Subject: Linux-Development Digest #7
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, 9 Aug 94 08:13:06 EDT

Linux-Development Digest #7, Volume #2            Tue, 9 Aug 94 08:13:06 EDT

Contents:
  Re: SLIP, CSLIP, PPP and modems (Michael J Graven)
  Re: Read Info docs thru WWW...Was: As; The GNU assembler docs {Q} (Robert Lipe)
  Re: kernel walkthrough? (Henry Ware)
  Re: wish: Removable IDE-support (Sverre H. Huseby)
  Re: DIP has fallen and cant get up (Paul Cadach)
  Re: memory segmentation problem (Lawrence Foard)
  Re: IFS (Inherited File System) (Werner Almesberger)
  Re: Repost: *MATROX* Support ? (Dirk Hohndel)
  Re: Linux backup of MSDOS? (Jim Balter)
  Re: memory segmentation problem (Jim Balter)
  FPE signal problem (Help) (Studnitzky Boaz)
  Re: Pentium version of gcc? (Ronald Hindmarsh)
  how to read/write to I/O? help! (Augustyn  Robert)
  Re: No Free Inode on 1GB harddisk!! (Barry Yip kam-wa)
  iso9660 file system post (Riku Saikkonen)
  Re: Help needed compiling SLIPLOGIN (Bart Kindt)
  Re: How to get DOSEMU 5.2 to work on recent kernels (Teemu Kilpivuori)

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

From: mjg@cs.stanford.edu (Michael J Graven)
Subject: Re: SLIP, CSLIP, PPP and modems
Date: 9 Aug 94 03:15:02 GMT
Reply-To: mjg@cs.stanford.edu (Michael J Graven)

bart@dunedin.es.co.nz (Bart Kindt) writes:

> Hmmm. FTP is a protocol which is designed to cooperate with other TCP/IP 
> traffic on the same line.  

Allow me a bit of pedantry: as I recall, FTP /uses/ TCP to implement the
filestream betweeen the client and the server.  A TCP can be written
with or without significant optimizations that result in throughput
enhancement.  The interested reader may wish to consult "Congestion
Avoidance and Control" by Van Jacobson (in the SIGCOMM Proceedings 1988,
pp.314-329) for a more thorough explanation.


--


-Michael                "See that big line in the sky with the blue triangles?
mjg@cs.stanford.edu      That's why it's colder now."  -- Roger Treamstra, WGN

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

From: robertl@arnet.com (Robert Lipe)
Subject: Re: Read Info docs thru WWW...Was: As; The GNU assembler docs {Q}
Date: Tue, 9 Aug 1994 04:26:35 GMT

robertl@arnet.com (Robert Lipe) writes:

(Don't you just hate people that follow up their own postings?)

On my previous message regarding converting info docs to HTML/WWW,
a temporary home has been given to to them at ftp.skypoint.com 
(Thanx, Ken!).  At this moment, they exist in the /incoming 
directory, but may be moved at whim.

There was a shar file containing the scripts that do the conversion,
a .tar.gz that contain my local GNU docs, and a pair of READMEs.

Enjoy.
----
Robert Lipe    Sr. Software Engr.   Arnet Corp.     robertl@arnet.com

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

From: hware@bronze.coil.com (Henry Ware)
Subject: Re: kernel walkthrough?
Date: 8 Aug 1994 16:15:07 -0400

In article <3243b5$kf9@gsusgi2.gsu.edu>,
George P. Magiros <libgpmx@gsusgi2.gsu.edu> wrote:
>On another note, linux's posix
>compatiblity extends to the kernel system calls? - if there is 
>a posix standard down to the system calls.  (funny how VMS can say it's
>posix compatible.)

Posix 1003.1 is only concerned with the functions a program can call.  
It is defined as a C interface.  In Linux & other Unixes, these are often 
system calls; however the standard explicitly doesn't care if they are 
actually system calls, library calls, or whatever.  Just so long as they 
work.    

Cheers,
Henry
-- 
That does it!  I'm putting me back in my kill file!

