Subject: Linux-Development Digest #914
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:     Thu, 14 Jul 94 15:13:06 EDT

Linux-Development Digest #914, Volume #1         Thu, 14 Jul 94 15:13:06 EDT

Contents:
  Re: 1.1.28: SCSI disks inverted ! (Chris Adams)
  Re: Why no AFS? (Yuri Trifanov)
  Re: Is XFree86 3.X from X11R6 ? (Dirk Hohndel)
  Re: binutils-2.4 (ld,nm) under Linux (Eric Youngdale)
  Re: 1.1.28: SCSI disks inverted ! (Eric Youngdale)
  1.1.28: SCSI (Heinz Kranefeld)
  Re: Bugs in 1.1.27 Makefiles (Kevin Lentin)
  Re: Is XFree86 3.X from X11R6 ? (Dan Logue)
  Re: Multicasting. (Cheng-Kang Wen)
  Re: Driver out for Xircom parallel-port ether adapter? (Frank Lofaro)
  Re: Energy Star Screen Saver for X? (Avinoam Shmuelin)
  Massive GCC error detected (Chris Worley)
  Re: 1.1.28: SCSI disks inverted ! (Rene COUGNENC)
  1.1.28: SCSI disks inverted ! (Rene COUGNENC)
  Returning free'd memory (Kent A Vander Velden)
  Re: Massive GCC error detected (NetDog)
  Re: Programming in Linux (Michael Will)
  Intelligent Digiboards & Linux (Robert P Glamm)
  Re: Massive GCC error detected (Curt L. Olson (Admin))

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

From: racerx@vespucci.iquest.com (Chris Adams)
Subject: Re: 1.1.28: SCSI disks inverted !
Date: 14 Jul 1994 00:27:11 -0500

In article <301e38$5f@renux.frmug.fr.net>,
Rene COUGNENC <cougnenc@blaise.ibp.fr> wrote:
>Yes, I realize that.
>I never have boot problems because as an old Linux user (since 0.12),
>I always have two (at least) boot floppies handy:
>
>       - One with a "stable" kernel
>       - One with the last kernel I am testing.
>
>It is generally a good idea to test the new patchlevels from a boot
>floppy, and make them the default kernel on the hard drive *only* when
>you are sure that everything works proprely.
. 
. 
. 
>PS: You may be able to specify the boot device with the <tab> key under LILO,
>    I am not used to this stuff so I don't comment it...

If you are using LILO, then you can always have a "stable" kernel handy
right on your hard disk!  Just add another entry when you have a new
kernel to test out, specify it when booting, and, if it works, you can
make it the default.  If it doesn't work, just boot up the old kernel.
No need for floppies.

I have done this lots of times.  Even when I first installed Linux and
recompiled my kernel, I left the one that was put there by the
installation (SLS 1.2 - Ugh!) as an option under LILO in case I screwed
up the configuration part.  It just makes good sense.  There is no law
that ALL kernels must be named zImage after all.
-- 
Chris Adams
racerx@iquest.com
United States Space Academy Simulations Director and Trainer
"With friends like these, who needs halucinations?" - Buddy, 'Night Court'

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

From: yuri@shimari.cmf.nrl.navy.mil (Yuri Trifanov)
Subject: Re: Why no AFS?
Date: Wed, 13 Jul 1994 08:37:56 GMT


One would have to start out from scratch from the outdated protocol
documentation and packets stolen from the network.

AFS is a continually evolving product entirely proprietary to 
Transarc. The source is fairly expensive and non-redistributable.

Given Transarc's agenda concerning DCE, and the accumulated cruft in
the AFS work, it might be wiser to redesign a competing free
product which incorporates some of AFS's better features.

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

From: hohndel@aib.com (Dirk Hohndel)
Subject: Re: Is XFree86 3.X from X11R6 ?
Date: Thu, 14 Jul 1994 11:59:48 GMT

Dan Logue (dlogue@starbase.neosoft.com) wrote:

: : It's X11R6 based. At this point of time I don't know what new features
: : will be supported, but I expect S3-864/964 support for some boards and a
: : few other new servers like P9000 and a few SVGA ones.


: I just ordered a miro 20S video card that uses an S3-864. I take it from this
: remark that the board will not be able to be used with the existing XFree86 2.0
: S3 accelerated driver.  Is this right?  If so, is there any temporary work around for the time (such as a patch for S3-864) or must I default to basic VGA support?

