Subject: Linux-Development Digest #22
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, 13 Aug 94 15:13:05 EDT

Linux-Development Digest #22, Volume #2          Sat, 13 Aug 94 15:13:05 EDT

Contents:
  Re: Pentium version of gcc? (Laurent Chemla)
  Re: How to get DOSEMU 5.2 to work on recent kernels (Edward Doolittle)
  Re: term and news/mail (Patrick Reijnen)
  Re: 1.1.38 broken from PCI/Pentium (Eberhard Moenkeberg)
  Re: Registrar for major device #s? (Eberhard Moenkeberg)
  Mcd Bug ? (Liberty Line Sas)
  Re: Suggestion to LILO folks (Werner Almesberger)
  Adaptec AHA 2940 (Alexander Nareyek)
  Re: Suggestion to LILO folks (Rob Janssen)
  Re: Does anyone use SONY CDU-535 CD-ROM anymore? (Kris Ong)
  SCSI & PCMCIA driver: kernel rescan request (Roger C. Pao)
  SLIP routing problems (Brian Watts)
  X11R6 build on linux: won't; help! (Bruce Parkin)
  Re: Proposal to write a SAMBA fs client for the kernel (Volker Lendecke)
  Re: BOCA ioAT66 serial card?? (Rob Janssen)
  Is there a long-term plan for kernel development? (Corey Ladas)

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

From: laurent@brasil.frmug.fr.net (Laurent Chemla)
Subject: Re: Pentium version of gcc?
Date: 10 Aug 1994 15:48:57 GMT

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

:       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. 

--
Please forgive my poor english.

I made this old 2.4.0 up to 2.5.8, and got quite a big speed improvement, at
least on the dhrystone standard test (I'm backuping on a slow scsi
magneto_optical drive right now, so the results would'nt be useful, but if
I remember well, i obtain around 87000 dhrytones with the stock 2.5.8 and
around 95000 with my pentium 2.5.8 version).
That was not such a big deal to bring the Intel 2.4.0 patched up to 2.5.8,
but I had to make a lot of work 'by hand' because the patch utility was lost
in proceeding 2.4.5->2.5.0 patch. I made some other hand made modifications
when patch reported rejects, but not so much, and lastly a big one in
changing back some obscure inline assembling option that was preventing me
from compiling Linux.
I also uploaded this GCC version on tsx-11 one month ago or so, but it looks
like T'Tso didn't received it well, because it never appeared anywhere. I
dunno why. It's available on my bbs, if any French reader cares about :-)
As I only have a dialup Internet acces, I wonder if I have to put them on
tsx-11 again or if someone can find them out of the incoming directory...

Hope this helps.
--
Laurent Chemla : chemla@cnam.cnam.fr or laurent@brasil.frmug.fr.net
Brasil BBS  - +33 1 44 67 08 44 -  Atari France developpers support

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

Crossposted-To: comp.os.linux.help
From: dolittle@math.toronto.edu (Edward Doolittle)
Subject: Re: How to get DOSEMU 5.2 to work on recent kernels
Date: Wed, 10 Aug 94 19:41:42 GMT

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

>> 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.

In article <Cu9CKA.AJ@utu.fi> teekilpi@utu.fi (Teemu Kilpivuori) writes:

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

There is another way: comment out all video lines in dosemu.config.
Of course, this isn't ideal. :)

Ed
==================================================================
"There are no differences but          | dolittle@math.toronto.edu
 differences of degree between degrees | dolittle@math.utoronto.ca
 of difference and no difference."     |
 --William James                       | Ed Doolittle

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

From: patrickr@cs.kun.nl (Patrick Reijnen)
Subject: Re: term and news/mail
Date: Sat, 13 Aug 1994 14:52:51 GMT

In <lsmolyanCuGHs7.Kw2@netcom.com> lsmolyan@netcom.com (Larry Smolyansky) writes:

>I need to get some help from anyone who has been had any experience with 
>term and Linux.

>1.  How can I use pine locally to read mail

Hmm, I don't know pine. But i not two programs which will handle this:
term.mailerd+smail.tar.gz
BCRMAILHANDLER

Both are available from sunsite.unc.edu. /pub/Linux/apps/comm/termstuff or /pub/Linux/apps/comm/term/term/extra

>2.  How can I use tin locally to read my news groups

On thing which is needed is to redirect ort 119 to your remote newsserver.
Tredir 119 <remote_serve>:119. This you have to do being root. Otherwise you have to redefine ports in your /etc/services, as ports upto 1024 are priviledged to superuser use only.
Furthermore, you need the file /etc/nntpserver. This file should contain one line stating the full name of your remote newsserver.
Last , get yourself the tin pathes in one of the directories mentioned above. Recompile tin with these patches patched in and there you go.