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

From: sverrehu@ifi.uio.no (Sverre H. Huseby)
Subject: Re: wish: Removable IDE-support
Date: 8 Aug 1994 22:10:54 +0200


In article <31s8hg$19e@bbs.datasoft.com>, mark@datasoft.com (Mark Buckaway) writes:
> In article a8a@menja.ifi.uio.no, sverrehu@ifi.uio.no (Sverre H. Huseby) writes:
> >
> >I have an IDE-disk (SyQuest) with removable cartridges, but it doesn't
> >seem that the kernel supports removable IDEs. Before mounting new
> >cartridges, I have to run a little program that tells Linux to reread
> >the partition table of my /dev/hdb (using ioctl()).
> >
> >I think it would be nice to be able to tell the kernel (or whoever is
> >appropriate) that this IDE is actually removable. A new mountflag, perhaps...
> 
> Actually, the correct method of dealing with removible media (CDROM's, floppies, 
> etc) is to umount the drive, remove the media and insert a new one, and remount
> the drive.

Actually, this is what I do, but as I said, it doesn't work without forcing a
re-read of the partition table before the second mount. Please let me know if the
kernel is supposed to do this right -- maybe I have a hardware-problem (I'm not
one of those kernel-hackers, but I looked at the source without finding anything
about (auto-) re-reading partition tables for IDE-drives. Found it for SCSI, though).


Sverre.


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

Crossposted-To: comp.os.linux.misc,comp.os.linux.help
From:  Paul Cadach <paul@paul.east.alma-ata.su>
Subject: Re: DIP has fallen and cant get up
Date: Sat, 6 Aug 1994 19:15:34 GMT
Reply-To: paul@paul.east.alma-ata.su

Hi Rob,

You forgot to specify your e-mail address in message header - it was
you@somehost.somedomain (Your Name Here), because I post it here.


You wrote:
> My experience with linux and unix networking is very limited so please
> bear with me.

I've use Linux about half year long (My brother got SLS 1.03 distribution
at November 1993 when he was in London). Now I've use 'compound' Linux
(i.e. some from SLS 1.03, some from Slackware 1.2.0). We waiting until
Slackware Pro 2.0 arrived here (in exSU) to buy it.

> I'm trying to connect to a dynamically allocated host via CSLIP and
> dip.   I've finally gotten the dip -t option to work, but I'm not
> exactly sure what to tell it the IP addresses are.  The IP address 
> changes each time I call, so I can't preset any given address.

Try to use 'get $locip remote [<timeout>]' for your IP's address or
'get $rmtip remote [<timeout>]' for IP's address of your provider.
I don't know how to set up DNS (or /etc/hosts file) to work with
dynamic addressing.

Before 'get ...' use 'wait <blah-blah-blah> <timeout>' for text which
you get before IP address.

It may looks like it:
        ....
        # wait for string like 'your address is aaa.bbb.ccc.ddd'
        wait is 15
        # scan your IP address from remote responce
        get $locip remote 15
        ....
or like it:
        ....
        # wait for string like 'my address is aaa.bbb.ccc.ddd'
        wait is 15
        # scan remote IP address from its responce
        get $rmtip remote 15
        ...


I think it must help you.


Sorry my english again,

With best regards,
Paul.
-- 
Paul Cadach        System Programmer | kADA^ pAWEL   sISTEMNYJ PROGRAMMIST
e-mail:   paul@paul.east.alma-ata.su | FIDO:    2:5083/11.39
phone:    +7-(3232)-66-0538          | TEL.:    +7-(3232)-66-0538
=====================================+====================================

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

From: entropy@world.std.com (Lawrence Foard)
Subject: Re: memory segmentation problem
Date: Tue, 9 Aug 1994 05:50:14 GMT

In article <325g9p$dqs@search01.news.aol.com>, EHippo <ehippo@aol.com> wrote:
[stuff deleted]
>  memset(p, size);
[stuff deleted]

Here is one possible problem, memset needs three arguments
memset([pointer],[character to fill with],[size])

-- 
====== Call the skeptic hotline 1=900=666=5555 talk to your own personal . 
\    / skeptic 24 hours/day.     Just say no to victimless crimes.      . .
 \  / High quality Linux application development available.            . . .
  \/ Violence is a lousy substitute for sex and drugs.                . . . .

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

From: almesber@nessie.cs.id.ethz.ch (Werner Almesberger)
Subject: Re: IFS (Inherited File System)
Date: 8 Aug 1994 20:13:01 GMT

In article <320i2j$rar@tadpole.fc.hp.com> laf@sde.hp.com writes:
> Is anyone doing anything with IFS these days?

Not quite, but I plan to do an (experimental) NFS-based IFS later this
year. The kernel-based one has the problems of a) depending on too many
other kernel internals and b) being a bit complex.

Sorry that I don't have any better news at the moment.

