Subject: Linux-Development Digest #975
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:     Tue, 2 Aug 94 08:13:06 EDT

Linux-Development Digest #975, Volume #1          Tue, 2 Aug 94 08:13:06 EDT

Contents:
  Re: Fatal Signal 11  - reproduceable ! (Richard L. Goerwitz)
  Re: Voice Mail cards. (Pete Chown)
  Re: XFree86 Source Code (Dirk Hohndel)
  Need pattern matcher (olav woelfelschneider)
  Re: patching scsi (Joseph H. Julicher)
  Assembly inp() Question (Jon J. Allen)
  Re: DOS EMU under 1.1.34 (Jia-Ning Luo)
  Re: patching scsi (Joseph H. Julicher)
  Re: No Free Inode on 1GB harddisk!! (Thomas Weber)
  Interesting idea for lilo developers (Bill Kress)
  Re: link() system call (Jim Balter)
  Re: lint on linux? (Elan Feingold)
  Re: [hd] Do you recognize this error message? (Hannes Reinecke)
  Re: lint for linux (Jim Balter)
  Re: Does the Adaptec 2940 SCSI driver exists ? (Harald Milz)
  Re: Need pattern matcher (Jim Balter)
  Massive Performance drop with Laptop (Adam DePrince)
  mmap() for MAP_FILE|MAP_SHARED (Mike Black)
  Problems with graphic modes and virtual consoles on a video 7 1024i (Bernd Anhaeupl)
  Problems with graphic modes and virtual consoles on a video 7 1024i (Bernd Anhaeupl)
  Re: Fatal Signal 11 - reproduceable ! (nye)
  Re: Linux backup of MSDOS? (Tom Oehser)
  Re: Voice Mail cards. (Klaus Weidner)

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

From: goer@quads.uchicago.edu (Richard L. Goerwitz)
Subject: Re: Fatal Signal 11  - reproduceable !
Reply-To: goer@midway.uchicago.edu
Date: Sun, 31 Jul 1994 06:01:41 GMT


In one article iwj10@cus.cam.ac.uk (Ian Jackson) scolds:
>
>1. This is answered in the FAQ.
>2. You're posting to completely the wrong group.

In another he continued:

>>Sorry if this is not the appropriate newsgroup.
>
> No, you're clearly not, because you haven't taken any time to find out
> what the right newsgroup is.

This is the sort of griping that gives Linuxers a bad name.  You live
here on Usenet, you put up with a few stupid questions.  At the very
least, Ian, why don't you keep the griping to e-mail instead of pretend-
ing that we all just hang on your every sour word?

-- 

   -Richard L. Goerwitz              goer%midway@uchicago.bitnet
   goer@midway.uchicago.edu          rutgers!oddjob!ellis!goer

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

From: pc@dale.dircon.co.uk (Pete Chown)
Subject: Re: Voice Mail cards.
Date: Fri, 29 Jul 1994 17:59:22 GMT

In article <314hmv$4in@pbaron.connected.com> baron@pbaron.connected.com (Joe Portman) writes:

> As far as I know, no voice mail vendor is going to allow source
> code to be released, or programming details of their cards to be
> made public. Sorry, but they have big dollars invested in this
> stuff, and it is their decision.

Well, let me include an edited quote from the mgetty+sendfax manual:

Features
========

   * behaves like a normal answering machine for human callers

   * remote playback of messages via a DTMF code

   * speech synthesizer support - add the date and time to messages

   * voice conversion utilities - play messages on /dev/audio


What you need
=============

   Obviously, you need a modem that supports the voice commands. I have
tried only two different models so far, I would be interested in
hearing about others.

   The ZyXEL works well, as long as you don't have a ROM older than
6.01. The support for the new (>=6.10) ROMs isn't well tested yet, but
it seems to work now.

   The Dolphin modem works with the following changes ...

===============================================================================

Clearly mgetty is not doing everything that is needed for voice mail,
but it shows that comparatively few changes would allow it to work in
that capacity.  If I was working on a voice mail system, this is where
I would start.