There is a hacked version on some FTP servers, but I cannot say whether
it works with your board or not. We don't support it.

: Also, I know it may be too early to give a date when version 3.0 is out

3.0 is out (part of X11R6) but almost unusable on Linux

: but could you give aaaany estimate ...2 months, 3 months, 6 months?
: I sure will miss my 1024x768 by 256 colors if I must revert to basic VGA for a
: while.

I cannot give you a date for 3.1, as we don't have a schedule yet. But
we're working on it.

        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: ericy@cais.cais.com (Eric Youngdale)
Subject: Re: binutils-2.4 (ld,nm) under Linux
Date: 14 Jul 1994 11:58:28 GMT

In article <BASILE.94Jul12193921@rosser.soleil.serma.cea.fr>,
Basile STARYNKEVITCH <basile@rosser.soleil.serma.cea.fr> wrote:
>
>Did someone manage to make GNU binutils-2.4 work under Linux?  I did
>compile (with default configure -v --prefix=/usr) it but the resulting
>ld didn't work (something like m486 (or perhaps 386) emulation not
>found - I don't remember the exact message - it is curious since I have a
>486DX2/66). Also, the resulting nm dumps core.

        There have been a lot of changes since the last public release.  
The latest snapshot now has these features:

        1) Linux QMAGIC format now used as default output.
        2) Strip program can successfully strip either ZMAGIC or QMAGIC.
        3) Linux shared libraries now fully supported.
        4) Linker now supports linking against ELF shared libraries.
        5) Default configuration for linux has both support for a.out and ELF
(a.out is the default).

Patches not yet merged into official distribution:

        6) Support for generating ELF shared libraries nearly done - 
        I need about one more evening, and I can probably finish it so
        that my test library correctly links.
        7) Patches exist for assembler to accept PIC operands.

        I do not know when the next public release will be - certainly I
would suggest that we wait until all of the patches have been integrated,
but this decision is up to the maintainer.

        Incidentally, the dual ELF/a.out support in the BFD library has 
been present for quite some time - most people do not know this, but the 
gdb that people are using contains support for debugging ELF binaries.

-Eric


-- 
"The woods are lovely, dark and deep.  But I have promises to keep,
And lines to code before I sleep, And lines to code before I sleep."

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

From: ericy@cais.cais.com (Eric Youngdale)
Subject: Re: 1.1.28: SCSI disks inverted !
Date: 14 Jul 1994 12:07:42 GMT

In article <301e38$5f@renux.frmug.fr.net>,
Rene COUGNENC <cougnenc@blaise.ibp.fr> wrote:
>Ce brave Bernd Mielke ecrit:
>
>> rene@renux.frmug.fr.net (Rene COUGNENC) writes:
>> >Funny :-)
>> >The two disks sda and sdb are inverted; I find the mirror stuff in /usr,
>> >the var stuff in /home...
>>
>> This Error is not funny :-<
>> It prevents to boot my machine!!!
>
>I hope that Eric Youngdale sent you the bug fix, if not let me know,
>I'll post it to you.

        No need.  Get 1.1.29, and the fix is included.

        In case anyone is curious, I am slowly attempting to rework the 
scsi code so that it would be possible to load a driver as a loadable 
module - as part of this, I changed the array of devices into a linked 
list, but I was inserting new devices at the wrong end of the linked list.

        Please avoid 1.1.28.  There was a patch inadvertently included
which was a preliminary attempt to fix the swapping problem with
fdomain/aha152x/st0x cards, but it was not right and it would lead to the
system hanging if you start swapping.  I believe that this problem has now
(finally) been completely solved with 1.1.29.  If you have one of these 
cards, please try 1.1.29 and see whether swapping is now stable.

-Eric

-- 
"The woods are lovely, dark and deep.  But I have promises to keep,
And lines to code before I sleep, And lines to code before I sleep."

