> <[email protected]
.oxy.edu> <[email protected]>
Sender: [email protected] (The news facility)
Reply-To: [email protected] (Pete Chown)
Organization: U of Cambridge Comp Lab, UK
Lines: 29

In article <[email protected]> [email protected]
du (Oren Kastner) writes:

>As far as I know, when it comes to directories, the permissions are as
>follows:
>r - you may `read' the directory, meaning, you can do an `ls' on it and see
> the files in it.
>w - you can write to the directory, meaning you can create or delete files
> in it. (which is what happened in this case)
>x - you can `pass thru' the directory, meaning you can `cd' to it or anything
> below it (providing the permissions on the object that follow allow to be
> `cd'ed to.

In fact, if you have execute permission on a directory you can access
files which are in it. But you can't do an ls to find out which file
you want unless you have read access. Someone here created a program
which should really have been setuid, but they couldn't be bothered to
insert the code to swap ids all the time, so they just gave execute
access (only) to the program and to the directory where it kept its
private files. So you couldn't find out the name of the file with ls,
and you couldn't find it by running strings on the person's program,
because that had no read access. It wasn't marvellous but it did seem
to work...

I wonder what happens if you try to ptrace() a binary that is
execute-only?
--
__________________________________________________ No one cares how
Pete Chown, mail [email protected] (Internet) | efficiently you get
or [email protected] (Janet ๐Ÿ™‚ _| the wrong answer


[next article]
From: [email protected] (Pete Chown)

Subject: Re: Floating point problem (Where did it go? ๐Ÿ™‚
Summary: C problem?
Message-ID: <[email protected]>
Date: 2 Jun 92 20:18:36 GMT
References: <[email protected]>
Sender: [email protected] (The news facility)
Reply-To: [email protected] (Pete Chown)
Organization: U of Cambridge Comp Lab, UK
Lines: 34

In article <[email protected]> [email protected] (Dan Miner) writ
es:

[ stuff deleted ]

>Here is a blurp from my code...
>
>double scalar;
>[...]
> scalar = 480 / 640;
>
> You would expect a value between 0 and 1. You
>print it.. What do I get? 0.00000
>What is going on here?

The code you have written is equivalent to:

scalar = (double) ( ((int) 480) / ((int) 640) );

(I have put the stupid casts of 480 and 640 to integer in to make
clear what is going on)

So you are dividing the integer 480 by the integer 640, giving the
answer 0, and then converting it to a double. What you should do is
either write:

scalar = 480.0 / 640.0;

or else put explicit casts of the integers to doubles:

scalar = ((double) 480) / ((double) 640);
--
__________________________________________________ No one cares how
Pete Chown, mail [email protected] (Internet) | efficiently you get
or [email protected] (Janet ๐Ÿ™‚ _| the wrong answer


[next article]
From: [email protected] (Hans-Juergen Knobloch)

Subject: Re: How to create X11 applications with shared libraries ?
Sender: [email protected] (The Usenet-News System)
Message-ID:
Date: Wed, 3 Jun 1992 07:47:40 GMT
References: <[email protected]> <[email protected]>
Nntp-Posting-Host: dtpc2.dtro.e-technik.th-darmstadt.de
Organization: Institut fuer Datentechnik, T.H. Darmstadt
Lines: 12

The problems, I had last weekend, to create X11 applications with shared
libraries are solved.

I reinstalled the "old" GCC version 2.11c and deleted all stuff of the
GCC test versions 2.12a, 2.12b and 2.12c on my system and my first X11
test programrun immediately.

This morning I saw, that the above mentioned test versions of GCC were
removed from banjo.concert.net /pub/Linux/GCC, may be that some other
behalve me run into the same problem.

Hans-Juergen Knobloch


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

Subject: Re: Elm -- STILL giving Segmentation errors
Message-ID:
Date: 2 Jun 92 00:16:15 GMT
References:
Organization: Currently Lacking Organization
Lines: 21
X-NewsSoftware: Amiga GRn V1.29 4/25/92 (Beta) by Mike Schwartz

In article [email protected] (Steven M. Palm)
writes:
.
.
.
>it generates mail, and elm can read this. However, when I hit the S)end key
>in Elm, I get segmentation errors.

