Subject: Linux-Development Digest #920
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, 16 Jul 94 10:13:04 EDT

Linux-Development Digest #920, Volume #1         Sat, 16 Jul 94 10:13:04 EDT

Contents:
  Re: Massive GCC error detected (Ian McCloghrie)
  Info needed on hd.c (Karl Hakimian - staff)
  Re: building a PCMCIA SCSI adaptor (Ralph Becker-Szendy)
  Re: PCMCIA Kernel (Jay Denebeim)
  Re: Is XFree86 3.X from X11R6 ? (Dan Logue)
  Re: Linux 1.1.29: telnet shells dying??? (Wolfgang Siebeck)
  Re: non SCSI CD-ROM (Panasonic) and Linux (Eberhard Moenkeberg)
  Re: serious defect on linux (Bill Hogan)
  Memory crunch in 1.1.24? (Richard Whittaker)
  Re: Linux Performance Enhance ? (Jim Balter)
  gcc 2.6.0 (David Miller)
  mt tell not working (Dirk Sturzebecher)
  Re: Is XFree86 3.X from X11R6 ? (Frank Lofaro)
  Re: Massive GCC error detected
  Re: 1.1.28: SCSI disks inverted !

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

From: imcclogh@cs.ucsd.edu (Ian McCloghrie)
Subject: Re: Massive GCC error detected
Date: 15 Jul 94 16:39:37 GMT
Reply-To: ian@ucsd.edu

grante@reddwarf.rosemount.com writes:
>It'll be back.  You just managed to avoid the bad spot or access
>pattern in the L2 cache, or in RAM.  Other people reported that the
>problem only occurred when compiling certain versions of the kernel,
>or with certain versions of gcc.  As far as I know, it still turned
>out to be a hardware problem.

Given the current month (July), I suspect it's probably a heat
problem.  I ended up having to replace the case of my system last
summer because of overheating problems.  (Not that this was a bad
thing, I was very glad to get rid of that mini-tower)

--
____
\bi/  Ian McCloghrie      | FLUG:  FurryMUCK Linux User's Group
 \/   email: ian@ucsd.edu | Card Carrying Member, UCSD Secret Islandia Club
GCS (!)d-(--) p c++ l++(+++) u+ e- m+ s+/+ n+(-) h- f+ !g w+ t+ r y*

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

From: hakimian@haney.eecs.wsu.edu (Karl Hakimian - staff)
Subject: Info needed on hd.c
Date: Fri, 15 Jul 1994 16:36:11 GMT

Who is the current "expert", "maintainer", etc of the

/usr/linux/devices/block/hd.c (etc.)

file(s).

I am trying to understand the section that complains when it sees a disk with
more than 16 heads.

I tried an experiment where I removed the check and then setup my ide drive to
use the ILB (?) translation (this makes the drive look like it has 32 heads).
The result was Linux could see the drive when it boots (this was not the case
before the change) but an attempt to mount that disk give a kernel panic
(trying to write to an out of bound sector, or some such message). Since I did
not re-newfs the disk I did not find this that surprising but a couple of
things I saw did not make sense.

1) I mounted the disk readonly, why was it trying to write?

2) When Linux came up it thought the disk has a geometry with 16 heads. When I
   don't use the translation this is what it should see. When I use the
   translation it was complaining about it being 32 heads. When I remove the
   check it then comes up and sees it as 16 heads. What is happening?

Anyone have any suggestion on where to go from here?

Thanks.
--

Karl Hakimian
hakimian@eecs.wsu.edu

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

Crossposted-To: comp.os.linux.misc
From: ralph@falcon.SLAC.Stanford.EDU (Ralph Becker-Szendy)
Subject: Re: building a PCMCIA SCSI adaptor
Date: Thu, 14 Jul 1994 18:44:14 GMT

In article <1994Jul5.181959.18730@lamont.ldgo.columbia.edu> 
sid@ldgo.columbia.edu (sidney hellman) writes:
>       A few months ago I posted some questions about SCSI pcmcia
>adaptor for linux, and discovered that as of now there is nothing
>available.  We are about to give up on getting the programming info
>from Adaptec and their not-so-cooperative competitors, and are
>considering making up our own SCSI pcmcia adaptor.  WE ahve two
>questions at this point.

