Subject: Linux-Development Digest #80
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, 9 Sep 93 20:32:37 EDT

Linux-Development Digest #80, Volume #1           Thu, 9 Sep 93 20:32:37 EDT

Contents:
  Re: Lockups. Where to look for the trouble. (Bill Henning)
  can't build pdksh-4.8 (Cameron L. Spitzer)
  Re: Status of the NET-2 port (Tor Arntsen)
  Re: Idea, donate a Pentium machine to Linus, anyone? (Kim C. Callis)
  Re: ** IDE Multiple Mode Patch ** (Mark Lord)
  Re: ** IDE Multiple Mode Patch ** (Mark Lord)
  Linux/NOVELL co-existence? ("Dan Newcombe")
  Re: ** IDE Multiple Mode Patch ** ident drive failure ** (Mark Lord)
  Re: ** IDE Multiple Mode Patch ** (Mark Lord)
  Re: ** IDE Multiple Mode Patch ** (Mark Lord)
  Linux on Atari? (Laurent Chemla)
  Re: ** IDE Multiple Mode Patch ** (Eric J. Schwertfeger)
  Re: Idea, donate a Pentium machine to Linus, anyone? (Pirate (Anthony Taylor))
  bug in XFree 1.3 (XDrawArc) (Joerg Czeranski)
  Re: ** IDE Multiple Mode Patch ** (Paul J. Brewer)

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

From: bhenning@vanbc.wimsey.com (Bill Henning)
Subject: Re: Lockups. Where to look for the trouble.
Date: 9 Sep 1993 12:49:24 -0700

Even after applying Linus's patches to 99pl12 (from SLS 1.03) I still
experience lockups under heavy disk/paging activity... for example
running a clean compile of a kernel in one xterm and starting up
xdvi on the FAQ in another xterm almost guarantees a hard crash -
drive LED stays on, system frozen.

Fortunately so far e2fsck always seems to fix the few problems that
showed up in the filesystem as the result of such lockups.

Hardware: Amd386/40+64k cache+Cyrix387+8Mb dram+170WD and 340WD, STB 4com
          mitsumi cd, tseng4k svga

Bill


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

From: cls@truffula.sj.ca.us (Cameron L. Spitzer)
Subject: can't build pdksh-4.8
Date: Thu, 9 Sep 1993 14:45:07 GMT


I downloaded the new pdksh-4.8 from sunsite.  There are no
linux-specific install or readme files, but there is a
makefile.linux in the sh subdirectory, which produces the
following error:

gcc -pipe -O2 -fomit-frame-pointer -s -D_SYSV -D_POSIX_TERM -DNO_MMAP -D_POSIX_SOURCE -I../std/h    -c version.c -o version.o
In file included from stdh.h:48, from version.c:9:
../std/h/stddef.h:33: redefinition of `size_t'
/usr/include/stdio.h:73: `size_t' previously declared here
In file included from /./usr/include/sys/types.h:4, from ../std/h/sys/types.h:11, from stdh.h:51, from version.c:9:
/usr/include/linux/types.h:26: redefinition of `ptrdiff_t'
../std/h/stddef.h:34: `ptrdiff_t' previously declared here
make: *** [version.o] Error 1

I've got gcc-2.4.5 and libc-4.4.1 on 0.99pl12, a great system that
compiles practically everything!
Was pdksh-4.8 made for an older system?
Have I done something dumb (not the first time :-) ?
Should I just comment out the extra defs?
Has anyone built pdksh-4.8 on this system?  What did you do?  TIA

Cameron
cls@truffula.sj.ca.us


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

From: tor@spacetec.no (Tor Arntsen)
Subject: Re: Status of the NET-2 port
Date: 9 Sep 1993 19:05:57 GMT
Reply-To: tor@spacetec.no

In article 747592922@pbhrzx.uni-paderborn.de, n62274@pbhrzx.uni-paderborn.de (Kl.Schaefers) writes:
> yeah, you are honest, however. There are several other people working
> on the so called net-2 stuff, but they grab bsd sources, and only
> remove the header :-) 

Okay, I can see the smiley.  But I still think you should qualify such a statement.