I found the problem.... ๐Ÿ™‚

in date.c, function get_arpa_date, I believe, it is wanting to call a
function "timezone" to retrieve the local timezone. Linux's time.h file
defines timezone as an extern long or similar. Well, it seems that when
it tried to execute a function that really wasn't.... Seg error.

Can anyone enlighten me on a timezone function, or if it's in Linux, how
to get to it? For the time being I've compiled it in.

--
-- [email protected] -----------------------------------

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


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

Subject: Re: fdformat
Message-ID:
Date: 2 Jun 92 00:19:18 GMT
References: <[email protected]>
Organization: Currently Lacking Organization
Lines: 20
X-NewsSoftware: Amiga GRn V1.29 4/25/92 (Beta) by Mike Schwartz

In article [email protected] (Michael Pereckas)
writes:
>[email protected] (David.L.) writes:
>
>>Compiled 'fdformat.c' and ran it on 1.44 & 1.2 Mb. diskettes ๐Ÿ™‚ happy me ๐Ÿ™‚
>>BUT - the results are unreadable by mtools or dos ๐Ÿ™
>
>fdformat only low-level formats the disk. It does not write any
>filesystem information. To use the fdformated disk for a DOS filesystem,
>run mformat on the disk. Mformat writes DOS filesystem information to
>the disk, but does not low-level format it. To put a Linux filesystem
>on the disk, use mkfs.

On my system, fdformat works fine on already-formatted disks, but gives
reset-floppy errors and dies on a never-formatted disk. Both 5.25" and 3.5"
disks.

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


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

Subject: Re: Floating point problem (Where did it go? ๐Ÿ™‚
References: <[email protected]>
Date: 3 Jun 92 08:20:01 GMT
Organization: Rechenzentrum Uni-Duisburg
Message-ID:
Lines: 38

In <[email protected]> [email protected] (Dan Miner) writes:

[deleted som stuff]
>Here is a blurp from my code...

>double scalar;
>[...]
> scalar = 480 / 640;

> You would expect a value between 0 and 1. You
>print it.. What do I get? 0.00000
>What is going on here?

> I have had alot of little problems with 2.11c. Someone
>thinks that tar is the fault. I think I had replaced for untarring
>the files. Has anyone had similar problem?

> (I have no 387...)

That's no problem, that's absolute correct C behaviour.
What happens is:

480 is divided by 640, but since both 480 and 640 are integers, the /
is an integer division, so you get 0. Then this zero is converted to
double and this gives your 0.00000.

So you should use

scalar = 480.0 / 640.0;

That will do it.
Hope that helped you,
Michael
--
Michael Staats
Theoretische Tieftemperaturphysik
Universitaet Gesamthochschule Duisburg
[email protected]


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

Subject: uemacs3.11c
Keywords: microemacs 3.11c available ftp banjo.concert.net
Message-ID:
Date: 3 Jun 92 08:50:30 GMT
Organization: Netcom - Online Communication Services (408 241-9760 guest)
Lines: 8

I uploaded a Linux port of uemacs3.11c to banjo.concert.net.

/pub/Linux/Incoming/ue311c.tar.Z <- executable
/pub/Linux/Incoming/ue311c.s.tar.Z <- source


Toshiyasu Morita
[email protected]


[next article]
From: Paul Allen

Subject: Re: MGR and X
Message-ID: <[email protected]>
Sender: [email protected] (Mr Background)
Reply-To: [email protected]
Organization: The Internet
Date: Wed, 3 Jun 1992 01:58:32 GMT
Lines: 65


[email protected] (Kevin C. Dorff) wrote:

|I know what X is and what it gives you, but if someone could quickly
|compare and contrast MGR and X for me, along with their speed, disk
|requirements, capabilites, memory requirements, etc, I would REALLY
|appreciate it.

If X is BSD, MGR is V7. If X is the Cadillac, MGR is the VW. Building
X takes hundreds of megabytes of disk, although you can get by with
much less if you're selective and practice some tape-swapping.
(And, if you don't mind not having sources, you can have a working
basic X setup in 10Mb or so.) Building MGR only needs about 10Mb or so
for the whole thing. X has resources, and zillions of applications,
and multiple toolkits, and industry-wide battles over GUI
look-and-feel. MGR has never heard of a "resource file", comes with only
a smattering of clients, has no toolkits, and is massively ignored by the
industry titans who worry about look-and-feel. X is complex. MGR is simple.