>So far I have ftp, gopher, etc. working but I can't seem to nail those 
>two.  If you have any suggestions, please let me know.

>Thanks

>Larry
>-- 
>                                             lsmolyan@netcom.com
>                                             Larry Smolyansky

Hope this helps,

PAtrick Reijnen

--
************************* Patrick Reijnen *************************
* Department of Computer Science, Catholic University of Nijmegen * 
* Email:  patrickr@{sci,cs}.kun.nl                                *
* WWW:    http://{atlas,zeus}.cs.kun.nl:4000/homepage.html        *

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

Date: Sun, 07 Aug 1994 11:36:33 +0200
From: Eberhard_Moenkeberg@p27.rollo.central.de (Eberhard Moenkeberg)
Subject: Re: 1.1.38 broken from PCI/Pentium


Hello James Norton and all others,

on 07.08.94 James Norton wrote to All in USENET.COMP.OS.LINUX.DEVELOPMENT:

JN> I have an 60Mhz Pentium with an Intel motherboard.  I have not been able
JN> to get patches 38, 39, or 40 to boot.  I have tried compiling the kernel
JN> using gcc 2.5.8 and 2.6.0.  They both compile the kernel fine.  But when
JN> I try to boot the kernel, I get a message like "Booting the kernel now."
JN> Then nothing.

I see the same here - 60 MHz Pentium, gcc 2.5.8.

Last visible action is: the cursor goes one line up (to the first column
of the line "uncompressing kernel", if I remember right).

Greetings ... Eberhard


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

Date: Wed, 10 Aug 1994 00:14:45 +0200
From: Eberhard_Moenkeberg@p27.rollo.central.de (Eberhard Moenkeberg)
Subject: Re: Registrar for major device #s?


Hello David Hinds and all others,

on 09.08.94 David Hinds wrote to All in USENET.COMP.OS.LINUX.DEVELOPMENT:

DH> Does anyone know who I should talk to about "registering" a major
DH> device # for a Linux device driver?

Linus has to decide about inclusion into <linux/major.h>, and Nick
Holloway should know about it for his MAKEDEV utility.

If you just need a free number, but your driver is of less interest
to others, look into major.h.

Greetings ... Eberhard


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

From: Liberty Line Sas <MC5171@mclink.it>
Subject: Mcd Bug ?
Date: Sat, 13 Aug 94 15:18:46 CEST


 
Hi all, I think I've found a problem with the mitsumi cd-rom driver
included 
whith the standard kernel distribution ( 1.1.44 )

if I do 

        dd if=/dev/mcd of=/dev/null 

the system completely hangs after a minute or so. 

I tried two different systems with different drives and controllers
(single and double speed) so i think the problem is not related to 
the drive model.

Also , the old mitsumi driver (kernel < 1.1.25 ) works well. 

Any help would be appreciated, 
Thanks.

                                Daniele Paoni
                                mc5171@mclink.it

 

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

From: almesber@nessie.cs.id.ethz.ch (Werner Almesberger)
Subject: Re: Suggestion to LILO folks
Date: 13 Aug 1994 16:42:36 GMT

In article <1994Aug13.061749.13223@iowegia.dsm.ia.us> kjhoule@iowegia.dsm.ia.us (Kevin Houle) writes:
> Well, this install, I put LILO on. Little did I realize it
> would fry the video card on the first boot.

LILO only uses the BIOS functions to display characters (14), to get the
current display mode (0xf) and to set the display mode (0). So a card
that is destroyed by this is not very likely to survive DOS either ...

The video mode selection in the kernel "prologue" (boot/setup.S) may
possibly do more aggressive operations. If that hardware failure was
preceded by a change in the kernel version, you may want to post its
version number and the video mode it was trying to set. (Get this
either from your /etc/lilo.conf or with rdev -v)

Maybe some others who were unfortunate enough to buy a Diamond Stealth
Pro may want to try to reproduce the problem. (It'd surely be a nice
way to get rid of the card ;-)

Because any obvious problem in that area would probably have yielded a
bunch of complaints already, I tend to believe your card broke for some
unrelated reason. Warranty should cover that.

