From:     Digestifier <Linux-Activists-Request@news-digests.mit.edu>
To:       Linux-Activists@news-digests.mit.edu
Reply-To: Linux-Activists@news-digests.mit.edu
Date:     Thu, 25 Jun 92 13:30:17 EDT
Subject:  Linux-Activists Digest #341

Linux-Activists Digest #341, Volume #2           Thu, 25 Jun 92 13:30:17 EDT

Contents:
  Logitech mice problem revisited. (Claude Lecommandeur)
  freeze gcc to 2.2.2... (Chuck Boyer)
  Bug in fsqrt emulation (Linus Benedict Torvalds)
  Re: Logitech mice problem revisited. (Linus Benedict Torvalds)
  ps (Paul Douglas Page)
  your idea - good & bad (Mike Vore W3CCV)
  Re: need X11 for DOS (SHEHAB AM)
  need help with olvwm (Brian K. Talley)
  Re: your idea - good & bad (David Boyce)
  Tape drives/drivers (Andrew Zwerin)
  mcopy won't work??? (Randall A Helzerman)
  Re: .96b, shoelace, sharlibs, etc (Cory West)
  determining which library an executable needs (Claude Morin)
  Re: Error installing Root-0.96 (Martin Junius)

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

From: lecom@slhp1.epfl.ch (Claude Lecommandeur)
Subject: Logitech mice problem revisited.
Date: 25 Jun 92 09:18:50 GMT


    I did much investigation on why my logitech mouse (MouseMan M-MC13)
doesn't work with X11. This mouse is supposed to be Microsoft compatable
but isn't. The normal sequence (microsoft) for button1 click without
moving is hexa : 60 0 0 (seven bits, no parity, 2 stop bits) i verified
it is true with another microsoft compatable mouse (genius). But the
logitech one send, (or at least i read on /dev/ttys1) : 60 40 40.

    This mean that the seventh bit on the 2 last bytes are set to one
and they should be set 0. I was told by Logitech that actually, there
is a difference between their mice and microsoft : there is only one
stop bit instead of two. But setting -cstopb does nothing at all.

    Should it be possible that the missing stop bit set the seventh
bit to 1 ? Should it be possible that there is a bug in the low level
serial driver in Linux ?

    I tried hundreds of settings on /dev/ttys1 without success. I would
be very interested in knowing has investigating this problem.


  Thanks for any hints.

claude.

-- 


                    Claude Lecommandeur
                    Service Informatique Central
                    Ecole Polytechnique Federale de Lausanne
                    1015 LAUSANNE (SWITZERLAND)
                    E-Mail : lecom@sic.epfl.ch
                    Tel : (41 21) 693-22-97

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

From: boyer@sumax.seattleu.edu (Chuck Boyer)
Subject: freeze gcc to 2.2.2...
Date: 25 Jun 92 09:24:20 GMT


 
I strongly agree and reiterate; we 'must' agree at some stopping
off point on the compiler, gcc2.2.2 would be a good point to 
freeze. Stabilize the Linux compiling process......
boyer@sumax.seattleu.edu

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

From: torvalds@klaava.Helsinki.FI (Linus Benedict Torvalds)
Subject: Bug in fsqrt emulation
Date: 25 Jun 92 11:52:06 GMT

I won't make a patch2 for this one, but 0.96b(.pl1) has a one-character
bug in the fsqrt emulation code that results in fsqrt resulting in zero
all the time.  The following patch corrects it (thanks to Rick Sladkey
for noting it: I hadn't tested the code, so I never saw it...)

========== cut cut ==========
--- sqrt.c.orig Thu Jun 25 00:00:00 1992
+++ sqrt.c      Wed Jun 24 23:59:37 1992
@@ -55,7 +55,7 @@
        src[3] = s->b;
        d->exponent = 0;
        d->a = d->b = 0;
-       if (exponent)           /* fsqrt(0.0) = 0.0 */
+       if (!exponent)          /* fsqrt(0.0) = 0.0 */
                return;
        if (!src[2] && !src[3])
                return;
========== cut cut ==========

That is, add the missing '!' and recompile the kernel.  This bug
naturally doesn't show up on machines with a co-processor, and isn't
noticeable unless you have programs compiled with gcc versions > 2.1, as
earlier versions didn't use the fsqrt instruction. 

At least one other known bug in 0.96b.pl1 is that a serial line open
that fails due to EBUSY (ie another serial line using the same irq-line
is active) will still incorrectly add 1 to the tty->count.  You don't
usually see this bug, but it can result in bogus EBUSY errors later on. 
This and the fsqrt bug are already fixed in my version, and I'll make
patch2 available when I have some more fixes. 

Other things added by patch2 will include correct CS5-8 and parity
handling on the serial lines (and HUP on setting the speed to 0). 