I would not start by signing a non-disclosure agreement, in order to
get access to some company's proprietary interface.  It doesn't mean
you shouldn't, but I would feel I was being taken for a mug.  I mean,
normally if a company needs a driver writing for some card, they pay
to have it done.  We'll do it for free if by doing so we can work
towards the GNU goal of a freely hackable operating system.  If we do
it for free without working towards that, because it is object code
only, then we are just acting as programmers but for no wages.

Also, we really have to push hardware developers to publish their
interface specifications, otherwise Linux will not really be free at
all.  Linux is becoming a big operating system; hardware developers
will want their stuff supported by it.  We have a lot of leverage over
them...
--

==========================================================================
Pete.Chown@dale.dircon.co.uk          "The Pen is mightier than the Quill"
                                      -- anonymous

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

From: hohndel@aib.com (Dirk Hohndel)
Subject: Re: XFree86 Source Code
Date: Tue, 2 Aug 1994 01:27:46 GMT

First M. Last (First_Last@brown.edu) wrote:
:       I am curious about the XFree86 server implementation. Can anyone tell
: me where I can find the source code? Thanks.

Source code can be found as patches against X11R5 on physics.su.oz.au
and on ftp.x.org

        Dirk
-- 
Dirk H. Hohndel <hohndel@aib.com>                 Phone: (703) 430-9247  
AIB Software Corporation                          Fax:   (703) 450-4560 
46030 Manekin Plaza, Suite 160, Dulles, VA 20166

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

From: wosch@rbg.informatik.th-darmstadt.de (olav woelfelschneider)
Subject: Need pattern matcher
Date: 1 Aug 1994 19:09:53 GMT

Hi folks,

for my still (slowly) growing multimedia project, I need a piece of code...

What I'm searching for is a c-function which matches a filename against
a pattern and returns true or false, like:
 int match_file(char *file_name, char *pattern);
which could be used in a form like:
 if (match_file(file_name, "*.voc")) { ...some code... }

I've seen some regexp.c files around in some software packages, but they
are all undocumented and hard to understand.

Maybe there is something around with a small docu describing how2 call the
code?


For the impatient: xplaycd and xmixer are available from most sunsite mirrors
under /pub/Linux/apps/sound/multimedia1.0.tar.gz.
xplaysound, an x-frontend to play sound files of different flavours, is
currently under construction.


Thank you,
--
/======================================\
| Olav "Mac" Woelfelschneider          |
| wosch@rbg.informatik.th-darmstadt.de |
+--------------------------------------+
| I refuse to grow up,                 |
| I don't want to lose my humor...     |
\======================================/

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

From: julichjh@malachi.Rose-Hulman.Edu (Joseph H. Julicher)
Subject: Re: patching scsi
Date: 1 Aug 1994 17:28:34 GMT

In article <CtuqCr.729@pe1chl.ampr.org> rob@pe1chl.ampr.org (Rob Janssen)  
writes:
> It should work as-is.  What is the kernel reporting during boot?
>
The kernel tells me that I have 2 scsi drives sda and sdb
These correspond to the LUN 0 of the two drive drawers. 
The host adapter is an adaptec eisa scsi card.
I forget the number but is has the integrated floppy and is no longer made.
> 
> Rob
> -- 

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

From: jonallen@cs.buffalo.edu (Jon J. Allen)
Subject: Assembly inp() Question
Date: Mon, 1 Aug 1994 17:23:52 GMT


1) I am looking for general information on GNU assembler commands.  
I have books on Microsoft assembly.  GNU (LINUX) assembly seems
to be a bit different in places.  Really I only need port I/O commands.


2) I am porting code from Microsoft Windows code to Linux.  I need
a code sample for inpw()and outpw() 16 bit commands.  Already have
8 bit versions working...

Thanks...




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

From: deer@tem.nctu.edu.tw (Jia-Ning Luo)
Subject: Re: DOS EMU under 1.1.34
Date: 31 Jul 1994 04:55:11 GMT


  when I use Linux 1.1.3x with dosemu 0.53pre6, the key "=" will become

 "9". it's strange enough. When I use my old kernel 1.1.29, it works fine..