A. Are you aware of the complications involved in building a PCMCIA card?
The fact that you have about 6 in^2 of area for the whole circuit are just
the beginning. Everything has to be surface mount and rather low height, the
connector is hard to find (maybe impossible to find in small quantities),
and assembly of such a board is nearly impossible for all but the most
advanced amateurs. Due to space constraints, glue logic has to be pretty
much in an EPLD. Maybe you took all this into account.

B. Adaptec is not the only manufacturer of PCMCIA-SCSI cards. Here are a few
hints about others.  First, there was a one-page article in "Electronic
Design" about PCMCIA-SCSI cards (June 13 issue, page 127). Second, there is
a PCMCIA standards association, which can be reached at (408)720-0107. The
PCMCIA-SCSI working group is headed by Larry Barn of NCR; he might know what
the status of the PCMCIA-SCSI standard is, and what vendors implement it.

C. The PCMCIA-association knows of the following manufacturers of
PCMCIA-SCSI cards (I'm not sure the list is complete, and errors in the
spelling and the phone numbers are all mine):
- Adaptec (which we all know and dislike),
- Lytronic (714)545-6649
- Magic RAM (213)413-9999
- New Media (714)453-0100
- Premex (714)851-8242
- QLogic (714)668-5365
- TWD (719)634-8187

>       2) what SCSI controller chips do people prefer.  We are
>thinking about using the WD3393A because our Engineer is familiar with
>it, but are there other chips that are preferred by Linux? (i.e.
>Fujitsu or NCR perhaps)
I did one project with the WD 33C93A, and found it to be HORRIBLE. The
documentation is lousy. The chip has a multitude of bugs, and getting bug
information from WD is like pulling teeth, but more difficult. The chip is
very sensitive to glitches on the SCSI lines (which trip up its internal
sequencer, which can be taken care of in software, but is a lot of extra
work). On the other hand, once you get it to work (which is harder than with
other chips), it's performance is suppesedly very good, compared to other
SCSI chips of similar age (such as the NCR 53C90 series, or the ubiquitous
53C80). Definitely not a chip for a small project. I would recommend the NCR
53C90 series (which I used to replace the 33C93A), which is much easier to
work with (partly because the guys at NCR are very friendly and helpful), or
maybe one if its more modern successors.

-- 
Ralph Becker-Szendy                                 RALPH@SLAC.STANFORD.EDU
Stanford Linear Accelerator Center                      RALPH@SLACVM.BITNET
M.S. 95, P.O. Box 4349, Stanford, CA 94309                    (415)926-2701
My opinion. This is not SLAC, Stanford U, or the US DoE speaking. Just me.

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

From: denebeim@deepthot.cary.nc.us (Jay Denebeim)
Subject: Re: PCMCIA Kernel
Date: 16 Jul 94 03:11:18 GMT

In article <305im6$1t1@koudai.cs.titech.ac.jp>,
Rafael Kazumiti Morizawa <rafael@cs.titech.ac.jp> wrote:
>I am trying to reconfigure my kernel on my T3400 to use the PCMCIA cards
>and I got a compilation error. It seems that there is a file missing in 
>the directory '/usr/src/linux/drivers/pcmcia' called 'cisreg.h'.
>
>The kernel version I am using is 1.1.27, and the package of the kernel card
>services is level 2.10, developed by David Hinds. In the README file
>that was in the package was written that it worked with kernel version
>1.1.20 

This is more of a 'me too', I tried it on 1.1.20 (I use CVS here so I
can get basically any version to run for me)

Jay

P.S. I have a DE-650 and a compaq concerto (cirrus chip), does anyone
know of any other drivers that work with it?  I used the older DE-650
driver from Mr. Hines, but although it recognizes the board, it won't
work with it properly.

