Subject: Linux-Development Digest #770
From: Digestifier <Linux-Development-Request@senator-bedfellow.MIT.EDU>
To: Linux-Development@senator-bedfellow.MIT.EDU
Reply-To: Linux-Development@senator-bedfellow.MIT.EDU
Date:     Sat, 28 May 94 13:13:05 EDT

Linux-Development Digest #770, Volume #1         Sat, 28 May 94 13:13:05 EDT

Contents:
  Re: Another Huge Security Hole! (Stefan Rodenstein)
  Re: Problems compiling Linux 1.1.15 (Mike Dowling)
  Routing requests and 1.1.15 (Mike Dowling)
  SQ-270-Meg-MicroHOW2 (Matthias Nott)
  Re: How to print faster (Christopher Etz)
  Re: Motif Development (Rob Janssen)
  Anyone speeding up the Mitsumi driver? (Peter Desnoyers)
  Kernel change summary 1.1.15 -> 1.1.16 (Russell Nelson)
  Re: Handwriting recognition in Linux? (yes I have) (Kiriakos X Georgiou)
  Re: The visual bell: patch and RFD (Michael Shields)
  Re: Graphic image copyrights (Ricardo Guimaraes)
  Re: Bug in LOADLIN 1.4 (Steve McMahon)
  Buslogic driver and Linux 1.1.16 (Buffat Marc)
  Re: Why is my Emacs binary so big (2.2M) (Sander van Malssen)
  Re: Does linux work in laptop w/ IBM BIOS? (Steve McMahon)
  Re: Motif question (Bill Foster)
  Re: 8k nfs performance (Amrik Thethi)
  Handwriting recognition in Linux? (Kevin Burtch X8534 Ppppp)
  Re: Zombie problems (Matthias Urlichs)

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

From: stefan@delphi.central.de (Stefan Rodenstein)
Crossposted-To: comp.os.linux.admin
Subject: Re: Another Huge Security Hole!
Date: 25 May 1994 11:36:31 GMT

Uri Blumenthal (uri@watson.ibm.com) wrote:
: Hi,
:       It's Linux-1.1.14. The problem is: when you
:       log in as <whoever>, it still gives you UID
:       [you guessed it :-] 0.  I.e. to become root
:       on Linux-1.1.14,  you just have to login to
:       the box...

Strange... my Linux-1.1.14 works correctly.. the users get the right IDs...
are you sure it's the kernel?
ciao
        Stefan

--

        Stefan Rodenstein     EMAIL: stefan@delphi.central.de
                                     ptsr@asterix.rz.tu-clausthal.de

Entropy isn't what it used to be.

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

From: mike@MooCow.math.nat.tu-bs.de (Mike Dowling)
Subject: Re: Problems compiling Linux 1.1.15
Reply-To: on.dowling@zib-berlin.de
Date: Fri, 27 May 1994 10:52:30 GMT

>>> On 26 May 1994 19:28:48 GMT, lindheim@ccsf.caltech.edu (Jan Lindheim) said:

Jan> Linux 1.1.15 will not compile for Sound Blaster support on my system.  It
Jan> fails on sbpcd.c.  The changes to sched.h in include/linux seems to have
Jan> caused this.

Strangely, I had no problems.

Jan> Also, is there a simple way to bump up the number of processes that can
Jan> run at one time?  As soon as I get up to 78 processes, fork fails.

I had a similar problem while using v1.0, although I suspect that the kernel is
not at fault.  My system was dead for about 30 secs, and when I finally could
log in, the current process no was ~550.   Crond was implicated, and, with the
latest version, I had no further problems.

I also had problems with mound and nfsd.  My process table would overflow.  My
solution was not to allow my filesystem to be mounded over the net.


--
Dr. Michael L. Dowling                    (__)       on.dowling@zib-berlin.de
Abteilung fuer Mathematische Optimierung  (oo)
Institut fuer Angewandte Mathematik        \/-------\
TU Braunschweig                             ||     | \
Pockelsstr. 14                              ||---W||  *
38106 Braunschweig, Germany                 ^^    ^^    Ph.: +49 (531) 391-7553


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

From: mike@MooCow.math.nat.tu-bs.de (Mike Dowling)
Subject: Routing requests and 1.1.15
Reply-To: on.dowling@zib-berlin.de
Date: Fri, 27 May 1994 10:59:49 GMT

When I boot I get 3 messages to the effect

Warning: obsolete routing request made.