                Linus

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

From: torvalds@klaava.Helsinki.FI (Linus Benedict Torvalds)
Subject: Re: Logitech mice problem revisited.
Date: 25 Jun 92 12:36:36 GMT

In article <3264@sicsun.epfl.ch> lecom@slhp1.epfl.ch (Claude Lecommandeur) writes:
>
>    I did much investigation on why my logitech mouse (MouseMan M-MC13)
>doesn't work with X11. This mouse is supposed to be Microsoft compatable
>but isn't. The normal sequence (microsoft) for button1 click without
>moving is hexa : 60 0 0 (seven bits, no parity, 2 stop bits) i verified
>it is true with another microsoft compatable mouse (genius). But the
>logitech one send, (or at least i read on /dev/ttys1) : 60 40 40.
>
>    This mean that the seventh bit on the 2 last bytes are set to one
>and they should be set 0. I was told by Logitech that actually, there
>is a difference between their mice and microsoft : there is only one
>stop bit instead of two. But setting -cstopb does nothing at all.

Ok, it does indeed seem to be the bad support for CS7 in the linux
kernel.  It seems I'll have to make patch2 available earlier than I
thought: I have implemented full CS7 support now, and if that is what it
takes to get some mice to work with linux, patch2 should correct it. 
Expect the new patch tomorrow. 

Thanks for a good report: these kinds of things really help when it
comes to correcting kernel features.  As it happened, I added the
support before I saw this post, but that's just luck. 

Patch 2 also supports the CSTOPB bit, which earlier linux versions
ignore (along with PARENB/PARODD and the CSx settings).  I hope the mice
problems will be gone after this patch. 

                Linus

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

From: pdp@ariel.its.unimelb.EDU.AU (Paul Douglas Page)
Subject: ps
Date: 25 Jun 92 12:45:15 GMT

Does anyone know if their is a later version than ps0.95c.tar.Z ?

I can't seem to get that to compile with linux 0.96b .
The error message says :-

In function show_regs :

structure has no member named 'alarm'


Any ideas?

Thanks in advance.

















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

From: Mike Vore W3CCV <mvore@fedix.fie.com>
Subject: your idea - good & bad
Reply-To: mvore@fedix.fie.com
Date: Thu, 25 Jun 1992 13:22:13 GMT



|         I was just thinking... when and if TCP/IP support is added to Linux
| (and I believe someone is working on it) then presumably ftp will
| work on it. At that point, could not someone who has a Linux

Yes someone is working on it.  A friend of mine just downloaded all the
tcp/ip sources from me last night and is currently trying to get the
kernel code for a 'real' ftp (not ka9q type of application).


| machine at an Internet site set it up as an aftp site, allowing
| unrestricted access to the entire file structure? Normally
| people setting up an anonymous ftp site go to great lengths
| to restrict access, but since every part of Linux is freeware
| anyway this wouldn't be necessary (the user might want to protect
| some private files, but regular file permissions should suffice for
| that).

Would you leave your system open for any hacker to come in and walk
through - with no restrictions???

| -> Working config files available as examples. As "shipped",
| files like /etc/rc and /etc/inittab are generic. It could be
| very instructive for novice users to grab a customized, working
| copy from some site

These could be tar'd into  generic-setup.tar without having access to
ones personal - and although they are used by everyone - they are far
from generic.  The generic version is just a starting point - just as
config.sys and autoexec.bat on Ms.Dog systems are not generic.


| -> De facto placement conventions. It's nice to release a
| document saying "such-and-such file belongs in such-and-such
| directory", but realistically many people won't read or remember
| something like that. I don't think anyone would have a problem
| figuring out that if they grab a library from directory /X/Y/Z
| on the ftp machine, it belongs in directory /X/Y/Z on theirs.

If one does a "cd /bin" inside an ftp session, then "get login" or
whatever, that does _not_ put it into /bin on the local macine!  It
still goes into the current directory.  The user then still has to
remember where that dammed file goes.

-- 
/*---------------------------------+--------------------------------*/
  Michael Vore  W3CCV              |   mvore@fedix.fie.com
   System Administrator            |   mvore@digex.com
   Federal Information Exchange    |   voice (301)975-0103
/*---------------------------------+--------------------------------*/

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

From: ams90@ecs.soton.ac.uk (SHEHAB AM)
Subject: Re: need X11 for DOS
Date: 25 Jun 92 10:10:38 GMT

Ok, point taken that X might need an Unix server in background. BUT is
there a PD X for DOS?

ams...

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

From: talley@ashleigh.Kodak.COM (Brian K. Talley)
Subject: need help with olvwm
Date: 25 Jun 92 12:20:12 GMT

Hi -

   What, exactly, should the Linux version of olvwm do, anyway?  On my Sun
it manages virtual screen displays.  When I first saw it for Linux, I thought
olvwm would do roughly the same thing, but when I run it, it just starts twm
with no virtual window manager. (One concern of mine was that I'm not running
olwm at all.  If it exists for Linux, sould someone please tell me where I can
get it?)