MGR provides multiple overlapping resizeable windows, each of which
can display both text and graphics. Text can be cut and pasted between
windows. An application running in a window can control the font used
to show text, select which events it wants to hear about, iconify itself,
etc. MGR has a simple mechanism for creating and using pop-up menus.
MGR uses the same client/server arrangement used by X, but MGR
can work over any reliable byte-stream without knowing anything about
the underlying protocol. I have an MGR client on my Sun at work that
gathers NFS statistics and produces a moving strip-chart. I can run
this client while logged in from home over a 2400 baud modem, and the
graphics work fine. I currently run MGR on my 4Mb 386 under Minix, and
have enough memory for half a dozen windows. (Minix doesn't have shared
text, shared libraries, or paging, so MGR may do a little better under
Linux.) I have a Trident-based SVGA card. Text scrolling is, well, slow.
It's possible that some work on the bitblt code will help this, but I
think the real fix is to buy a better card. The X folks all seem pretty
happy with their ET4000's.

|I would LOVE to run windows under Linux if possible, but since I am
|not networking or developing X apps (at least right now) MGR may be
|enjoy.

If you're not wedded to X, MGR might be a reasonable alternative.
It has all the basic stuff a window system must provide, but nothing
that's not absolutely essential. Experienced X programmers will probably
find it frustrating.

|ALSO what video cards they will run on will be VERY helpful.

I'm on shakey ground here, but I have a vague recollection that the
Minix port of MGR worked on any video card that could be made to look
like an EGA. It turns out that some VGA modes, including 800x600x16,
put the adapter in a state that looks just like a really big EGA.
Since a hardware vendor would have to go out of their way to make their
800x600x16 mode *not* look like an EGA, it's a good bet that *all*
SVGA's work essentially the same in this mode. The differences would
be in the specific values you need to stuff in the registers in order
to achieve the proper mode. One determines these values with a little
program that puts the adapter in the desired mode and then reads out
the register values. I use 800x600 for MGR under Minix, and that's
about as high as I want to go on my 14" screen.

I hope this helps!

Paul Allen
[email protected]


[next article]
From: Brad Taplin

Subject: Re: your problems
Message-ID: <[email protected]>
Summary: Nicely please?
Sender: [email protected]
Organization: Indiana University
Date: Tue, 2 Jun 1992 21:16:13 -0500
Lines: 25

[email protected] writes:
>|Hello. I downloaded these files: install.doc
>|and then unzipped them on our rawrite.doc
>|Gateway 386sx16 SVGA drive d: rawrite.exe
>|and then used rawrite to make bootimag.zip
>|a 360K bootable disk from the rootimag.zip
mvore writes...
>But did you read the install.doc file - You need HD disks for both
>the root and boot images!! They must be the same density!!...
>I hate to say it but RTFM

Got the information from a dozen helpful sources. Thanks all!
For the record, the install.doc file includes NOTHING about the
required disk size. I request you subtract the F from your RTFM.
Is there a more complete and comprehensive install file around?

>I have 0.95 on about 7 megs - a lot of 'stuff' that needs cleaning out.
>So a bare minimum will probably fit, enough to convince your feancee to
>let you spring for a good size disk - or even a bigger and better
>machine.. Good luck!!

Thank you. Sincerely. ๐Ÿ™‚
--
[email protected] or simply [email protected] will
appreciate your intelligent suggestions, anecdotes, comments & offers.


[next article]
From: [email protected] (Vern Paxson)
Newsgroups: comp.os.linux,gnu.gcc.bug,gnu.utils.bug

Subject: Re: flex bug?
Summary: input() fix is on 2.4 to-do list
Keywords: flex
Message-ID: <[email protected]>
Date: 2 Jun 92 16:34:09 GMT
References:
Reply-To: [email protected] (Vern Paxson)
Followup-To: comp.os.linux,gnu.gcc.bug,gnu.utils.bug
Organization: Lawrence Berkeley Laboratory, Berkeley
Lines: 25
To: [email protected]
NNTP-Posting-Host: 128.3.112.42

In article [email protected] (Ben Cox) writes:

>
> According to the lex manual (in the BSD4.3 manuals), input() should
> just get the next character off the input stream and return it,
> without putting it into yytext or anything.