It looks to me as if this results from the following 3 lines in my rc.inet1

/sbin/route add localhost
/sbin/route add tubs-net eth0
/sbin/route add default gw 134.169.2.1 1

If so, can anyone please tell me how this is now supposed to be done?
(Everything nevertheless works fine.)

        Mike Dowling
--
Dr. Michael L. Dowling                    (__)       on.dowling@zib-berlin.de
Abteilung fuer Mathematische Optimierung  (oo)
Institut fuer Angewandte Mathematik        \/-------\
TU Braunschweig                             ||     | \
Pockelsstr. 14                              ||---W||  *
38106 Braunschweig, Germany                 ^^    ^^    Ph.: +49 (531) 391-7553


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

From: mnott@gwdu03.gwdg.de (Matthias Nott )
Subject: SQ-270-Meg-MicroHOW2
Date: Fri, 27 May 1994 11:00:57 GMT

This microfaq will explain how you can use a 270-Meg-syquest-drive
with linux.

Your drive (270 Meg sq) will be supported by linux. You have to do the
following:

        If you use a ST02 host adapter, make sure it is jumpered
        to IRQ 5 (there is a Jumper #3 on the card: 1-2 is IRQ-request 3;
        you need 2-3 == IRQ-request 5).

        If you havent got the scsi functions enabled while installing
        linux, you will have to recompile the kernel:

        login as root
        cd /usr/src/linux
        make config
        say yes or no several times and activate scsi (you need ST01/ST02)
        make dep
        make zlilo 2>&1 | tee make.prot

        Then fdisk /dev/sda1. Use x (extended features) to choose the
        correct parameters (1024 Cylinders, 39 Heads, 16 Sectors).
        Use Cylinders 0002 to 1024 for the partition if you want to
        use the whole cartridge for linux (btw ive got the impression
        that anyway the sqprep utility wont work correctly if there is
        a non-dos-partition on the cartridge - so go ahead and use all :)
        If you try to look up the partition table information again, you
        might be thrown out of fdisk. Don't know whether this was a bug.

        Then format the cardridge:

        mkfs -t ext2 -c -v /dev/sda1

        And mount it:

        mount -t ext2 /dev/sda1 /SQ

        will mount it and make it accessible through the directory /SQ.

        Naturally you can as well add a line to each /etc/fstab and /etc/mtab
        in order to get the cardridge mounted at boot time (i do so). BUT:
        If there is a non-linux cardridge (non-ext2), eg. NO CARDRIDGE,
        in the drive, you will get a nice error message, be logged in as
        root on bash# and the drives will be read only. Just correct the
        problem and type exit at the bash#-prompt in order to be logged in
        normally.

        Matthias - mnott@gwdg.de



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

From: cetz@cetz.rhein-main.de (Christopher Etz)
Subject: Re: How to print faster
Date: Fri, 27 May 1994 21:26:42 GMT

David - Foulds (foulds@shell.portal.com) wrote:
: The printer driver lp.c seems to need a longer busy-loop,
: at least on my machine (486-SX33) and with my printer
: (Panasonic KX-P4410 laser with 2.5MB).  The busy loop
: that waits to send the next char in lp_char_interrupt
: only counts to 3 before giving up.  I increases the count
: to 30 or so (and also increased LP_BUFFER_SIZE to 2048)
: and reduced the time to print ghostscript 200K raster
: files (or whatever you call them) i.e. images of postscript
: pages after having been ghostscripted from 4 minutes to 50 seconds.
: This was while crunching the postscript in another process.
: On an empty machine the time per page went from 80 seconds 
: to 40, cpu utilization from 90% to 30% or so.