----

                           Jia-Ning Luo,   deer@office.ee.ttit.edu.tw

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

From: julichjh@malachi.Rose-Hulman.Edu (Joseph H. Julicher)
Subject: Re: patching scsi
Date: 2 Aug 1994 04:04:56 GMT

I found the problem with not recognizing the LUNs.

The scsi driver stops looking for LUNs if the version # of the drive is 0
and the emulex bridgeboards return revision 0.
So I will patch the version 0 check to check for emulex bridgeboards.
Joe Julicher

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

From: tom@pandemonium.saar.de (Thomas Weber)
Subject: Re: No Free Inode on 1GB harddisk!!
Date: Mon, 1 Aug 1994 20:41:49 GMT

In article <4167@win.or.jp>, Barry Yip kam-wa <g609296@win.or.jp> wrote:
>We have a 1GB scsi harddisk used for storing a full newsfeed. After
>getting about one week of news, there is no more free inode space left
>and I think the only solution is to get another bigger one harddisk even
>we still have enough harddisk space. Is there any way to increase the
>available inodes in this situation. Also, will it be better if the ext2
>filesystem can have a bigger limit on the max. no. of inodes allocated.
>Any suggestion and comments welcome.

man mke2fs.
Have a look at the -i option.

Too lazy to read the man page?? Well, get a bigger disk ;-)))

tschau,
  Thomas

-- 
Halt Wanderer, halt, geh langsam hier;  Wie du jetzt bist, so waren wir;
Wie wir jetzt sind musst du einst sein, Zu folgen uns, ruesst du dich fein!
                                                                 -atrocity-

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

From: kress@kentrox.com (Bill Kress)
Subject: Interesting idea for lilo developers
Date: Wed, 27 Jul 1994 00:23:59 GMT

lilo is a great system. The problem is obvious, there is
no input that you can set when you turn your computer on
so that you can go get a drink while your system boots
to the operating system/configuration of your choice.

I notice that you guys tried to do this with the scroll
lock/caps lock keys but, of course, just before going
to the hard disk my ROM resets these keys.
(By the way, since caps/scroll/num works so well for this,
why not have each key boot a differnt system?)

Anyway, back to my point, there IS a piece of hardware
that can retain its state through the whole boot sequence.
The normally useless joystick can often be de-springed (a
switch on many joysticks) and left sitting in one of the
four corners. This way, up to four (or more) entries in
lilo could be chosen before you even hit your computers
power switch.

I know this works, I used to use it to start differnt
dos programs from autoexec.bat.

This way you could boot your OS/2, Linux 25x80,
Linux 50x80, or Linux init level 6 (x-windows)
just by repositioning your joystick.

Well, it would be NICE if you could pass info
like the run level to the Linux kernel, but that's
another problem to battle later, you could always
read the joystick position from inittab and act on
it then...

If this is the wrong newsgroup for this, would somebody
please forward it to the lilo developers?  I assume that
they are the same people that do linux development, but
I may be completly wrong.

Thanks for listening,
   Bill Kress.

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

From: jqb@netcom.com (Jim Balter)
Subject: Re: link() system call
Date: Tue, 2 Aug 1994 07:07:19 GMT

In article <31gblr$dbo@news.u.washington.edu>,
Tim Smith <tzs@u.washington.edu> wrote:
>Michael Edward Chastain <mec@shell.portal.com> wrote:
>>I agree.  A long time ago I read an article (I think written by Dennis
>>Ritchie), saying that some very old (pre-V7) Unix allowed links to
>>directories, but use of this feature lead to unmanageable chaos.  I am
>
>I don't know what everyone is smoking, but I wish I had some!  Unix has
>allowed hard links to directories much later than V7.  System V Release 2
>allowed it.  System V Release 3 allowed it, I seem to recall.  Ultrix
>seems to allow it (according to the man page for link--I don't have
>the root password on any Ultrix system so I can't actually check).

In the same sense that root is allowed to unlink /usr without emptying it first.
R&T moved many functions out of the kernel by making root a special agent.
root was given the ability to link directories solely in order to implement
directory operations such as mv/mkdir/rmdir.