Yes, this is basically a bug (though it's possible to read POSIX as
allowing it). Fixing it is on the to-do list for release 2.4, which
still has no scheduled release date. The problem is that input()
restores the character that was overwritten by the '\0' that terminates
yytext; why it works on the Sequent is a mystery to me.

> /* Note: I used \"[^"]* instead of \"[^\"]* because the BSD4.3 manual
> (yes, I know that's for lex and not flex) says " is not special inside
> [], although my nutshell book says to use `\"[^\"]*': this part works,
> so nutshell must be wrong... ๐Ÿ™‚ */

The only magic characters inside of []'s are: '\', ']', and newline,
so '"' does not need to be escaped (though it doesn't hurt to do so).

Vern

Vern Paxson [email protected]
Computer Systems Engineering ucbvax!ee.lbl.gov!vern
Lawrence Berkeley Laboratory (510) 486-7504


[next article]
From: [email protected] (Claude Lecommandeur)

Subject: SCSI and Future Domain controler
Message-ID: <[email protected]>
Date: 3 Jun 92 07:37:53 GMT
Sender: [email protected]
Organization: Ecole Polytechnique Federale de Lausanne
Lines: 22


I'have just read in the SCSI FAQ that patches for the Future
Domain controler exists. Where can i find them ? I have just
tried the last version of SCSI from headrest.woz.colorado.edu,
but it doesn't seem to see my Syquest disk (scsi: 0 disks, 0
tapes at boot).

config : I have a 486/33 running Linux 0.96, and, i #defined
whatever string containing "FD" in my config.h file before
rebuilding the kernel.

Thanks for any hints

--


Claude Lecommandeur
Service Informatique Central
Ecole Polytechnique Federale de Lausanne
1015 LAUSANNE (SWITZERLAND)
E-Mail : [email protected]
Tel : (41 21) 693-22-97


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

Subject: Info on Esprit Systems Executive 10 Terminal
Message-ID: <[email protected]>
Date: 3 Jun 92 08:04:51 GMT
Sender: [email protected] (Mr Background)
Reply-To: V5068U%[email protected]
Organization: The Internet
Lines: 15

Hi all:
I am trying to get an Esprit Systems Executive 10 terminal working on
Linux and I am having trouble getting the termcap together. Does anyone have
control code info on this beast? Or (*gasp*) a termcap entry for it???
A related question: has anyone used the %c format in termcap and gotten
it to work? I need to set the cursor position code to E^Q%p2%c%p1%c (push the
second parameter (col) and output as a char, push the first parameter and
output as a char (row)). This seems rather trivial but the cursor does not end
up in the right spot (micro-emacs scribbles on the MIDDLE of the screen ๐Ÿ™ )
I can do this manually using echo commands but curses programs barf. Likewise,
termcap entries do not seem to like spaces in them, either literally or in
octal form. This terminal has LOTS of escape sequences with spaces in them. I
have some limited documentation on this beast but need a full list of codes.
Any hints/tips anyone?
jim [email protected] ( moe is sick. )


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

Subject: X11 & xterm
Message-ID:
Date: 2 Jun 92 16:43:03 GMT
Sender: [email protected] (USENET News Owner)
Organization: Coventry Polytechnic, Coventry, UK
Lines: 21
Nntp-Posting-Host: cc_sysh


I have got X up and running in the past few days, getting over the problems
with the wrap screen, I see others are having problems with, by changing
the video card cycle var's (I think) in the config file.

But now I have another problem... On running xterm the xterm-window opens
and the prompt comes up but I cann't type anything into it.


Has anyone any ideas what the problem maybe ?