Thanx for your advice.  It helped me, too.  My machine is a 486-33
with a printer called Canon LBP4plus.  I've modified the busy loop
to be executed "LP_CHAR(minor)" times (LP_CHAR (minor) is not used
anywhere else in the interrupt driver).  This way, I can change the
value using tunelp and got the following results:

With LP_CHAR (minor) = 15, I get the best performance, larger values
don't change the performance, smaller numbers make printing slower.
With LP_CHAR (minor) = 10, I only get about half the throughput.
Best throughput means that printing is about five times faster as with
the original kernel (version 1.0 still).

: Any other improvements possible?

I would be interested, too.

: Cheers, David

: p.s.  all the above is with interrupts enabled.

Christopher
-- 
____________________________________________________________________________
Christopher Etz            Kopernikusstr. 28          D-65929 Frankfurt/Main
cetz@cetz.rhein-main.de    Tel.: +49 69 318091        Telefax: +49 69 318091

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

From: rob@pe1chl.ampr.org (Rob Janssen)
Subject: Re: Motif Development
Reply-To: pe1chl@rabo.nl
Date: Fri, 27 May 1994 21:40:54 GMT

In <1994May27.155118.17581@imec.be> buytaert@imec.be (Steven Buytaert) writes:

>tf49665@delphi.com wrote:
>:    Where can I buy a low cost Motif Development library for Linux? Please
>: reply via email.

>  ARGHHHH, I just edited a response to this beaten to death
>  question, 2 minutes ago (no joke) !!!

>  NO, I don't e-mail answers. People should read the article
>  of 'Emily Postnews Answers your questions on Netiquete.'
>  Watch if fly by news.answers every month.

Keep calm :-)

I just ignore questions with a "please e-mail answers, I don't have time
to read this group".  It seems a better solution that to spend time on
reacting to those.

Rob
-- 
=========================================================================
| Rob Janssen                | AMPRnet:   rob@pe1chl.ampr.org           |
| e-mail: pe1chl@rabo.nl     | AX.25 BBS: PE1CHL@PI8UTR.#UTR.NLD.EU     |
=========================================================================

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

From: peterd@sensorite.dev.cdx.mot.com (Peter Desnoyers)
Subject: Anyone speeding up the Mitsumi driver?
Date: Fri, 27 May 1994 13:42:12 GMT

Is anyone out there hacking on the Mitsumi driver? I've noticed that
it's as slow as a dog (40-50kbyte/s on a double-speed Mitsumi), no
doubt partly due to those 4 10-ms sleeps during each block read...

Anyway, if there's anyone out there working on it, or who knows how to
get specs on the drive, I'd like to hook up with them and take a stab
at it too.

                                Peter Desnoyers

-- 

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

From: nelson@crynwr.crynwr.com (Russell Nelson)
Subject: Kernel change summary 1.1.15 -> 1.1.16
Date: 27 May 1994 20:43:55 GMT

sbpcd drive structure renamed
Fixed a few problems in QIC02 driver mostly related to debugging output.
tty driver changes:
  o properly wakes up processes when it gets a newline.
  o no need to wake anyone else up if we've recieved canonical data.
  o Drop RTS when we drop DTR (and don't generate interrupts either).
  o if line discipline open returns positive value, return it.
  o panic only if it was negative.
3c501 and plip added to modular code.
added Makefile entries for DE650 and 3c589 (PCMCIA Ethernet)
Change PPP memory allocation to GFP_ATOMIC.
Added /proc/net/snmp.
The device driver now sets the type of the packet.
If the packet was not sent to us, don't IP forward it, and don't accept it
        for our reception.
Rebuild the TCP header when forwarding it.
If you remove the IPX device, that will remove IPX routes via it.
Added Hedrick's TCP changes to avoid taking back window allocations.
Missed a test for SUBNETSARELOCAL (should be CONFIG_INET_SNARL)
Scavenge memory for the kernel boot image uncompressor from the already-read
        compressed data
Stripped out some unused code from the uncompressor.
--
-russ <nelson@crynwr.com>      ftp.msen.com:pub/vendor/crynwr/crynwr.wav
Crynwr Software   | Crynwr Software sells packet driver support | ask4 PGP key
11 Grant St.      | +1 315 268 1925 (9201 FAX)    | Quakers do it in the light
Potsdam, NY 13676 | LPF member - ask me about the harm software patents do.