- Werner
-- 
   _________________________________________________________________________
  / Werner Almesberger, ETH Zuerich, CH      almesber@nessie.cs.id.ethz.ch /
 /_______________________________________________almesber@bernina.ethz.ch_/

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

From: hohndel@aib.com (Dirk Hohndel)
Subject: Re: Repost: *MATROX* Support ?
Date: Tue, 9 Aug 1994 06:16:18 GMT

Michael Teper (mt3q+@andrew.cmu.edu) wrote:
: I posted a question regarding support for Matrox MGA based video cards
: under Linux and have not received any sort of a response.  I find it
: hard to believe that NOONE has both a Matrox card and a system running
: Linux (especially considering that Matrox makes some of the fastest
: video accelerators on the market).

Might be a reason that this question has been asked (and answered) so
many times, that no one bothered to answer again?

: I would like to know what is the status of Matrox support under linux:
: planned, in works, or implimented ?

: Any sort of a reply would be appreciated.

: I am also curious how one would contact the XFree development team with
: a video card support request.

The contact address would be XFree86@XFree86.Org, but I can give you the
answer right away: No, we don't support Matrox and most likely we never
will, as Matrox is not giving away the programming information without a
non disclosure agreement, which is a no-go for a FreeWare project like
XFree86.

        Dirk
-- 
 _____ _         __  _____            ___  __    ___          _        _   
|_   _| |_  ___  \ \/ / __| _ ___ ___( _ )/ /TM | _ \_ _ ___ (_)___ __| |_ 
  | | | ' \/ -_)  >  <| _| '_/ -_) -_) _ \ '_ \ |  _/ '_/ _ \| / -_) _|  _|
  |_| |_||_\___| /_/\_\_||_| \___\___\___/\___/ |_| |_| \___// \___\__|\__|

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

From: jqb@netcom.com (Jim Balter)
Subject: Re: Linux backup of MSDOS?
Date: Tue, 9 Aug 1994 06:38:30 GMT

In article <325hp2$7hi@sun.cais.com>,
Tom Oehser <toehser@cais2.cais.com> wrote:
>>>> >the idea is to fill the unused space with zeros so that gzip will do a
>>>> >better job.
>
>*Actually*, having started this misbegotten thread, the *idea* was not to 
>make gzip do a better job.  Since what I am doing is putting 540mb of 
>data onto a 525mb tape, the *idea* was to find a much dumber faster 
>compression program than gzip --fast, so as to not waste time compressing 
>it down to 300mb of data.  I would like gzip to do a much worse job.  Is 
>there a compression program that uses much less cpu for less compression, 
>but which will still fit 540 on 525?  -Tom

compress compresses faster than gzip, as do the LZRW algorithms, among others.
Even RLE is likely to get you the small amount of compression you are asking
for, at near-cat speeds.

Perhaps you should be asking over in comp.compression.



-- 
<J Q B>

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

From: jqb@netcom.com (Jim Balter)
Subject: Re: memory segmentation problem
Date: Tue, 9 Aug 1994 06:54:15 GMT

In article <325g9p$dqs@search01.news.aol.com>, EHippo <ehippo@aol.com> wrote:
>    matrix[i] = (double *)getmem(n*sizeof(double *));

That should be sizeof(double).

>The program compiles correctly, and runs "as is" on another system, so I
>thought I'd ask here.  I hope this is the right newsgroup.

Nope, sure isn't.  Take questions like this to comp.lang.c.


-- 
<J Q B>

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

From: s3176015@techst02.technion.ac.il (Studnitzky Boaz )
Subject: FPE signal problem (Help)
Date: Tue, 9 Aug 1994 09:47:11 GMT

Hi, i am wondering if anyone can tell me why the following code
produces an FPE exception after the addition ....

#include <stdio.h>
#include <signal.h>

void fp_ex(int sig)
{
  printf("FPE\n");
}

void init_signals(void)
{
  struct sigaction a;
  a.sa_handler=fp_ex;
  a.sa_mask=0;
  a.sa_flags=SA_RESTART;
  a.sa_restorer=NULL;
  sigaction(SIGFPE,&a,NULL);
}

main()
{
  double a,b,c;
  init_signals();
  a=1;
  b=0;
  printf("1\n");
  c=a/b;
  printf("2\n");
  a=b+1;
  printf("3\n");
  printf("a=%f\n",a);
}

I would appreciate any comments on the above code, or the behavior.