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

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

From: theboss@cs.tu-berlin.de (Alexander Nareyek)
Subject: Adaptec AHA 2940
Date: 13 Aug 1994 16:54:31 GMT

Hi,

I'm looking for a driver for the Adaptec AHA 2940.

If any exists, where can I find it?
If not, is it under construction and when will it
be available?

Thanks a lot, Alex.


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

From: rob@pe1chl.ampr.org (Rob Janssen)
Subject: Re: Suggestion to LILO folks
Reply-To: pe1chl@rabo.nl
Date: Sat, 13 Aug 1994 14:11:15 GMT

In <1994Aug13.061749.13223@iowegia.dsm.ia.us> kjhoule@iowegia.dsm.ia.us (Kevin Houle) writes:

>Well, this install, I put LILO on. Little did I realize it
>would fry the video card on the first boot. It sure would be
>nice to have some kind of obvious message warning of possible
>system damage when using LILO with certain hardware. Sure, I
>can't use the Diamond with Linux, but I can with other OS's,
>and that's a couple hundred bucks down the drain. 

LILO won't fry your video card.  If yours got fried, it was not because
of using LILO.  It must just have been broken.

(trying to run X without reading the guidelines on how to configure it
is another thing.  even that is not very likely to fry your videocard,
but it is not impossible with some cards)

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

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

From: krismon@hh.sbay.org (Kris Ong)
Subject: Re: Does anyone use SONY CDU-535 CD-ROM anymore?
Date: 12 Aug 1994 22:04:57 -0700

Hans de Hartog (dehartog@ccult1.comcons.nl) wrote:
: jrc01@maroon.tc.umn.edu (Jeffrey Comstock) writes:

: >In article <31v4q6$50f@solaris.cc.vt.edu>,
: >David La Croix <dlacroix@guilder.bevd.blacksburg.va.us> wrote:
: >>I just got a SONY CDU 535 type CD-ROM drive, and I'd like to use it under
: >>LINUX... I'm using 1.1.38, and the patch I found sony535-0.5 doesn't patch
: >>cleanly into the 1.1.38.  I would like to know if anyone is using one of these 
: >>drives, and also, if this patch, (if done by hand) will cause problems with
: >>the 1.1.30 and up kernels.  I'd like to use 1.1.39 if possible.
: >>
: >>Could support for these drives be put into the stock kernel?  (since this is
: >>extremely close to the CDU 31A type drives.)


: >I am using the yggdrasil kernel, and it works with my CDU535.  I don't know
: >which patch level the ygg kernel is at..
: >--
: >Jeffrey Comstock
: >INET: jrc@brainiac.mn.org               AT&T: 317-578-0884
: >CW:   -. .-. ----- -..             

: YES! PLEASE keep the Sony535 as a standard in future kernels!
: I'm currently stuck with 1.0.<something> because of this!

I'm running 1.1.41 with my 535 running fine.. just apply the last patch, and
fix the errors manually.

Kris.
-- 
krismon@hh.sbay.org            The Electronic Battlefield Series FAQ    
kristoff@gas.uug.arizona.edu   http://www.hh.sbay.org/~krismon/public.html 
Prodigy: GRTV13B               ftp://hh.sbay.org/users/krismon/ebsfaq

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

From: rpao@paonet.org (Roger C. Pao)
Subject: SCSI & PCMCIA driver: kernel rescan request
Date: Sat, 13 Aug 1994 17:29:09 GMT

I'm looking at modifying a SCSI driver to work with a PCMCIA SCSI
adapter.  David Hinds would prefer I use his loadable modules model.
This sounds like a great idea as I don't want to write a point enabler
and conflict with his code in the future (esp. if/when he puts it into
the kernel).

The problem I have is, if I do this, how do I notify the kernel that
there is now a new SCSI adapter and devices out there?  I need to be
able to kick the kernel into rescanning for devices.

This would be great for non-PCMCIA adapters as well since I could then
leave my MEOD off as usual and turn it on when I need it without
rebooting Linux.  All I would need to do is run a program which makes a
system call to rescan.  The Tahiti 1 is just way too noisy to leave on
all the time.  I can barely work next to this ST41200NM here.