To quote "The UNIX Time-Sharing System" by R&T:

"The directory structure is constrained to have the form of a rooted tree.
Except for the special entries ``.'' and ``..'', each directory must appear as
an entry in exactly one other directory, which is its parent.  The reason for
this is to simplify the writing of programs that visit subtrees of the directory
structure, and more important, to avoid the separation of portions of the
hierarchy.  If arbitrary links to directories were permitted, it would be quite
difficult to detect when the last connection from the root to a directory was
severed."

Actually, this justification is confused and backwards, since you aren't
allowed to remove a directory unless it is empty.  The real problem would be
that, once you created a link to a directory, you would not be able to delete
it without emptying the directory, and if you created a loop, you would never
be able to delete *any* directory in the loop.  Unless you were the superuser
and knew what you were doing.

-- 
<J Q B>

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

From: feingold@avette.zko.dec.com (Elan Feingold)
Subject: Re: lint on linux?
Date: 1 Aug 1994 17:30:46 GMT
Reply-To: feingold@avette.zko.dec.com (Elan Feingold)



> <sigh> This question comes up every two weeks. No there is no 
> lint for Linux. 'gcc -Wall' should do everything you need though.

I beg to differ.  Check out LCLint, from MIT.  

..
In addition, to run LCLint you will also need,

        o lsl --- Larch Shared Language checker

If you do not already have lsl, lsl needs to be built first.
Instructions for building lsl are included at the end of this file.

Here is an example command sequence for extracting the source package:

   % ftp larch.lcs.mit.edu
   Name (larch.lcs.mit.edu:evs): anonymous
   ftp> cd pub/Larch
   ftp> bin
   ftp> get lclint1.3.src.tar.Z
   ftp> quit
..

I haven't had good luck with it, but it's out there :)

elan

--
===============================================================================
|| Elan Feingold (Cornell '94)    || "Two of the most famous products of     ||
|| Software Engineer II           ||  Berkeley are LSD and Unix. I don't     ||
|| Digital Equipment Corporation  ||  think that is a coincidence."          ||
|| Work: 603.881.1115             ||                       - Anonymous       ||
===============================================================================

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

From: hare@zarquon.mathi.uni-heidelberg.de (Hannes Reinecke)
Subject: Re: [hd] Do you recognize this error message?
Date: 02 Aug 1994 08:52:24 GMT


Yes, I've seen it several times, back in the good ol' days of 0.98 .
I got it myself while playing with an rather old 386 with MFM-HD.
Two possible explanations:
- Your HD is to slow. You could try to fix this in changing the line

#define HD_DELAY        0

( linux/drivers/block/hd.c line 31 )
in

#define HD_DELAY        100 

(or something like this, probably more)
This gives your HD some time to relax between the different request
(And also slows your machine down :-( ).

- Your HD is sort of defect. Try to make a "e2fsck -cvrs /dev/hdXX"
  (Not sure about the switches; take a look in the man-pages and be
  sure to take the switch for 'check bad blocks').
  Naturally assuming you've got an ext2fs.

Hope this helps,

Hannes
=======
Hannes Reinecke                      |
<hare@vogon.mathi.uni-heidelberg.de> |  XVII.: WHAT ?
                                     |  
PGP fingerprint available            |          T.Pratchett: Small Gods
see  'finger' for details            |          

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

From: jqb@netcom.com (Jim Balter)
Subject: Re: lint for linux
Date: Tue, 2 Aug 1994 07:14:18 GMT

In article <31i600$5dj@senator-bedfellow.mit.edu>,
Derek Jones <D.G.Jones@scuna.dircon.co.uk> wrote:
>Sigh, again,
>
>I have responded privately to some of the people who have
>said "gcc does everything" over the last few months
>but now feel it's time for the list.
>
>gcc -Wall -pedantic does *not* catch some of the (even
>more trivial) errors that a decent lint will catch. *Esp.*
>if those errors are *calling problems between* source files.

Using prototypes, plus gcc -Wstrict-prototypes -Wmissing-prototypes to enforce
their use, can go a long way toward solving that particular problem.