As far as I know Fred and the other guys working on the Linux net code release 2 are
writing *every* piece of the kernel code by themselves.

Not amused,
Tor 
---
+--------------------------------------------------------------------+
| Tor Arntsen   -  Spacetec A.S, N-9005 Tromso, Norway +47-83-84500  |
| LA1RHA - tor@spacetec.no (1) - tor@tss.no (2) - tarntsen@bbb.no (3)|
| Linux user                                                         |
| #include <disclaimer.h>                                            |
+--------------------------------------------------------------------+


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

From: callis@noc.usfca.edu (Kim C. Callis)
Subject: Re: Idea, donate a Pentium machine to Linus, anyone?
Date: 9 Sep 1993 11:15:54 GMT

Dag Asheim (dash@ifi.uio.no) wrote:
: >Anyone also comes out with such an idea? It comes into my mind
: >5 seconds ago so I just present it :-) 

: No, we don't want that.  Then he'll just make the system to slow to
: use on a 486.  But at least you didn't suggest to give him an Alpha.
: With a Pentium we would be binary compatible, even if it is a little
: on the slow side.  :-)

:                                       Dag

: PS:  Actually, if Linus would promise not to forget the days he had
: a 386, it might be a good idea.  He deserves it!

Of course this would never work, but just think if everyone who has
Linux going or anyone who used to use Linux donated a single $1 bill...
We probably would be able to get a brand new, nifty machine for Linus.
Just think, with a Pentium(tm) based machine, he could start designing
Linux: The Next Generation (coming soon to a theatre near you!).

I do realize that this is wistful thinking. Why would anyone in their
right mind want to donate a dollar for something that is available all
over the Internet for free? Nevermind that we are speaking of a mere
fraction of what it cost to get Un*x from SCO or ATT... Sorry, I just
thought that something would come out of such a statement.
--
*******************************************************************
* Kim C. Callis                                                   * 
* Univ. of San Francisco, San Francisco, CA                       *
* EMAIL: callis@usfca.edu or callis@dons.ac.usfca.edu             *
*******************************************************************
DISCLAIMER: As long as nothing I say attempts to represent the 
views of the Univ. of San Francisco, they could care less what I 
have to say. Meaning that anything said, like it or not, is strictly
my own personal opinion and views.


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

From: mlord@bnr.ca (Mark Lord)
Subject: Re: ** IDE Multiple Mode Patch **
Date: Thu, 9 Sep 93 20:08:27 GMT

In article <c9108932.747559444@peach.newcastle.edu.au> c9108932@peach.newcastle.edu.au writes:
>mlord@bnr.ca (Mark Lord) writes:
>
>>In article <1993Sep7.030025.25084@bmerh85.bnr.ca> mlord@bnr.ca writes:
>>>This patch (unofficial) can be applied to enable "multiple sector mode"
>>>operation in systems with IDE drives.  Virtually all recent IDE drives include
>
>yeah the identifcation biso tells me some stuff about my drive including:
>Model = Maxstor7120AT, firmware=305730
>bfftyope= dualportcache,  b-size=64k, max_mult_sec=16
>dblwordIO= yes,


Please email me the whole output.. I am assembling a database of results,
especially with drives like this and some older conner ones..
-- 
mlord@bnr.ca    Mark Lord       BNR Ottawa,Canada       613-763-7482

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

From: mlord@bnr.ca (Mark Lord)
Subject: Re: ** IDE Multiple Mode Patch **
Date: Thu, 9 Sep 93 20:06:30 GMT

In article <1993Sep8.124239.23222@hemlock.cray.com> lord@pecan.cray.com writes:
>
>This gave me about a 35% speed up using iozone (with 19 Mbyte files),
>but it also meant that I couldn't touch my modem and disk at the same
>time, it must be missing those interrupts like crazy. I experimented
>with MAX_NO_INTR all the way down to 2, the interrupt loss was still
>there (less frequently), but the throughput gain disappeared
>all together :-(, in fact dropping MAX_NO_INTR below 10 meant that
>most of the speed up disappeared.