Another possibility is to have a register/callback design for the PCMCIA
SCSI driver.  The driver could register with the kernel specifying itself
as a new SCSI adapter.  The kernel/SCSI subsystem could then call the
driver as if it were boot-time loading.

I'd still like the rescan system call as well though for non-PCMCIA SCSI
adapters.

rp94
-- 
Roger C. Pao <rpao@paonet.org>

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

From: brian@xp.psych.nyu.edu (Brian Watts)
Subject: SLIP routing problems
Date: 10 Aug 1994 16:28:40 GMT


Kernel: 1.1.35 -- dip version: 3.3.7duri

I am having difficulty getting SLIP to work over the network.
Using dip/sliplogin, I can connect to a Linux box from another
Linux box and establish communication between the two systems.
(BTW, neither dip nor sliplogin configures the interface correctly,
this has to be done by hand on both side -- the reason is a mystery
to me.)
The host has a leased line connection to the Internet through
a router: 100.35.249.254.  I can do anything between the two
machines: 199.35.249.1 and 199.35.249.2, BUT I can't connect
to other machines on the network (ping doesn't even work) and
I can't connect to anything via the router.

I also have the same problem when connecting an MSDOS box to the
host using WINSOCK -- i.e. I can communicate directly, but to no other
machines.

SLIP client:

lo        Link encap Local Loopback  
          inet addr 127.0.0.1  Bcast 127.255.255.255  Mask 255.0.0.0
          UP LOOPBACK RUNNING  MTU 2000  Metric 1
          RX packets 0 errors 0 dropped 0 overrun 0
          TX packets 146 errors 0 dropped 0 overrun 0

sl0       Link encap Serial Line IP  
          inet addr 199.35.249.2  P-t-P 199.35.249.20  Mask 255.255.255.0
          UP BROADCAST POINTOPOINT RUNNING  MTU 296  Metric 1
          RX packets 206 errors 0 dropped 0 overrun 0
          TX packets 386 errors 0 dropped 0 overrun 0

Kernel routing table
Destination     Gateway         Genmask         Flags Metric Ref Use    Iface
199.35.249.20   0.0.0.0         255.255.255.255 UH    0      0        0 sl0
127.0.0.0       0.0.0.0         255.0.0.0       U     0      0      146 lo
0.0.0.0         199.35.249.20   0.0.0.0         UG    0      0       33 sl0


SLIP host:

lo        Link encap Local Loopback  
          inet addr 127.0.0.1  Bcast 127.255.255.255  Mask 255.0.0.0
          UP BROADCAST LOOPBACK RUNNING  MTU 2000  Metric 1
          RX packets 0 errors 0 dropped 0 overrun 0
          TX packets 34 errors 0 dropped 0 overrun 0

sl0       Link encap Serial Line IP  
          inet addr 199.35.249.1  P-t-P 199.35.249.20  Mask 255.255.255.0
          UP POINTOPOINT RUNNING  MTU 296  Metric 1
          RX packets 562 errors 0 dropped 0 overrun 23
          TX packets 351 errors 0 dropped 0 overrun 29

eth0      Link encap 10Mbps Ethernet  HWaddr 00:00:21:29:97:06
          inet addr 199.35.249.1  Bcast 199.35.249.255  Mask 255.255.255.0
          UP BROADCAST RUNNING  MTU 1500  Metric 1
          RX packets 52489 errors 0 dropped 0 overrun 0
          TX packets 76775 errors 0 dropped 0 overrun 0


Kernel routing table
Destination     Gateway         Genmask         Flags Metric Ref Use    Iface
199.35.249.20   0.0.0.0         255.255.255.255 UH    0      0        0 sl0
199.35.249.2    199.35.249.20   255.255.255.255 UGH   0      0       43 sl0
199.35.249.0    0.0.0.0         255.255.255.0   U     0      0      612 eth0
127.0.0.0       0.0.0.0         255.0.0.0       U     0      0       34 lo
0.0.0.0         199.35.249.254  0.0.0.0         UG    0      0    71896 eth0

If anyone can help, I'd be most grateful.

Thanks.

Brian

======================================================================
brian@xp.psych.nyu.edu
______________________________________________________________________
Contact                      | Brian Watts
World Trade Markets          | email: bwtm@aol.com
109 W 27th Street            | tel: (212) 229 2585

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

From: parki005@gold.tc.umn.edu (Bruce Parkin)
Subject: X11R6 build on linux: won't; help!
Date: Sat, 13 Aug 1994 18:24:12 GMT

I've followed all of the instructions the best
I can, and hacked around and mucked a lot; but
I'm not getting anywhere. Any help would be 
appreciated.