Cheers



--
---
Matthew Roderick /Email: [email protected] | \ //\ |`. |
"""""""""""""""" /Interests: Fine Art & Good Times | \/\//--\|_| |
/Quote: If I felt any worst I'd be dead, If I felt any better I wouldn't be me


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

Subject: Re: SCSI and Future Domain controler
Message-ID: <[email protected]>
Sender: [email protected] (The Daily Planet)
Nntp-Posting-Host: kinglear.cs.colorado.edu
Organization: University of Colorado at Boulder
References: <[email protected]>
Date: Wed, 3 Jun 1992 12:54:22 GMT
Lines: 16

In article <[email protected]> [email protected] (Claude Lecommandeur) write
s:
>
> I'have just read in the SCSI FAQ that patches for the Future
>Domain controler exists. Where can i find them ? I have just

omicron.cs.unc.edu:/pub/faith is Rick's preliminary Future
Domain driver, which uses the command() interface. The drawback
of this interface is that it is not interrupt driven, and
processes will not be scheduled during disk access.

I've been told that Rick is done with the interrupt driven driver
that lacks these drawbacks - when this is made available, it will
be incorporated into the SCSI package which will be fed back into
the "stock" distribution.




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

Subject: Re: Linux and large (SCSI) hd drives
Message-ID: <[email protected]>
Sender: [email protected] (The Daily Planet)
Nntp-Posting-Host: kinglear.cs.colorado.edu
Organization: University of Colorado at Boulder
References: <[email protected]>
Date: Wed, 3 Jun 1992 13:06:30 GMT
Lines: 22

In article <[email protected]> [email protected] (Jan Nicolai Langf
eldt) writes:
>
>I have read the faq and the scsi faq.
>
>Does linux suffer under the same restrictions on hd geometry that dos
>does? I'm looking at a good deal for a 660Mb (HP 97548) SCSI disk, and
>was wondering if I can actually use the whole disk? (and btw, what IS
>the restrictions if any?)

As far as disk size supported, the SCSI routines support SCSI disks up
to 2 terrabytes in size.

The limitations stem from the Minix filesystem, which has a 64M
limit (you can only have one filesystem per partition),
and the the fact that at this time the SCSI drivers only support the four normal
partitions.

So, until changes are made, you can only use ~256M of it under Linux.

However, I will put in support for extended partitions RSN,
and people are working on alternate file systems.



[next article]
From: [email protected]

Subject: X386 enhanced
Message-ID: <[email protected]>
Sender: [email protected]
Nntp-Posting-Host: vger.cheme.cornell.edu
Reply-To: [email protected]
Organization: Cornell University School of Chemical Engineering
Date: Wed, 3 Jun 1992 13:16:04 GMT
Lines: 4

I read about an enhanced X386 1.2 that's being developed. Sounds good.

Elan Feingold
Cornell University


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

Subject: Re: Linux swapping
Message-ID: <[email protected]>
Organization: Printer Products Eastman Kodak
X-Newsreader: Tin 1.1 PL3
References: <[email protected]>
Date: Fri, 29 May 92 11:19:35 GMT
Lines: 41

[email protected] (Bluebeard) writes:
>
>I have 2megs of memory and 4megs of swap.
>
>Does this give me a total of 6megs or only 4 ?
>
>A unix expert, told me that the actual memory maps onto the swap, so only the
>swap space above the system memory is available.
>Is this true on linux ?
>
>I'm considering upgrading to 4meg of RAM, but as I've only got a 40meg
>partition, I wasn't intending to increase the swap space.
>
>
>Thanks for any info on the above.
>
>
>--
>***********************************************************************
>* FLESH: Charlie Freckleton * JANET: [email protected] *
>* ALIAS: Bluebeard * or : [email protected] *
>***********************************************************************

this is not true under linux. under bsd systems there are two memory
management schemes used which use the disk: swapping and paging. paging
is what linux uses: when there is no physical memory for a page, one
is evicted to the disk (our swap space) to make room. swapping is the
wholesale motion of a process to disk, and is done by the swapper process.
if your process doesn't do anything for a while the system decides its
better to hedge it bets and evict the whole thing to disk, betting on
locality of reference on a process granularity, i guess.

linux simply does paging, using as much swap as is available. if there
isn't any available, the program waits. linus, correct me if i'm wrong,
but this is what i remember when i hacked with it.