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

From: pthk@ptb1.pt.tu-clausthal.de (Heinz Kranefeld)
Subject: 1.1.28: SCSI
Date: Thu, 14 Jul 1994 11:55:00 GMT

Hi guys,
with the kernel 1.1.28 i got funny partition tables from my
SCSI disks.
The disks are:
   Fujitsu M2624-512      sda
   DEC     DSP3105-S      sdb
   IBM     OEM 0664M1H    sdc
Controler: Adaptec 1542
A DOS partion is on sda1. fdisk shows wrong block counts and during
boot up i got the message:
   VFS: Can't find a valid MSDOS filesystem on 0x0803
The second disk is o.k. i think (hope). From the third disk fdisk only
shows the first three partitions (no logical partition of the extended
partition is shown).
If i boot another kernel (1.1.27) all things are correct with the disks. 
Thanks for your time
Heinz


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

From: kevinl@bruce.cs.monash.edu.au (Kevin Lentin)
Subject: Re: Bugs in 1.1.27 Makefiles
Date: 14 Jul 1994 12:39:07 GMT

Matthias Urlichs (urlichs@smurf.noris.de) wrote:

> > 2. The drivers for de600 and de620 get recompiled every time make is run,
> > even though they have not changed since the last make.
> > 
> Probably because if you're majing with -j3 (or whatever), ar gets called
> multiple times on the same archive -> confusion.

This happened to me for quite a while too. I think I had to empty the
MODULES file to stop it happening.

-- 
[==================================================================]
[ Kevin Lentin                   |___/~\__/~\___/~~~~\__/~\__/~\_| ]
[ kevinl@bruce.cs.monash.edu.au  |___/~\/~\_____/~\______/~\/~\__| ]
[ Macintrash: 'Just say NO!'     |___/~\__/~\___/~~~~\____/~~\___| ]
[==================================================================]

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

From: dlogue@starbase.neosoft.com (Dan Logue)
Subject: Re: Is XFree86 3.X from X11R6 ?
Date: 14 Jul 1994 13:09:04 GMT

Dan Logue (dlogue@starbase.neosoft.com) wrote:
: Dirk Hohndel (hohndel@aib.com) wrote:
: : javet@didec30.epfl.ch wrote:
: : :  Hello,


: : :  Could someone tell me what's in XFree86 3.X ? Is the X11R6 version or what ?
: : :  If not, what's new ?


: : It's X11R6 based. At this point of time I don't know what new features
: : will be supported, but I expect S3-864/964 support for some boards and a
: : few other new servers like P9000 and a few SVGA ones.


: I just ordered a miro 20S video card that uses an S3-864. I take it from this
                                                       ^^^
Sorry, the chip I believe is the S3-964 (there are patches for the 864 on 
ftp.cdrom.com). The change made to support the S3-864 was just to accept the
S3-864 as a S3-928, I wonder if the same would work for the S3-964?

Anybody tried this for an S3-964 board yet?

 Dan     dlogue@neosoft.com

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

From: ckwen@pc210.ee.ncku.edu.tw (Cheng-Kang Wen)
Subject: Re: Multicasting.
Date: Thu, 14 Jul 1994 11:31:49 GMT

In article <1994Jul5.111543.5335@uk.ac.swan.pyr> iiitac@uk.ac.swan.pyr (Alan Cox) writes:

>In article <2vb687$hcc@news.ysu.edu> s0017210@cc.ysu.edu (Steve DuChene) writes:
>>Alan Cox (iiitac@uk.ac.swan.pyr) wrote:
>>      Fogive my dumb question, but what is multicasting vis-a-vis networks?

>Sending data to a group of machines not just one or all. Its used extensively
>by the MBONE research projects (get the MBONE.FAQ from somewhere).

>ALan


Can Linux run mrouted or be used as router supporting IP multicast in the 
future work ? 

Cheng-Kang


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

From: ftlofaro@unlv.edu (Frank Lofaro)
Subject: Re: Driver out for Xircom parallel-port ether adapter?
Date: Wed, 13 Jul 94 03:01:44 GMT

In article <CsuJ5F.3r6@pe1chl.ampr.org> pe1chl@rabo.nl writes:
>In <Csspv5.n3F@news.cv.nrao.edu> juphoff@tarsier.cv.nrao.edu (Jeff Uphoff) writes:
>
>>The subject says it all...I just bought a laptop, and my employer has
>>several Xircom parallel-port ethernet adapters here (so I'd like to use
>>one of them and not have to buy a different one).  The README.DLINK file
>>in the kernel net drivers states (for the D-Link drivers):
>
>>        This is a set of Ethernet drivers for the D-Link DE-600/DE-620
>>        pocket adapters, for the parallel port on a Linux based machine.
>>        Some adapter "clones" will also work.  Xircom is _not_ a clone...
>
>>So, since the D-Link drivers don't work, has anyone been working
>>on/created a driver for the Xircoms?
>
>Xircom adapters are in the same ballpark as Diamond video cards:
>the manufacturer does not provide any info without non-disclosure
>agreement, hence no drivers for Linux are available and it is not
>likely they will ever be...
>
>Rob
>-- 