   Anyway, please reply via email (to talley@acadia.kodak.com) if you have
olvwm working as it should (first define "should").  *PLEASE* also reply if you
have a different window manager (other than twm) running.

Thanks!

--Brian

P.S.  I grepped the FAQs for olwm, olvwm, look, openlook and openwindows but
found nothing.  Either I've got old copies of the FAQs, or this isn't a problem
many people experience.

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

From: dsb@world.std.com (David Boyce)
Subject: Re: your idea - good & bad
Date: Thu, 25 Jun 1992 14:58:14 GMT

In article <1992Jun25.132213.2379@athena.mit.edu> mvore@fedix.fie.com writes:
>Would you leave your system open for any hacker to come in and walk
>through - with no restrictions???

    For one thing, I wasn't talking about login access. I was talking
about aftp (read-only) access. And yes, considering that I use
my Linux machine only for working on Linux, at least for now,
I would (but I only have a dialup connection, so I can't).
Sure, there will be people who will have private things, but
I bet a lot of the big contributors are like me in that their
Linux machines are used for Linux development, period.
    Second, what's to say someone couldn't dedicate a machine
to something like this? When you look at the price of a
minimal Linux 386 box, it's a tiny contribution compared to
the time all these presumably well-paid (or at least well-payable)
people are contributing. I don't mean anyone should actually
have to buy anything, but I suspect most large universities
and institutions might have an unused PC lying around.
    What I'm saying is that certain people have already volunteered
to support "standard" releases. These people will presumably
test them a few times before they release them. Thus they would
presumably have a test box which they install on, and they
probably don't keep anything else important there, since
frequent reinstallations are dangerous. When the last test
is done and the "release" is made, these machines will in
effect be reference platforms. Why not, then, open them up for
reference?

>| -> Working config files available as examples. As "shipped",
>| files like /etc/rc and /etc/inittab are generic. It could be
>| very instructive for novice users to grab a customized, working
>| copy from some site
>
>These could be tar'd into  generic-setup.tar without having access to
>ones personal - and although they are used by everyone - they are far
>from generic.  The generic version is just a starting point - just as
>config.sys and autoexec.bat on Ms.Dog systems are not generic.

    If you can't think of a situation where unsuccessful user X might
not want to have a look at successful user Y's Xconfig or inittab, I don't
think you're thinking very hard. And demonstrating that this can be done
other ways is a waste of bandwidth - I know it can be done
other ways. I'm saying this way is simple, natural, and
easily explained, and asking why NOT do it.

>| -> De facto placement conventions. It's nice to release a
>| document saying "such-and-such file belongs in such-and-such
>| directory", but realistically many people won't read or remember
>| something like that. I don't think anyone would have a problem
>| figuring out that if they grab a library from directory /X/Y/Z
>| on the ftp machine, it belongs in directory /X/Y/Z on theirs.
>
>If one does a "cd /bin" inside an ftp session, then "get login" or
>whatever, that does _not_ put it into /bin on the local macine!  It
>still goes into the current directory.  The user then still has to
>remember where that dammed file goes.

    I don't recall saying that it did go to /bin. What I said was that
it would be pretty obvious that it belonged there if you had just
seconds ago cd-ed  to /bin iin order to get it.

>/*---------------------------------+--------------------------------*/
>  Michael Vore  W3CCV              |   mvore@fedix.fie.com
>   System Administrator            |   mvore@digex.com
>   Federal Information Exchange    |   voice (301)975-0103
>/*---------------------------------+--------------------------------*/


-- 
David Boyce     dsb@world.std.com       617-576-1540

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

From: zwerin@cis.ohio-state.edu (Andrew Zwerin)
Subject: Tape drives/drivers
Date: 25 Jun 92 16:07:35 GMT

Can someone tell me if there are any tape drive drivers already
written for linux and which interfaces they are for.

I am considering purchasing an Archive VP60i, with a QIC02 interface
because my department has several QIC11/24 drives, with SCSI
interfaces.  I would like to be able to use the drive to move new
linux releases from the net to my PC (much faster then by modem) as
well as backing up my system.

Also, someone mentioned that gnu tar might be a problem, i.e. the
department has Suns and I have a PC and this is supposed to be a
problem?  Can anyone shed some light on this matter.

Any info would be appreciated, I do not know much about tape drives.

thanks,
andy

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

From: helz@ecn.purdue.edu (Randall A Helzerman)
Subject: mcopy won't work???
Date: 25 Jun 92 16:09:38 GMT

I downloaded mtools and everything seems to work fine except for
mcopy.  mdir, mdel, ect. all do what they should, but when I try

        mcopy a:foo.bar foo.bar

nothing happens.  No errors, no messages of any kind--but no files
are copied either.