Jay


-- 
Sig under construction
Jay Denebeim
jay@deepthot.cary.nc.us                 denebeim@deepthot.cary.nc.us
denebeim@deepthot.cybernetics.net       duke!wolves!deepthot!denebeim

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

From: dlogue@starbase.neosoft.com (Dan Logue)
Subject: Re: Is XFree86 3.X from X11R6 ?
Date: 14 Jul 1994 01:52:53 GMT

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
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?

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

Dan     dlogue@neosoft.com

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

From: siebeck@rmi.de (Wolfgang Siebeck)
Subject: Re: Linux 1.1.29: telnet shells dying???
Date: 16 Jul 1994 11:56:22 +0200

slouken@cs.ucdavis.edu (Sam Oscar Lantinga) writes:

>Note: The hang occurred with Sun OS machines.

I did the following:

- log in via telnet/ethernet to a Sun Sparc II running SunOS 4.1.1
- start the following script:
  > :
  >#
  >#    Testscript zum Ueberwachen einer Telnet-Verbindung
  >#

  >while true
  >do
  >     date
  >     sleep 60
  >done

No problem, running for about two hours. Version of Linux is:

  Linux artax 1.1.29 #2 Fri Jul 15 09:49:26 MET DST 1994 i486

  (eth0: WD80x3 at 0x280,  00 00 00 07 02 72 assigning address 0xd0000 WD8003-old, IRQ 9, shared memory at 0xd0000-0xd1fff.)

Wolfgang

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

Date: Thu, 14 Jul 1994 23:03:03 +0200
From: Eberhard_Moenkeberg@p27.rollo.central.de (Eberhard Moenkeberg)
Subject: Re: non SCSI CD-ROM (Panasonic) and Linux


Hello Laurent BOUAZIZ and all others,

on 13.07.94 Laurent BOUAZIZ wrote to All in USENET.COMP.OS.LINUX.DEVELOPMENT:

LB> Is there soft that would allow me to listen my CD's on a Panasonic
LB> (CR-562B) CD-ROM. Workman seems to require SCSI support.

workman -c /dev/sbpcd

Greetings ... Eberhard


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

From: bhogan@crl.com (Bill Hogan)
Subject: Re: serious defect on linux
Date: 16 Jul 1994 04:10:44 -0700

James Smith (smith@meishan.animal.uiuc.edu) wrote:
: i just run 13,973 K code on linux with 16 mb ram
: and it use swap memory.  The 1 minute cpu time
: become 2 hr run time.   This is very serious defect
: on swap memory.  Can anyone fix this?
: my friend found that when use swap memory the result is
: incorrect in previous kernel  but for 1.1.28 it work fine.
: I don't test in .29 yet.
: ========================

  Hello James.

  Excuse me, but _what_ 1-minute cpu time became a 2-hour run time?

  Do you mean perhaps that a program that ran in 1 minute when it was
dealing with a small dataset took 2 hours and a lot of swapping when you
gave it a larger dataset and you think that should not have happened?

  If so, before I would be willing to venture an opinion about whether it
should have happened or not, I would have to see the program you are
talking about. 

  Bill
  
-- 
  Bill Hogan
{echo "Subject: get bhogan@crl.com" | mail pgp-public-keys@pgp.mit.edu}

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

Crossposted-To: comp.os.linux.admin
From: rwhittak@orion.docwhitehorse.doc.ca (Richard Whittaker)
Subject: Memory crunch in 1.1.24?
Date: Sat, 16 Jul 94 06:49:47 GMT

Greetings from Whitehorse!

I've got an AMD386-40DX with 8MB of RAM, and didn't notice until recently
that Linux appears to only be looking at 4MB of the RAM that I have. Now, I
believe this to be a recent event, because previously I've been able to use
all 8Megs of RAM that I have installed. 