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

From: kxg7283@cs.rit.edu (Kiriakos X Georgiou)
Subject: Re: Handwriting recognition in Linux? (yes I have)
Date: Sat, 28 May 1994 14:34:57 GMT

Yes, I have done work on this, but I guess it's not in any way limited to 
Linux, since it's C source code. (It works on Linux and SunOS so far)

Dont expect GUI or anything, I have just written an engine to recognise 
SEGMENTED alphas and numerals.  I am currently reviewing the segmentors used in
NIST-1 and NIST-2 Conferences, after that I will implement the segmentor.

Then hopefully some of you wonderfull guys on the NET will write a GUI, and we 
have a good project of OCR/ICR.

Blah, I am also working on OMR (Optical mark recognition) but I guess this is
of no "general" use.

Kyriakos


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

From: mjshield@nyx.cs.du.edu (Michael Shields)
Subject: Re: The visual bell: patch and RFD
Date: 28 May 1994 09:46:26 -0600
Reply-To: Michael Shields <mjshield@nyx.cs.du.edu>

In article <9405151111.AA25882@ipvvis.unipv.it>,
Alessandro Rubini <rubini@ipvvis.UNIPV.IT> wrote:
> +     for (i=0;i<2;i++)
> +       for (p=screenstart+1;p<screenend+1;p+=2)
> +                 *p = (*p & 0x88) | ((*p << 4) & 0x70) | ((*p >> 4) & 0x07);

Couldn't this be done more cleanly by inverting the palette?
-- 
Vladimir.

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

From: guimar@pegasus.montclair.edu (Ricardo Guimaraes)
Subject: Re: Graphic image copyrights
Date: 28 May 94 01:26:24 GMT

thomas@melchior.frmug.fr.net (Thomas Quinot) writes:

>Paul Davis (Paul_Davis@mindlink.bc.ca) wrote:
>: I have captured all the card images, converted them to .XPM format and
>: am using them in a game.  What are the implications of releasing source
>: and object for such a game?
>You just violate Crimosoft's copyright (imho), the way you actually violate
>it doesn't matter...

The best solution is to include the script that converts the graphics with
the program and let the user do it. Of course, users would have to already own
Windows.

>-- 
>ThoThoThoThoTho
>        Totolitoto !

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

From: steve.mcmahon@lambada.oit.unc.edu (Steve McMahon)
Subject: Re: Bug in LOADLIN 1.4
Date: 28 May 94 15:52:12 GMT

In article <2s7hca$g53@hpunsca.canada.hp.com> chuckm@canada.hp.com (Chuck Munro) writes:
>
>You can avoid the upshifting of command line parameters by using a
>parameters file for LOADLIN, something like   LOADLIN  @FILE.PAR
>(This is the method I use for the  LINUX.BAT  file on my systems.)
>Note that when you use this option, *all* of your parameters go into
>the file.
>

Yes, it seems the problem is that DOS converts commnd lines in
CONFIG.SYS to upper case, thus rendering the kernel unable to identify
them. The author, Hans Lermen, also indicated that in a letter he sent
to me. The solution is to use a parameter file as he and you
indicated. This seems to work. Thanks to you both.

-Steve


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

From: buffat@europe.mecaflu.ec-lyon.fr (Buffat Marc)
Subject: Buslogic driver and Linux 1.1.16
Date: 28 May 1994 16:12:12 GMT