>I did some tests on this about 18 months ago and showed the
>problem with a very simple set of source files. I think at
>the time the discussion was on the turbo-c list or somesuch.
>
>I don't have access to a lint currently, (since I'm running
>Linux 8-), so I can't immediately repeat it for everyone
>but I may be able to in a few weeks.

You (and others) may be interested in LCLint, recently announced on
comp.compilers.  I know nothing other than what is contained in that
announcement, but it sounds promising.
-- 
<J Q B>

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

From: hm@seneca.ix.de (Harald Milz)
Subject: Re: Does the Adaptec 2940 SCSI driver exists ?
Date: Tue, 2 Aug 1994 07:46:48 GMT
Reply-To: hm@seneca.ix.de

Rob Janssen (rob@pe1chl.ampr.org) wrote:
: > In <31dphf$pri@infoserv.rug.ac.be> baekelan@intec.intec.rug.ac.be (Bart Baekelandt) writes:

: > >Could anyone answer the following question:

: > Probably yes.  But please note that it is a constant load on the developers
: > and other people in the newsgroup to ask this question over and over again.

: > It is better to look in the projects-FAQ and other FAQ documents first
: > to get your question answered.  You will probably learn a lot more at
: > the same time.

: > You find these documents in the comp.os.linux.announce newsgroup

If the respective article has already expired at your site, you can find it
in the normal locations for Linux FAQs. Plus, the latest version is always
available on ftp.ix.de:/pub/Linux/FAQ/Projects-FAQ.

-- 
Harald Milz                             office: hm@ix.de
iX Multiuser Multitasking Magazine      home:   hm@seneca.ix.de
Opinions are mine, not my employer's -- the answer is Forty-two
PS: I apologize for my bad return address - I hope this will be fixed soon.

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

From: jqb@netcom.com (Jim Balter)
Subject: Re: Need pattern matcher
Date: Tue, 2 Aug 1994 07:32:52 GMT

In article <31jha1$vrf@rs18.hrz.th-darmstadt.de>,
olav woelfelschneider <wosch@rbg.informatik.th-darmstadt.de> wrote:
>Hi folks,
>
>for my still (slowly) growing multimedia project, I need a piece of code...
>
>What I'm searching for is a c-function which matches a filename against
>a pattern and returns true or false, like:
> int match_file(char *file_name, char *pattern);
>which could be used in a form like:
> if (match_file(file_name, "*.voc")) { ...some code... }
>
>I've seen some regexp.c files around in some software packages, but they
>are all undocumented and hard to understand.
>
>Maybe there is something around with a small docu describing how2 call the
>code?

Search the sources of your favorite shell for "glob".  tcsh, for instance,
has a glob.c and a glob.3 documenting it.
-- 
<J Q B>

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

From: axd0822@hertz.njit.edu (Adam DePrince)
Subject: Massive Performance drop with Laptop
Date: Tue, 2 Aug 1994 08:29:51 GMT

I have a Compudyne DX/2-66 laptop.  Basicly a noname brand.
After running Linux for the first several minutes it switches itself
to a slower speed and stays there. 

Like other machines with AMI bios is supports the ALT+CTRL +/- to change
CPU speed.  In this respect it appears to be normal.

Part of the problem is due (I belive) to the "auto-detection" circuitry 
in most laptops.  This circuitry provides for raising or lower the speed
as needed to conserve the batteries.

I am considering writing code, perhaps a patch to sched.c, to adjust 
the clock rate to correspond to the amount of time spent in the idle process.
(I figure a beter indicator than any of CPU utilization).

The informtion I need is ...

1.      How to set and unset the "TURBO" mode on an AMI bios machine.
2.      How to manipulate the power saving features on such a machine
3.      Those working on similar projects whom I may collaborate with.
4.      If anyone else has had this problem and how they dealth with it.