Could someone tell me what I'm doing wrong?  My /etc/mtools file has
the single line:

A /dev/PS0

in it and I am running .95a.


Thanks alot for helping me solve this most frustrating problem.

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

From: corywest@rice.edu (Cory West)
Subject: Re: .96b, shoelace, sharlibs, etc
Date: Thu, 25 Jun 1992 16:48:52 GMT


In article <1992Jun25.070801.11566@sol.UVic.CA> pmacdona@sanjuan 
(Peter MacDonald) writes:

>   The following properties might be desirable:
>   - Tarred archives.Z of <= 1.1 meg could be produced by each supporter.
>   - These archives could be numbered or sequenced for identification.
>   - A master index (on the root disk) could indicate the contents of each
>     so that just the desired/minimum components might be downloaded.
>   - An install program might be provided, but mostly, just tar -x... from /
>     would be needed.
>   - No source (except for the system) would be provided.  Or perhaps
>     source could be offered as a second series of disks.
>   - An additional series of contrib disks could supply less common items
>     like uucp, or ka9q.
>
>   Obviously the target would be to eliminate the daunting task of 
>   assembling a working Linux system.  But also:
>   - Allow easy/complete distribution, say to BBS's.
>   - Consolidate the user base to eliminate bug reports stemming from
>     a misconfigured system.
>   - Better utilize the efforts of Linux supporters.


        As a person that just installed Linux, I can definitely say that
this is a great idea.  It's a good organization scheme and would really
help distribution.  Anyway, I am not exactly a UNIX wizard, but I'd be
willing to help.


                Cory West






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

From: klode@syslab.csd.uwo.ca (Claude Morin)
Subject: determining which library an executable needs
Date: 25 Jun 92 16:49:01 GMT

In article <1992Jun25.040702.28263@pool.info.sunyit.edu> ujlh@pool.info.sunyit.edu (James Henrickson) writes:
>I've got another question.  (HELP!)  :-)  I have two or three sets of shared
>libs and would like to recompile my programs so they only use the latest.  Is
>there a way to find out which shared libs a program uses?  I'd hate to just
>delete the old libs and find out the hard way.

There are two ways: any easy way and a difficult way.
(isn't that always the case? :-)

Difficult) run 'strings' on the executable you're investigating.  Somewhere
           you should see a string of the form "/lib/libxxxx", where "xxxx" is
           usually "92.04.06" or "c_v_11c".  If you're examining a shared
           executable, this string is most likely the name of the shared
           library.  Do this for *all* your executables (yecchh).  Note that
           you can always resort to 'less' if you don't have 'strings'.

Easy)      to avoid this problem in the future, retrieve the file 'ldd.tar.Z'
           from tsx-11.  If you follow the instructions, you'll be able to
           type "ldd <executable>" to (L)ist the (D)ynamic (D)ependencies of
           the executable.  In other words, 'ldd' prints the names of the
           shared libraries the executable needs.  'ldd' returns nothing
           for static executables.  Note that this won't work for executables
           linked before installation of 'ldd'; the results are unpredicatble.
           This means you can look forward to *another* spate of linking .a's.
           Enjoy :-)
---
Claude Morin
(that's "klode", NOT "clod"...French :-)
London, Ontario, Canada

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

From: mj@dfv.rwth-aachen.de (Martin Junius)
Subject: Re: Error installing Root-0.96
Date: 25 Jun 92 15:53:50 GMT

Rob.Levin@f217.n3802.z1.fidonet.org (Rob Levin) writes:

>Is this available *somewhere* for FidoNet users in the United States? 
>We sure would be grateful.  I have not yet managed to put together a
>system that works reliably and this might help....

It's available for FIDO file request on 2:242/6, but it would be 
much cheaper, if someone with Internet access gets it in the U.S. 
and puts in on a FIDO node there.

Martin
-- 
 _____ _____
|     |___  |   Martin Junius           FIDO:      2:242/6.1
| | | |   | |   Communication Networks  Internet:  mj@dfv.rwth-aachen.de
|_|_|_|@work|   Aachen U of Technology  Phone:     ++49-241-8790220 (voice)

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


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: Linux-Activists-Request@NEWS-DIGESTS.MIT.EDU

You can send mail to the entire list (and comp.os.linux) via:

    Internet: Linux-Activists@NEWS-DIGESTS.MIT.EDU

Linux may be obtained via one of these FTP sites:
    nic.funet.fi				pub/OS/Linux
    tsx-11.mit.edu				pub/linux
    tupac-amaru.informatik.rwth-aachen.de	pub/msdos/replace

The current version of Linux is 0.96a released on May 22, 1992

End of Linux-Activists Digest
******************************