  Boaz Studnitzky
  s3176015@t2.technion.ac.il


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

From: ronald@cs.tu-berlin.de (Ronald Hindmarsh)
Subject: Re: Pentium version of gcc?
Date: 9 Aug 1994 10:35:51 GMT

lee@netspace.students.brown.edu (Lee J. Silverman) writes:


>       There is an "optimized" version of gcc 2.4.0 available at
>ftp.intel.com.  It looks like some employee or tech-nut hacked around
>with it to improve the optimizations for Pentiums, as well as
>everything else in the x86 family. 
[...]
>       Has anyone gotten anything to compile with this version of
>gcc?  If so, has it proven to be faster?  Does anyone out there (intel
>readers?!?!) know if someone's working on the optimizations in 2.6.0?

I complied the raytracer povray with gcc-i2.4.0p. It was not straight
forward to compile because gcc with full optimization settings
dumped core on some files. After compiling these files with lower
optimization, I got a working povray binary. But - the speedup compared to
gcc 2.5.8 -O6 was rather insignificant for most scenes (~10% max.). 

I also compiled the clinpack floating point benchmark. No problems to
compile, and the speed up is around 20%, which is about what Intel
tells you in i2.4.0p's readme files. Here are the results:


Pentium 90, gcc 2.5.8  -O6:
===========================
Rolled Double  Precision  9035 Kflops ; 10 Reps 
Unrolled Double  Precision  8803 Kflops ; 10 Reps 
Rolled Single  Precision 11075 Kflops ; 10 Reps 
Unrolled Single  Precision 10899 Kflops ; 10 Reps 

Pentium 90, gcc i2.4.0p -mpentium  -O4:
=======================================
Rolled Double  Precision 11075 Kflops ; 10 Reps 
Unrolled Double  Precision 10729 Kflops ; 10 Reps 
Rolled Single  Precision 12956 Kflops ; 10 Reps 
Unrolled Single  Precision 11257 Kflops ; 10 Reps 


I also asked the gcc team whether they plan to include Intel's 
patches into newer gcc versions. The answer was, that the
patches were not usable because they are sort of badly programmed
(indeed, this would explain the rather unstable i2.4.0p compiler).
There is some work going on to include own pentium optimizaton,
but this is in a very earley stage. So I do not expect
an official -mpentium gcc flag in the near future.

To summarize, I dont think the Intel gcc hack is useful for us.
Moreover, Intel discourages you to expect g++ or objc to work
in i2.4.0p (I use g++ for my own programs).

The gnu team will probably do the optimization job better ;-)

>--
>Lee Silverman, Brown class of '94, Brown GeoPhysics ScM '95
>Email to: Lee_Silverman@brown.edu
>Phish-Net Archivist: phish-archives@phish.net
>"Nonsense - you only say it's impossible because nobody's ever done it."
-- 
|  Ronald D. Hindmarsh, e-mail: ronald@cs.tu-berlin.de  |
|  TU Berlin, FR 3-9, Franklinstr. 29,  D-10587 Berlin  |
|  Phone: (+49) 30 314-73432   Fax: (+49) 30 314-21103  |

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

From: august1@server.uwindsor.ca (Augustyn  Robert)
Subject: how to read/write to I/O? help!
Date: Tue, 9 Aug 1994 10:36:44 GMT

Hi,
I want to read and write to the i/o port, and do not know where t
to start , any help will be greatly appreciated.
Thanks in advance.
Robert

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

From: g609296@win.or.jp (Barry Yip kam-wa)
Subject: Re: No Free Inode on 1GB harddisk!!
Date: 9 Aug 94 07:08:41 GMT

>In <4167@win.or.jp> g609296@win.or.jp (Barry Yip kam-wa) writes:

>>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

Thanks to many of you who kindly help me. I have to admit that I
overlooked the man page of mke2fs as I was too nervous. Now everything
works fine. It seems that 1 i-node per 2K is sufficient for news and for
ordinary files 1 i-node per 8K is more appropriate. Any comments.

--
Barry Yip
g609296@win.or.jp

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

Subject: iso9660 file system post
From: riku.saikkonen@compart.fi (Riku Saikkonen)
Date: Tue,  9 Aug 94 14:05:00 +0200

>Has anyone else noticed that the iso9660 filesystem is hosed in
>v1.1.41 (I compiled this version of the kernel and it hangs trying to
>mount my cdrom).  My cd is a Sony cdu31a - it worked up to patch level
>36 - I haven't tried the levels between this and 41.  Is there some
>explanation for this??