One can always try to reverse engineer the DOS driver.
It may be hard, but probably not impossible.
Once you figure out what needs to be done to activate the card's functions, 
writing a Linux driver should be straight-forward (albeit not necessarily 
simple)



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

From: shmueli@moomintroll (Avinoam Shmueli)n
Subject: Re: Energy Star Screen Saver for X?
Date: 14 Jul 1994 13:31:15 GMT
Reply-To: shmueli@student.umass.edu

In article r15@lyra.csx.cam.ac.uk, and1000@cus.cam.ac.uk (Austin Donnelly) writes:

|Erm, Christoph Rimek <chrimek@rimki.toppoint.de> posted here in April
|with his patch which does exactly what you plan. Look at:
|
|sunsite.unc:/pub/Linux/kernel/patches/console/linux.1.0.power-save.patch

Thanks Austin. It's there! I should have looked first. Now let's see
why it won't work under X

--Avi





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

From: cworley@galena.use.com (Chris Worley)
Subject: Massive GCC error detected
Date: Wed, 13 Jul 1994 02:57:00 GMT

Very stupid error.
gcc: internal compiler error: program cc1 got fatal signal 11
cpp: output pipe has been closed

This keeps happening no matter what I compile.  Near as I can
figure there is some kind of cumilative resource or memory
leak that causes an overload.  Notice, I wish to recommend
a change to gcc, such that on this error the current output
file is to be deleted.  Top does not report any unusual
allocations in memory or resources.  Error normally occurs
after about 10 minutes.  After a recieved a few of these,
I then recieved the following in order.
gcc: internal compiler error: program cc1 got fatal signal 6
cpp: output pipe has been closed

Kernel Panic

fsck inode size does not match
root may not log into this terminal (Console)
login timed out continueing (60 secs)

I now have about a 700-900kb unexplained hole in my memory.

This was gcc 2.5.8 Kernel 1.0.8 -> 1.1.23.
I reset the entire file system (198meg) three times with no success
so far.
-- 
===========================================================
Guild Commander                      cworley@galena.use.com
===========================================================

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

From: rene@renux.frmug.fr.net (Rene COUGNENC)
Subject: Re: 1.1.28: SCSI disks inverted !
Date: 13 Jul 1994 19:08:23 GMT
Reply-To: cougnenc@blaise.ibp.fr (Rene COUGNENC)

Ce brave Bernd Mielke ecrit:

> rene@renux.frmug.fr.net (Rene COUGNENC) writes:
> >Funny :-)
> >The two disks sda and sdb are inverted; I find the mirror stuff in /usr,
> >the var stuff in /home...
>
> This Error is not funny :-<
> It prevents to boot my machine!!!

Yes, I realize that.
I never have boot problems because as an old Linux user (since 0.12),
I always have two (at least) boot floppies handy:

        - One with a "stable" kernel
        - One with the last kernel I am testing.

It is generally a good idea to test the new patchlevels from a boot
floppy, and make them the default kernel on the hard drive *only* when
you are sure that everything works proprely.

I hope that Eric Youngdale sent you the bug fix, if not let me know,
I'll post it to you.

In order to recompile your kernel, you'll have to boot... The best
way is to boot from a rescue disk (or a installation disk, Slackware
provides many floppies like that), then mount your root partition
and edit whatever file you use to mount your hard drives.
(/etc/fstab or /etc/rc). Invert the disks names, reboot. It should work.

Then apply the patch, recompile the kernel, modify one more time the
devices you mount since you'll boot from a correct kernel now.

PS: You may be able to specify the boot device with the <tab> key under LILO,
    I am not used to this stuff so I don't comment it...
--
 linux linux linux linux -[ cougnenc@renux.frmug.fr.net ]- linux linux linux 

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

From: rene@renux.frmug.fr.net (Rene COUGNENC)
Subject: 1.1.28: SCSI disks inverted !
Date: 13 Jul 1994 01:48:15 GMT
Reply-To: cougnenc@blaise.ibp.fr (Rene COUGNENC)



Funny :-)

With 1.1.28, when I:

        mount /dev/sda1 /usr
        mount /dev/sda2 /home
        mount /dev/sdb1 /mirror
        mount /dev/sdb2 /var

The two disks sda and sdb are inverted; I find the mirror stuff in /usr,
the var stuff in /home...

This is with xiafs, Adaptec 1540B, kernel 1.1.28.