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


[next article]
From: [email protected]

Subject: X1.0/GCC2.11c Problems
Message-ID: <[email protected]>
Sender: [email protected]
Nntp-Posting-Host: vger.cheme.cornell.edu
Reply-To: [email protected]
Organization: Cornell University School of Chemical Engineering
Date: Wed, 3 Jun 1992 13:33:21 GMT
Lines: 14

o System: 386/SX 16, 4Mb, LINUX 0.96a, X1.0, Gcc2.11c (the magic combo)
o I get errno 3 & 4 when I run startx. "Unable to connect to server"
and "process does not exist" or something like that. It bugs me because
X0.1 worked fine, and I ran inst2.x and when I run an X client it
says "cannot open display :0" instead of "segmentation fault"
but I just can't figure out the problem. My /dev/tty0 and /dev/console
are fine too.

Hellllp!!!!!

Thanks,

Elan Feingold
Cornell University


[next article]
From: [email protected] (Rhys Weatherley)

Subject: Re: Dexxa mice
Message-ID: <[email protected]>
Date: 3 Jun 92 02:41:01 GMT
References: <[email protected]>
Sender: [email protected]
Reply-To: [email protected]
Lines: 39

In <[email protected]> [email protected] (Ken Corey - Operator) write
s:

>Anyone out there have the programming manuals for Dexxa Mice?

Not me. ๐Ÿ™

>I'm interested in finding out why Dexxa mice aren't Microsoft mice compatible
>without a driver. *grrr*

Simple. The mouse company is lying when it says it is Microsoft compatible.
The mouse company can produce a mouse driver for DOS that will make it *look*
like a Microsoft mouse to DOS applications, but if the protocol used between
the mouse and the computer on the serial link is not the same as that used by
true-blue Microsoft mice, then it's wrong to call it Microsoft compatible IMHO.
The Linux drivers in X and MGR (among others) need to talk in the protocol
used on the serial link - Linux doesn't have the benefit of a generic device
driver interface like DOS does.

Classic case of "DOS is the only thing that matters" syndrome on the part of
the mouse manufacturers. It's almost impossible to find real information on
the protocols used. ๐Ÿ™

Anyway, even though I don't have a Dexxa mouse, I'm trying to figure out the
protocol based on what Dexxa people tell me so I can include it in my mouse
library for Linux (which will eventually make it into MGR, and maybe X if
Thomas Roell includes it). I have a test program here that decodes some of
the Dexxa behaviour (mainly the buttons), and I'm looking for people to run
this and send me the output so I can stare at it and figure out the rest of
the protocol for the mouse movements. Please contact me if you want to help
out.

Any technical data on the Dexxa mouse protocol is *very* welcome.

Cheers,

Rhys.
--
Rhys Weatherley, University of Queensland, Australia.
[email protected] "I'm a FAQ nut - what's your problem?"


[next article]
From: [email protected] (Lawrence C. Foard)

Subject: Gnu lib man pages?
Message-ID: <[email protected]>
Date: 2 Jun 92 20:21:00 GMT
References: <[email protected]>
Sender: [email protected]
Organization: The Turing Project, Charlottesville Virginia.
Lines: 14

Are there man pages available for the gnu library? The man pages I
could find on tsx-11.mit.edu only seem to only be for the gnu programs.

.96a kernel bug? -
I recently set up my system to work over a modem, and it appears that
the kernel echo's a 127 for a delete rather than ' ', is this
a bug or a POSIX "feature"? If its a feature, is there a way of having
it send the ' ' sequence instead? For the time being I just
hacked the kernel to fix it. But I'm not sure if there is another way
to do it.

P.S. Having full source code makes life so much easier! Its so nice to
be able to fix it when its broken instead of banging your head
against the wall trying to figure a way around an OS problem ๐Ÿ™‚


[next article]
From: [email protected] (Jan Nicolai Langfeldt)

Subject: Linux and large (SCSI) hd drives
Message-ID: <[email protected]>
Date: 3 Jun 92 08:14:37 GMT
Sender: [email protected] (Nicolai Langfeldt)
Organization: Folkeaksjonen for (tungt) eple cider p} Inf. stua
Lines: 13
Nntp-Posting-Host: maud.ifi.uio.no
Originator: [email protected]


I have read the faq and the scsi faq.