-- 
===============================================================
Tic.                            |       Adam DePrince
                                |       axd0822@hertz.njit.edu

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

From: mike@rayleigh.aftac.gov (Mike Black)
Subject: mmap() for MAP_FILE|MAP_SHARED
Date: Tue, 2 Aug 1994 08:28:55 GMT

I'm involved in a project which requires file-based shared mapping
and was wondering who (if anyone) is working on this.  I'm interested
in helping out to get it running.  Specifically (as of patch 37 to 1.1)
specifying MAP_FILE|MAP_SHARED is an illegal operation whereas one
used to be able to declare MAP_SHARED.  The old results were that
the changes to shared memory were shared between processes but were
still not mapped to the file.
-- 
mike@rayleigh.aftac.gov

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

From: anhaeupl@late.e-technik.uni-erlangen.de (Bernd Anhaeupl)
Subject: Problems with graphic modes and virtual consoles on a video 7 1024i
Date: Tue, 2 Aug 1994 09:02:57 GMT

I am just implementing a (banked) monchrom driver for my old video seven 1024i 
graphics card for XFree86 2.1.1. Now everything seems to work with one 
exception. The symptoms are:

        1.  When I start up X in 1024x768 interlaced mode the screen
          is white. (The video timing seems to be correct, since my 
          Monitor syncs on the signal, I get the correct (black) 
          borders,...)

        2.  After switching to another virtual console using 
          <Alt-Ctrl-F1>  and back to X11 with <Alt-Ctrl-F7> I get
          the correct image on my screen.

        3.  All other methods to establish 1024x768 interlaced mode
          (switching resolution with <Alt-Ctrl-Gray+/->, calling
           'xrefresh -display :0' from another virtual console) result 
           in a white screen until I use method 2.

            Also calling xrefresh -display :0 from a real terminal
           connected through a serial port only 'refreshes' the 
           incorrect white screen, if it was incorrect before, and
           the correct screen, if it was correct before.  

        4.  The 640x480 modes with 1024x786 virtual resolution work
           correctly.

        5.  I have verified (by reading them back) that the additional 
           SVGA registers I am using in my driver always get the same 
           values when switching to 1024x768 interlaced mode, whether 
           I use method 2 (and get the correct image) or method 1/3 
           (and get the incorrect white screen).

Any ideas? I am especially interested in possible differences between 
switching virtual consoles through keystrokes (Method 2) and through
a programm (methods 1,3)

I am still using the Linux kernel version 1.0.7 with XFree86 2.1.1
on an old 80386 (C&T chipset (`NEAT`), 33Mhz, 8 MByte RAM).

Please respond via e-mail (anhaeupl@late.e-technik.uni-erlangen.de).
I am also interested in a working driver for the 1024i.

Bernd.

--
Bernd Anhaeupl                  Tel.:  +49 9131 857787
LATE - Uni Erlangen                     
Cauerstr. 7                     Email: anhaeupl@late.e-technik.uni-erlangen.de
91058 Erlangen

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

From: anhaeupl@late.e-technik.uni-erlangen.de (Bernd Anhaeupl)
Subject: Problems with graphic modes and virtual consoles on a video 7 1024i
Date: Tue, 2 Aug 1994 10:01:31 GMT

I am just implementing a (banked) monchrom driver for my old video seven 1024i 
graphics card for XFree86 2.1.1. Now everything seems to work with one 
exception. The symptoms are:

        1.  When I start up X in 1024x768 interlaced mode the screen
          is white. (The video timing seems to be correct, since my 
          Monitor syncs on the signal, I get the correct (black) 
          borders,...)

        2.  After switching to another virtual console using 
          <Alt-Ctrl-F1>  and back to X11 with <Alt-Ctrl-F7> I get
          the correct image on my screen.

        3.  All other methods to establish 1024x768 interlaced mode
          (switching resolution with <Alt-Ctrl-Gray+/->, calling
           'xrefresh -display :0' from another virtual console) result 
           in a white screen until I use method 2.

            Also calling xrefresh -display :0 from a real terminal
           connected through a serial port only 'refreshes' the 
           incorrect white screen, if it was incorrect before, and
           the correct screen, if it was correct before.  

        4.  The 640x480 modes with 1024x786 virtual resolution work
           correctly.

        5.  I have verified (by reading them back) that the additional 
           SVGA registers I am using in my driver always get the same 
           values when switching to 1024x768 interlaced mode, whether 
           I use method 2 (and get the correct image) or method 1/3 
           (and get the incorrect white screen).