I am running linux on a 486DX2 with 20 Meg. with a VESA Buslogic BT 445S
(with a 500 Meg Maxtor disk)
and a VESA S3 card. Until yesterday night, I was running a linux kernel 
1.1.12 with the adaptec SCSI driver. I then install the patch until pl16
and I configure the kernel to use the new buslogic driver instead of
the adaptec. I reboot and the kernel found my buslogic SCSI card , and
all seems correct until I try a "du" on the root partition. I begin
to get disk errors, and then I do a mistake. I try to run e2fsck on the
root partition, that corrupt definitively my root file system. It takes
me a day to get my system working, using the adaptec SCSI driver.
Has anybody try a BT 445S with the new SCSI driver on a linux box with more
than 16 Meg. I have read now hat there is some problems with DMA on
a BT 445S with > 16 Meg.
        Any suggestion will be appreciated.  
-- 
=============================================================================
Marc BUFFAT                                 ++++++++++++++++++++++++
Lab. Mecanique des fluides LMFA             |     CNRS URA 263     |
ECL, 36 av. Guy de Collongue                |     ECL Lyon         |
Ecully 69131, FRANCE                        |     UCB Lyon I       |
tel: (33) 72/18/61/61                       ++++++++++++++++++++++++
fax: (33) 78/64/71/45                      
email: buffat@mecaflu.ec-lyon.fr

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

Crossposted-To: comp.os.linux.help
From: svm@kozmix.hacktic.nl (Sander van Malssen)
Subject: Re: Why is my Emacs binary so big (2.2M)
Reply-To: svm@kozmix.hacktic.nl
Date: Sat, 28 May 1994 06:45:23 GMT

Michael MNUK <mmnuk@risc.uni-linz.ac.at> wrote:
> I compiled the Emacs 19.24 using the standard distribution (no extra packages
> dumped). The resulting binary is about 2.2M big which is about twice as much
> as the binary I got with the Slackware distribution. I only changed the
> compilation flag from "-g" to "-O2".
> 
> I observed the same with e.g. fvwm. The ratio there is even more dramatic.
> E.g. FvwmBanner is now about 250K whereas the old (Slackware distrib.) was
> about 30K.
> 
> Am I linking statically (-static option never specified!) or is
> something else wrong?

Are you sure the binary was stripped? (with strip(1) or by linking
with gcc -s ) ?

Sander
-- 
Sander van Malssen
svm@kozmix.hacktic.nl

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

From: steve.mcmahon@lambada.oit.unc.edu (Steve McMahon)
Subject: Re: Does linux work in laptop w/ IBM BIOS?
Date: 28 May 94 16:05:27 GMT

In article <940527165230@lambada> steve.mcmahon@lambada.oit.unc.edu (Steve McMahon) writes:
>
>I went poking arond the kernel, and I think I've found the reason for
>the above behavior. The IBM BIOS uses hard drive type 0 to indicate
>Auto. The kernel looks for the hard drive type to determine the number
>of hard drives attached. It assumes that type 0 means no drive is
>attached (see the code below), otherwise it assumes one or two drives
>are attached.
>
>As you can see, when CMOS_READ returns 0 as the type, NR_HD in the
>above code remains zero, and so linux assumes that no drives are
>attached.  Having identified the problem, how to fix it? I really
>don't know. If we disregard the above block, then how is the kernel to
>know the number of drives? Is there another value in CMOS perhaps that
>tells the number of drives? If so, then that one can be used instead.
>How do DOS, OS/2 and NT do it?
>

Somebody sent me a solution (I lost his name with the letter though,
sorry). It is really very simple. To hd.c add the line:

static void hd_geninit(void)
{
        int drive, i;
        extern struct drive_info drive_info;
        unsigned char *BIOS = (unsigned char *) &drive_info;
        int cmos_disks;

        if (!NR_HD) {
                for (drive=0 ; drive<2 ; drive++) {
                        hd_info[drive].cyl = *(unsigned short *) BIOS;
                        hd_info[drive].head = *(2+BIOS);
                        hd_info[drive].wpcom = *(unsigned short *) (5+BIOS);
                        hd_info[drive].ctl = *(8+BIOS);
                        hd_info[drive].lzone = *(unsigned short *) (12+BIOS);
                        hd_info[drive].sect = *(14+BIOS);
                        BIOS += 16;
/* add this line */     if (hd_info[drive].cyl) NR_HD++;
                }

What this means is simple, if the number of cylenders for that drive
in CMOS is not zero, this means a drive is attached. Bingo, instant
solution to all those IBM machines problems. Don't you think this
ought to be standard in the kernel? Linus?