Does linux suffer under the same restrictions on hd geometry that dos
does? I'm looking at a good deal for a 660Mb (HP 97548) SCSI disk, and
was wondering if I can actually use the whole disk? (and btw, what IS
the restrictions if any?)

(adaptek 1542 controler if it matters)

Thanks!

Nicolai


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

Subject: Re: X -> can't access display 0:
Message-ID: <[email protected]>
Organization: Printer Products Eastman Kodak
X-Newsreader: Tin 1.1 PL3
References: <[email protected]>
Date: Fri, 29 May 92 15:23:19 GMT
Lines: 39

[email protected] (Jay Lawrence) writes:
>
>I have gotten X up and running on my system with little trouble!
>Later I will post my Xconfig info for my system (I am not near
>my machine).
>
>Anyway, xinitrc works for the most part. I get xclock and xterm
>find, twm starts up. But then if I want to run any more xprograms
>I get an error to the effect of can't access display 0:. What
>have I done wrong? Well, I have set the /usr/X386/bin/* to root/root.
>My environment has DISPLAY="0:". I can't think of anything else.
>Hmmm.

perhaps you need to export this variable? i use:
DISPLAY=":0"
export DISPLAY
otherwise this won't get exported to another process' environment

>
>Next question, how _exactly_ do I join the xwindows mailing list???
>(Perhaps listing the command to get the different topics from the
> mailing list - ie/ send a command such as 'topics' to linux-activists-
> requests ... or list them in the help file?)

send mail to [email protected] with the message
X-MN-Admin: join x11

>
>Thanks for the great system that has been put together guys! I hope
>that I can be of some help in the future!!!
>
>Jay Lawrence
>UWaterloo

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


[next article]
From: [email protected] (Gregor Schmid)

Subject: weird problem with (fseek/ftell/fwrite/putc)
Message-ID:
Sender: [email protected] (News Manager)
Nntp-Posting-Host: fb3-s7.math.tu-berlin.de
Organization: FB3, TU-Berlin, Germany.
Date: Wed, 3 Jun 1992 11:33:06 GMT
Lines: 23

Hello,
as I tried to compile zoo using gcc 2.11c / linux 0.96a & patch1
I got a problem which I tracked down to the following:

when adding a file to an archive zoo first creates an empty directory
entry in the archive, then compresses the file and stores it and then
fseeks back to the position of the directory entry and overwrites it.

This works fine with standard compression.

If high compression is used, the compressed file is not stored using
blocktype read/write operations, but using putc. I inserted an ftell
into the code after the following fseek, which reports the right
position. Nevertheless the following fwrite doesn't go there, but
on the next 1024 bytes boundary. No need to say that this corrupts
the archive.

If this is a bug, is it in the gcc libraries or in the kernel?
Is there any immediate fix? I would like the zoo high compression
to work reliably to use zoo for my backups.

Thanks for 'listening',
Greg


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

Subject: Re: Stereo sound DAC(Was Re: Help for X11)
Message-ID: <[email protected]>
Date: 3 Jun 92 13:18:29 GMT
References:
Distribution: comp.os.linux
Organization: University of Bradford, UK
Lines: 31
Originator: [email protected]
Nntp-Posting-Host: ccw201

[email protected] (Erik Green) writes:

>B: It works on the parallel port, so we don't need to write a new kernel
> driver for it.

Unfortunately all my sound players and things (I wrote Modplay, and I also
use linux) need to hook the timer interrupt and they reprogram it to run
at many kHz. This isn't something thats going to work or be of use in
Linux.

I was thinking that it would be *really* easy to get a /dev/audio device
to output through sound cards that have DMA - like the Soundblaster (yuk)
and Covox Sound Master etc... I have code that will do DMA but I wouldn't
have a clue how to get a device driver working for linux... anyone interested
in having a go with me?

> Well, since so many people have mailed me asking for more information,
>I thought I'd post the sources for the DAC chips needed for the circuit.

The docs have lots of the IBM character set in them - not something to
post to a newsgroup! If you just want to get the hardware.doc without the
several hundred k of Modplay,

/misc/msdos/mp/hardware.doc on ftp.brad.ac.uk


Mark
--
Mark J Cox -----
Bradford, UK ---


 December 12, 2017  Add comments

Leave a Reply