Any ideas? I am especially interested in possible differences between 
switching virtual consoles through keystrokes (Method 2) and through
a programm (methods 1,3)

I am still using the Linux kernel version 1.0.7 with XFree86 2.1.1
on an old 80386 (C&T chipset (`NEAT`), 33Mhz, 8 MByte RAM).

Please respond via e-mail (anhaeupl@late.e-technik.uni-erlangen.de).
I am also interested in a working driver for the 1024i.

Bernd.

--
Bernd Anhaeupl                  Tel.:  +49 9131 857787
LATE - Uni Erlangen                     
Cauerstr. 7                     Email: anhaeupl@late.e-technik.uni-erlangen.de
91058 Erlangen

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

From: nyet@linuxftp.caltech.edu (nye)
Subject: Re: Fatal Signal 11 - reproduceable !
Date: 31 Jul 1994 08:53:35 GMT

dcflood@u.washington.edu (David Flood) writes:

>wosch@rbg.informatik.th-darmstadt.de (olav woelfelschneider) writes:

>>So you can not be absolutely sure wether this is a compiler bug or a linux bug.

>Or a hardware bug.  I had a bad swap partition that neither re-lowleveling
>(MFM) or mkswap -c would find the bug but by spliting the partition in 1/2
>eliminated the consistant signal 11's when compiling the kernel.  It also
>eliminated some serious disk thrashing.

Or a harder hardware bug. I had a bad SIMM that checked ok on my machine's POST
but gcc was barfing with sig 11's. Then put the simm in another machine
and it failed the POST. I got another good simm and the sig 11's went away
completely.

On another note, it looks like AMD dx2/66's don't like ghostscript - anybody
else see this?

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

From: toehser@cais.cais.com (Tom Oehser)
Subject: Re: Linux backup of MSDOS?
Date: 2 Aug 1994 10:42:15 GMT

>Tom Oehser <toehser@cais2.cais.com> wrote:

>>I am trying "dd if=/dev/hda | gzip --fast | dd of=/dev/nrmt0"

>What is the advantage of this over gzip </dev/hda >/dev/nrmt0 ?

Is there one?  I am just ignorant and *nix is my newest operating system.
I knew that some programs didn't like the raw devices, so I used dd for them.

>>(Why compress?  420mb/hda + 120mb/hdb > 525mb/mt0.  Besides, why not?)  

>Shudder.  You have a lot more faith in the data reliability of tape than I do.
>You might try looking into afio.

I think it all depends on the tape.  My Wangtek 5525ES with a new 3M 
cartridge... well, it's rated at 1 bit per 27,000 full 525mb tapes... I 
can live with that.  I _would_ like to find something that compresses 
very fast with only a little size reduction, since I'm fitting 540 on 525.  
Also, I _am_ doing tar backups of the filesystems: the binary image is 
really only to get back the boot records, partition tables, and 
functioning operating systems, all in one shot, to restore to.  I won't 
have to do it again unless I repartition, and it won't kill me if I lose 
it.  It would just make life a lot easier as far as reinstalling linux, 
OS2, stacker, lilo, bootmanager, swap partitions, etcetera.  I will look 
into "afio"- what is it? -Tom



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

From: klaus@snarc.greenie.muc.de (Klaus Weidner)
Subject: Re: Voice Mail cards.
Date: Mon, 1 Aug 1994 19:53:26 GMT

pc@dale.dircon.co.uk (Pete Chown) writes:
>Well, let me include an edited quote from the mgetty+sendfax manual:
[...]
>Clearly mgetty is not doing everything that is needed for voice mail,
>but it shows that comparatively few changes would allow it to work in
>that capacity.  If I was working on a voice mail system, this is where
>I would start.

Some people are intending to use vgetty as a voice mailbox with
faxback capability. I am planning to add the missing pieces to
zplay to make it possible to do this using shell scripts.

As far as I know, real voice mail cards have more capabilities than
voice capable modems. But for many purposes, the modem will probably
be sufficient.

Klaus
-- 
\ mail: klaus@muc.de, irc: tengu, http://www.muc.de/~klaus/
\ .signature error -- quote dumped

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


** 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
******************************