-Steve


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

From: bfoster@lisbon.nwest.mccaw.com (Bill Foster)
Subject: Re: Motif question
Date: 28 May 1994 00:54:22 GMT


I have recently bought Motif from Sequoia and it works fine. I have also  
heard that Motif from MetroLink works fine as well. There may not have  
been any answers because no one has both to actually compare them.

Hope this helps,
Bill Foster
bill@celtech.com

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

From: at@setanta.demon.co.uk (Amrik Thethi)
Subject: Re: 8k nfs performance
Date: Fri, 27 May 1994 14:16:29 GMT

Talking of 8K NFS, is anybody currently in the process of
implementing caching for the client? This is where the
real speed improvement will come. Is there any technical
reason why it would be difficult, or is it just a matter
of sitting down and hacking it out?

Cheers.
-- 
Amrik Thethi.                   Tel. +223 421 008 Fax. +223 421 024
Setanta Software Ltd.           Internet: at@setanta.demon.co.uk
Cambridge, UK.

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

From: kburtch@pts.mot.com (Kevin Burtch X8534 Ppppp)
Subject: Handwriting recognition in Linux?
Reply-To: kburtch@pts.mot.com
Date: Fri, 27 May 1994 15:19:21 GMT

Has anyone done anything with handwriting recognition in Linux? Does anyone out there
know of any source codes that might be helpful? (I haven't done any programming in a
while, but I might consider tackling this)

I am awaiting an order for a (few) NotePad computer(s), and I was considering using
Linux on it. (I would really like to) Running X would be possible, but questionable,
as it is a 386SL-20 w/8M RAM. Something based on the VGA libraries would probably
be better, but I'll take whatever I can get. :) (X _would_ be nice, but...) Then 
again, I heard something about another GUI (some time back) that was more efficient
than X, but couldn't be run remotely. I would appreciate any opinions from anyone who
uses this. ( whatever it's called, I remember it had an "M" in it :)

I also am considering playing with a (very inexpensive) 6"x9" digitizer I found that
can emulate a Summagraphics MM series or a Mouse Systems mouse. This would be used
with my loaded 486, so X is _no_problem_. :)  Handwriting recognition on this machine
would just be interesting, as it could be used from an X-terminal for a security 
system, or home automation system...  ( yeah, I know I've got some wierd ideas :)

I appreciate any help anyone can give me.   ( please forgive my babbling :)

Thanks!
Kevin




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

From: urlichs@smurf.noris.de (Matthias Urlichs)
Subject: Re: Zombie problems
Date: 28 May 1994 18:20:59 +0200

In comp.os.linux.development, article <2rtsi6$2dg@apollo.west.oic.com>,
  dillon@apollo.west.oic.com (Matthew Dillon) writes:
> 
>     When a child process exits, it *always* becomes a zombie *until* the
>     parent reads the child's exit code with wait(), waitpid(), wait3(), or
>     wait4().
> 
Except when you set the SIGCHLD handler to SIG_IGN.

>     If you want to fork off a child and then not have to worry about dealing
>     with his exit code later on, you can do a double fork:
> 
... or set SIGCHLD to SIG_IGN, or install a SIGCHLD handler which reaps
the signal.

> [ double fork P
>     Neat, eh?
> 
IMNSHO, the only sensible reasons why anybody would ever want to
double-fork, and thus break the existing association between processes
(and cause more load for the system), is (a) you want to exec() a process
which might not like to have stray child processes thrust upon it, or (b)
you're playing around with session IDs and/or process group IDs and/or
controlling terminals.
(Under some Unixes, setting these things only works with double-forking.)

-- 
Be sociable.
Speak to the person next to you in the unemployment line tomorrow.
-- 
Matthias Urlichs        \ XLink-POP N|rnberg  | EMail: urlichs@smurf.noris.de
Schleiermacherstra_e 12  \  Unix+Linux+Mac    | Phone: ...please use email.
90491 N|rnberg (Germany)  \   Consulting+Networking+Programming+etc'ing     42

Click <A HREF="http://smurf.noris.de/~urlichs/finger">here</A>.

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


** 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-Development-Request@NEWS-DIGESTS.MIT.EDU

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

    Internet: Linux-Development@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
    sunsite.unc.edu				pub/Linux

End of Linux-Development Digest
******************************