 
--
 linux linux linux linux -[ cougnenc@renux.frmug.fr.net ]- linux linux linux 

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

From: graphix@iastate.edu (Kent A Vander Velden)
Subject: Returning free'd memory
Date: 14 Jul 94 18:13:41 GMT

Are there any planed changes to the Linux kernel to actually completely give 
the memory back to the system which was free'd by the program before the 
program exits?



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

From: cdent@honors.indiana.edu (NetDog)
Subject: Re: Massive GCC error detected
Reply-To: cdent@indiana.edu
Date: Thu, 14 Jul 1994 02:35:35 GMT

>>>>> "B" == Beeblebrox  <M.S.Ashton@dcs.warwick.ac.uk> writes:

   B> grante@reddwarf.rosemount.com () writes:
   >> : gcc: internal compiler error: program cc1 got fatal signal 11

   >> This is almost always a hardware error.

   B> A friend of mine had a similar problem.  Believe it or not he
   B> solved it by _reducing_ a wait-state.  Fiddle with your CMOS
   B> settings.

I've only seen the problem while compiling the kernel, and have
stopped it by kill some process that aren't needed at that time (like
X). After that it is fine.

chris

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

From: michaelw@desaster.student.uni-tuebingen.de (Michael Will)
Subject: Re: Programming in Linux
Date: Tue, 12 Jul 1994 17:57:46 GMT

bass@cais.cais.com (Tim Bass (Network Systems Engineer)) writes:
>I have the exact same problem with Stevens code.  All of his books and code
>are very difficult to compile because the lib and header file are simply not
>there on my Linux box.  This is true for all Steven's book.  When I look
>at the header files, they have entries like:

>#ifdef notdef
>blah, blah
>#endif 

>I mean really, what is the "notdef" variable supposed to mean???
It is meant as an undefined variable - it is yet another way to comment
out code :-)

Cheers, Michael Will
-- 
. .         Michael Will <michaelw@desaster.student.uni-tuebingen.de> 
 .      cs-student in Tuebingen, Germany, Europe, Earth, Solar-System, [...]

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

From: glam0001@gold.tc.umn.edu (Robert P Glamm)
Subject: Intelligent Digiboards & Linux
Date: Thu, 14 Jul 1994 14:51:33 GMT

With all the discussions I've seen about intelligent Digiboards, I've just
gotta post something... ;)

Anyway, Digiboard is based just down the road from me in Eden Prarie, Minnesota.
I know two people that work there.  From the information I've gotten, to make
a intelligent Digiboard work with an OS (say, a concentrator with SCO or Linux)
you've gotta have two parts: the FEP & the kernel driver.

The kernel driver shouldn't be to hard to write (if you know the interface).
Unfortunately, the FEP gets uploaded into the intelligent Digiboard when you
start the machine, and is necessary to drive the card.  This FEP is only
distributed in binary form along with the kernel drivers that Digi ships
out.  

I think there also might be some proprietary serial chips on the boards as well.
This makes these things damn hard to reverse engineer a software driver for.

I've not read too many of the Digiboard posts, I've just seen the titles.  My
suggestion, if a driver hasn't been written, is to get a copy of the FEP and
try to hack a driver.  If you can't get the FEP, I would think that you'd be
SOL.

This is just for intelligent boards & at least some of their concentrators.

Non-technical questions (such as 'How can I get a *W@#$ manual from Digi
to write my own driver??') can be directed back at me.

Bob Glamm
glamm@msi.umn.edu


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

From: clolson@me.umn.edu (Curt L. Olson (Admin))
Subject: Re: Massive GCC error detected
Date: Thu, 14 Jul 1994 14:59:43 GMT

cdent@honors.indiana.edu (NetDog) writes:

>>>>>> "B" == Beeblebrox  <M.S.Ashton@dcs.warwick.ac.uk> writes:

>   B> grante@reddwarf.rosemount.com () writes:
>   >> : gcc: internal compiler error: program cc1 got fatal signal 11
>   >> This is almost always a hardware error.
>   B> A friend of mine had a similar problem.  Believe it or not he
>   B> solved it by _reducing_ a wait-state.  Fiddle with your CMOS
>   B> settings.
>I've only seen the problem while compiling the kernel, and have
>stopped it by kill some process that aren't needed at that time (like
>X). After that it is fine.
>chris

I had similar problems with my AMD 486/66DX2 which were fixed by adding
a cooling fan to the cpu.

Curt.

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


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