It works for me (SCSI CD-ROM, Hitachi CDR-1650S)...

>While I'm on the air, is there a version of gdb that works under the
>later 3* kernels??

This seems to work for me too (1.1.41, gdb 4.11, gcc 2.4.5, libc 4.4.4).

-=- Rjs -=- riku.saikkonen@compart.fi - IRC: Rjs
"`"Often is it seen," said Aragorn, "that in dangerous days men hide their
chief treasure." - J.R.R. Tolkien


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

Crossposted-To: comp.os.linux.help
From: bart@dunedin.es.co.nz (Bart Kindt)
Subject: Re: Help needed compiling SLIPLOGIN
Date: Wed, 10 Aug 1994 06:52:17 GMT

In article <31tk8v$iko@uuneo.neosoft.com> lellis@dmccorp.com (John Lellis) writes:
>From: lellis@dmccorp.com (John Lellis)
>Subject: Re: Help needed compiling SLIPLOGIN
>Date: 5 Aug 1994 15:01:51 GMT

>Bart Kindt (bart@dunedin.es.co.nz) wrote:

>: The program often (but not always) does not terminate after receiving a SIGHUP 
>: when a dialer disconnect, with the result the dial-in line is hanging forever.

>We've had a similar experience both using sliplogin and Karel Kubat's callback
>server.  I believe that it isn't related to sliplogin, but rather to the new
>serial drivers, since backing down from kernels 1.1.x to 1.0.9 alleviates the
>problem.  All kernels after 1.1.12 exhibit the behaviour you describe, while
>the 1.0.x kernels do not.  Try going back to the "trailing edge" series of
>kernels and see if it makes a difference.  It did the trick for us.  Hopefully,
>one of these days the serial drivers will be fixed and we can take advantage of
>the "bleeding edge" features, but until our dial-ins recycle reliably, we are
>stuck at kernel 1.0.9, I guess.  :-(

>--

>John Lellis (lellis@dmccorp.com)

I have some additional information on my original post.  The Sliplogin program 
*DOES* respond to the SIGHUP signal when a CD drops.  It then executes the 
scriptfile: /etc/slip.logout , which removes the route and the ifconfig.  (In 
my case it also appends a special log file).  But then, it does *not* execute 
the following commands, which KILL the program!
In the source of sliplogin, this looks like this:
...
system(...)
close(0)
exit(1)

The "system" command executes the scrip file; this works. But the "exit" 
command is not executed. ( I don't know about "close")  The problem is, that 
if I try to compile the Sliplogin program myself, I get warnings about the 
system() command!   Something like: Explicit command...     I managed to 
'solve' this warning by including a few library files which I found in the DIP 
program, which also uses system(). Now the warning is gone... But the system() 
command does not work in my home-compiled version. It *does* work in the 
pre-compiled version..  

What is wrong here?  Any help or sugestions very welcome.  

Bart.
=================================================
Bart Kindt (ZL4FOX/PA2FOX), Dunedin, New Zealand.
=================================================

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

Crossposted-To: comp.os.linux.help
From: teekilpi@utu.fi (Teemu Kilpivuori)
Subject: Re: How to get DOSEMU 5.2 to work on recent kernels
Date: Tue, 9 Aug 1994 07:52:57 GMT

In article <3407@aupair.cs.athabascau.ca> you wrote:
: Hi,

: I built DOSEMU 5.2 under kernel 1.1.19 and all went well. I could
: use it in an xterm and on a VC. Moved to kernel 1.1.35 and even
: though the thing still works under xterm, using it on a VC causes a
: segmentation fault. Recompiling/reinstalling has not helped. Same
: problem with kernel 1.1.40.

: Can anyone help?

Yep, you need to upgrade your dosemu to version 0.53preXX

It can be found at 
tsx-11.mit.edu:/pub/linux/ALPHA/dosemu/private/devel/pre053_11.tgz

Its a hidden directory, but its there.
At the same directory is also a file named
ansi_xterm.pre2.0.tgz ,which really is great for using dosemu under X.
Install that and use xdosemu script,that comes with dosemu 0.53,to use
dosemu under X.


--
   _________________________
 /______________________    | Address:Teemu Kilpivuori
|    _________________ /|   |         YO-Kyld 1B25 
|   | |_______________|_|   |         20540 Turku
|   |/______________________| Voice  :921-2543427
|__________________________/  Email  :teekilpi@utu.fi


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


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