I am trying to build X11R6 on an up to date
version of Linux, gcc, and all that. I've applied
all three public patches to the source tree (xc).

I seem to be running into a number of problems.
The first...

I do a "make World" and get:

=========>
making Makefiles in lib/X11...
        mv Makefile Makefile.bak
make[3]: Entering directory `/usr/src/X11R6/xc/lib/X11'
Makefile:673: *** commands commence before first target.  Stop.    
=================>

I start working on this problem, and I get a whole slew of
problems all concerned with the shared library build particular
to Linux.

I can seem to find no documentation to help me.

Really, any help would really be appreciated.

Thanks.

Bruce Parkin

---

bruce.parkin-1@umn.edu




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

From: lendecke@namu06.gwdg.de (Volker Lendecke)
Subject: Re: Proposal to write a SAMBA fs client for the kernel
Date: 13 Aug 1994 17:35:58 GMT


In article <PC.94Aug12123125@ISOlde.dale.dircon.co.uk> pc@dale.dircon.co.uk (Pete Chown) writes:

   This is already being done - not exactly in the way you anticipate,
   but as a client of userfs.  I suspect this may well be better, since
   it will keep as much code as possible out of the kernel, which is big
   enough already.

Hi,

maybe by "This is already being done" you mean my work, which
Andrew Tridgell has put nimbus.anu.edu.au. The version you see
there is something I did not really wanted to show to too many
others, I just mailed it to some for testing. In the last 4 weeks
I could not read mail, because I had a job where my parents
live. There I wrote a version which I dare to show to
others. There are 2 things missing: You cannot remove or rename
files.  These are really easy and should be done in the next few
days. I will keep you informed.

There's one thing to mention, for those already using
userfs-0.7.1:
I changed the protocol a bit, because I think in one special case
it is not sufficient. I did not tell it to Jeremy, who might have
a better solution for my problem. So you cannot use other clients
and my smbfs together.

Concerning performance, I have REALLY poor results. I did not
measure anything, but the pure reading performance is bad. I do
not have a clue where this might come from. Is it WfW, is it WfW
or linux TCP, or is it the linux kernel, which must schedule()
between smbfs and the reading client process, I do not know. So
do not expect too much.

I will send the finished version 0.1 to Andrew as soon as it is
ready.

bye, Volker

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

From: rob@pe1chl.ampr.org (Rob Janssen)
Subject: Re: BOCA ioAT66 serial card??
Reply-To: pe1chl@rabo.nl
Date: Fri, 12 Aug 1994 21:24:44 GMT

In <32g0a1$ros@scapa.cs.ualberta.ca> steven@gsb019.cs.ualberta.ca (Steven Charlton) writes:

>Has anyone hacked the kernel to properly support this serial card?  I
>have one, and must get it working in fairly short order.  This is a 6
>port, 3 UART (16550) serial card.  I noticed support for some BOCA
>serial card, but the IRQs and base addresses seemed wrong for this
>one.  Any and all help would be greatly appreciated.  E-mail response
>would be preferable.

You don't need to hack the kernel to support this card.  Just issue
6 "setserial" commands from your /etc/rc.serial, sepcifying the I/O
addresses and IRQs you assigned to the ports.  (IRQs may be the same)

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

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

From: creepus@connected.com (Corey Ladas)
Subject: Is there a long-term plan for kernel development?
Date: 12 Aug 1994 15:43:18 -0700

I'm just curious about where this is all heading.
It kind of seems like most work on the kernel (the official version anyway...)
is in the areas of improved device support and minor performance tweaks.
Given the direction the 'state-of-the-art' is heading (eg: multi-threaded,
real-time, distributed, microkernel, &c):

Is Linux going to be a viable alternative in 2 years?  ...in 5 years?

--
A concerned and curious Linux fan...
-- 
[brain off]
"your mother is like a box of chocolates..."
GAT -d+ H--- s g+ p2 au+ a- w@ v--- C++ U++ P++ L++>++++ 3 E N++ K++ W--- 
    M-- V-- po--- Y+ t+ 5- !j R G' tv b+++ D+ B--- e++ u* h- f r++ n y

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


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