Top reports my memory as being:

Load Averages 3.07 3.07 3.06
44 processes: 43 sleeping, 1 running, 0 zombie, 0 stopped
CPU states:  3.7% user,  0.0% nice, 13.9% system, 82.5% idle
Mem:   4768K av,  4356K used,   412K free,  1460K shrd,   992 buff
Swap:  8200K av,  4612K used,  3588K free

The Kernel itself on bootup reports memory as being:

<6>Memory: 4772k/8192k available (776k kernel code, 384k reserved, 2260k data)

2260k of data????

Is this normal behaviour?..

Any help is really appreciated...

                                        Cheers,
                                        Rich W.
--
Richard Whittaker: Snailmail: 1102 Pine St, Whitehorse YT Y1A 4E8
  Internet E-Mail: rwhittak@orion.docwhitehorse.doc.ca 
Geographic Coords: 60 Deg., 45', 53" N., 135 Deg., 7', 17" W. 
    Amateur Radio: VY1RW, VY1RW@VY1DX, VY1RW@VY1BBS, 145.010 MHz         

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

From: jqb@netcom.com (Jim Balter)
Subject: Re: Linux Performance Enhance ?
Date: Sat, 16 Jul 1994 07:44:33 GMT

In article <CsyB3C.8ML@pe1chl.ampr.org>, Rob Janssen <pe1chl@rabo.nl> wrote:
>In <302t3f$t6v@smurf.noris.de> urlichs@smurf.noris.de (Matthias Urlichs) writes:
>
>>In comp.os.linux.development, article <CsvDLF.tv@pell.com>,
>>  orc@pell.com (Orc) writes:
>>> In article <CsuMMo.49F@pe1chl.ampr.org>, Rob Janssen <pe1chl@rabo.nl> wrote:
>>> >
>>> >Ok ok...  I admit that my piece of code was not the same as the original.
>>> >So here is a re-try.  It uses no goto at all, generates the same or better
>>> >code as Matthias's version, and it also uses a feature of the C language
>>> >that is not well-known...
>>> >
>>> >static int try_to_free_page(int priority)
>>> >{
>>> >   int i = 5;
>>> >   static int what = 0;
>>> >
>>> >   switch (what)
>>> >   {
>>> >       while (i--)
>>> >       {
>>> >       case 0:
>>> 
>>At least use a "do { } while(i--)" loop...  ;-)
>
>But that is not the same...  at least it would have to be
>"do { } while(--i)" to get equivalent behaviour.

Nyet.  Since the top of the while is skipped by the switch, it really is a
bottom test even though it looks like a top test.  So "while (i--) { }" and
"do { } while (i--)" are equivalent (in this case), whereas "do { } while (--i)"
never executes the body with i == 0.  That would still be true if this
were a regular loop instead of a Duff's Device.  In that case,
"i = 6; while (i--) { }" and "i = 5; do { } while (i--)" would be equivalent.
Gotta get those idioms right!  The easiest way to do it is to determine what
the value of i is for the first iteration and what it is for the last
iteration.  For "while (i--) { }" it's i-1..0; for "do { } while (i--)" it's
i..0; for "do { } while (--i)" it's i..1 (assuming i > 0).  A couple of others
which generate good code (i mustn't be declared unsigned; very important!) are
"while (--i >= 0) { }" (i-1..0) and "do { } while (--i >= 0)" (i..0).
-- 
<J Q B>

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

From: davem@eden.rutgers.edu (David Miller)
Subject: gcc 2.6.0
Date: 16 Jul 94 04:39:09 GMT

Hello all,
        People who have been checking will notice that
gcc-2.6.0.tar.gz was just put up on prep.ai.mit.edu I was bug testing
the alpha 2.5.90 versions and learned very quickly that the kernel
will not compile because of the following cases.

in include/linux/string.h

the lines with:
        register int __res __asm__("ax");
or similar should be changed to:
        register int __res;
or gcc 2.6.0 will barf in almost every file including string.h because
it thinks that the ax register has spilled every time the string
functions are called. Linus, are you reading this? If not I will send
him mail.

Later,
David S. Miller
davem@eden.rutgers.edu

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

Crossposted-To: comp.os.linux.help,xmp.software.linux
From: dirk@flop.escape.de (Dirk Sturzebecher)
Subject: mt tell not working
Date: Wed, 13 Jul 1994 11:44:50 GMT

Hi,
I got a WANGTEK 5525ES streamer which does suport SCSI-2, but does report
to be SCSI-1. With slackware 1.0.3 (0.99pl12) mt tell used to work, since
upgrade to kernel 0.99pl15 mt tell gives me a:

mt: invalid tape operation `tell'

I thought it had to do with the tape being not recognized as a SCSI-2 
device (it was not recognized in 0.99pl12 either), but my try to force
it to be a SCSI-2 device is or not the right solution or not correct ( I
did some changes in scsi.c to force SCSI-2)

Now, does anybody know this problem, know the solution and especially where
I have to start fixing it. Could it be that mt is broken, as the error message
seams to be a mt problem ? I thought i did not update mt. If someone has a mt
where mt tell is working maybe he/she could send me a copy of it.

Cheers,
        Dirk

-- 
Dirk Sturzebecher                               Tel&Fax : [+49] (0531) 337077
Huttenstr. 10                                   E-Mail  : dirk@flop.escape.de
38114 Braunschweig
Germany

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

From: ftlofaro@unlv.edu (Frank Lofaro)
Subject: Re: Is XFree86 3.X from X11R6 ?
Date: Thu, 14 Jul 94 21:14:57 GMT

In article <CsxInp.GC1@aib.com> hohndel@aib.com (Dirk Hohndel) writes:
>
>3.0 is out (part of X11R6) but almost unusable on Linux
>

Exactly what is wrong with it? I know it is beta code. I could deal with 
some weirdness or bugs. Does it crash all the time or something like that? 
Is it woefully incomplete, missing key features? Or is it dangerous? 
(e.g. will I run the risk of frying something by running it. I don't have a 
screwey RAMDAC, so the card is safe, but if Xfree 3.0 tells the card to output 
150kHz horizontal, my monitor will explode, and I won't be happy :|) Or can 
people use it to hack root? (since its suid root)



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

From: grante@reddwarf.rosemount.com ()
Subject: Re: Massive GCC error detected
Date: Thu, 14 Jul 1994 22:06:48 GMT

NetDog (cdent@honors.indiana.edu) wrote:
: >>>>> "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.

It'll be back.  You just managed to avoid the bad spot or access
pattern in the L2 cache, or in RAM.  Other people reported that the
problem only occurred when compiling certain versions of the kernel,
or with certain versions of gcc.  As far as I know, it still turned
out to be a hardware problem.

--
Grant Edwards                                 |Yow!  Loni Anderson's hair
Rosemount Inc.                                |should be LEGALIZED!!
                                              |
grante@rosemount.com                          |

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

From: grante@reddwarf.rosemount.com ()
Subject: Re: 1.1.28: SCSI disks inverted !
Date: Thu, 14 Jul 1994 22:14:20 GMT

Chris Adams (racerx@vespucci.iquest.com) wrote:

: 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.

Definitely.  My LILO configuration has three kernels configured:

  "old" is the one from before the one called "linux".
  "linux" is the one I'm using now.
  "new" is the one that I'm building/testing.

After testing "new" for a while, I shift em' all over one.

Always, always, always, have a kernel that you _know_ works.

I broke that rule once, and sweated for a day and a half before I got
back up and running again.

The easiest way to screw yourself is to move the kernel files around
and then forget to the LILO config, so that then next time you boot,
it tries to load Aunt Sally's recipe for peanut butter cookies.  I
usually add the LILO config command to the /usr/src/Linux Makefile.

--
Grant Edwards                                 |Yow!  Loni Anderson's hair
Rosemount Inc.                                |should be LEGALIZED!!
                                              |
grante@rosemount.com                          |

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


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