I'm now recommending MAX_NO_INTR of 2 universally.  As for the "speedup",
the *only* reliable measure that seems to exist (for me) is with
        time dd if=/dev/hda2 of=/dev/null bs=1024k count=100

Where hda2 is any NON-root partition (to avoid cache hits on prior data).

iozone results vary widely from invocation to invocation, especially
when interspersed with the dd test.

*** OF greater importance than any IO speedup is the reduced kernel overhead,
*** which you should see regardless of MAX_NO_INTR.
-- 
mlord@bnr.ca    Mark Lord       BNR Ottawa,Canada       613-763-7482

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

From: NEWCOMBE@AA.csc.Peachnet.EDU ("Dan Newcombe" )
Subject: Linux/NOVELL co-existence?
Date: 9 Sep 1993 15:41:12 -0500

X-Mn-Key: dos

Here's my dilema...
    I am SICK of using DOS/Windows here at work.  The only thing I
    use windows for is the task switching.  Right now I am using
    CUTCP for logging into our Unix box.  Piece of cake (I actually
    like CUTCP.)  Anyway...

    The only thing that keeps me from totally switching over is the
    fact that to get my email, I have to access our Novell network.
    And every so often I need the network for other things too, but
    not much.  Has anyone been able to get any sort of access like this.
    I assume it'd have to be under DOSEMU, and I can live with that.
    Anyone?

      -Dan
--
Daniel A. Newcombe                                  Clayton State College
Computing Services                                  Morrow, GA 30260
E-Mail Address: newcombe@aa.csc.peachnet.edu        (404)-961-3421
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
"I thought maybe I've already gone crazy."
                            -Marillion,  "Blind Curve"

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

From: mlord@bnr.ca (Mark Lord)
Subject: Re: ** IDE Multiple Mode Patch ** ident drive failure **
Date: Thu, 9 Sep 93 20:14:19 GMT

In article <KEVINS.93Sep8221327@aragorn.ori.org> kevins@aragorn.ori.org writes:
>
>Running 99.12 kernel on a 386-40 w/ 8megs.  Two Conner IDE 
>drives.  One a 170 meg and other 109 meg.  System runs great
>been running linux since 99.6 without a problem.
>
>Patch applies and compiles clean, at boot, I get the message:
>
>hda: identify drive failed, ere_reg = 04
>hda: identify drive failed, ere_reg = 04

err_reg=04 means "Command Aborted", which implies "Command not supported".
But the identify-drive currently has nothing to do with multiple mode,
the two are attempted independantly.  But if you see no "mult mode enabled"
message, then your drive probably supports neither.
-- 
mlord@bnr.ca    Mark Lord       BNR Ottawa,Canada       613-763-7482

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

From: mlord@bnr.ca (Mark Lord)
Subject: Re: ** IDE Multiple Mode Patch **
Date: Thu, 9 Sep 93 20:11:53 GMT

In article <26mf4f$klq@uniwa.uwa.edu.au> oreillym@tartarus.uwa.edu.au writes:
>
>time dd if=/dev/hda of=/dev/null bs=16384 count=1024
>run 5 times (on an 8 meg machine), and takeing average of last 4 times
>gave  17.73 seconds.
>
>With multi-mode enabled (yes, it did show up at bootup), the average
>dropped to 17.55 seconds... 
>
>At the same time, %cpu time went from 25% w/o multi-mode, to 78%
>with... 
>
>Sorry, didn't check write speed.  Planning to do that in about 8 hrs
>when andrew finishes uploading
>       
>Mind you, the patches did improve thruput from 770K/s to 950K/s before
>the patches went in. That other stuff made a different, but the
>multi-sector mode doesn't seem to.

Yeah, I get slow results too, sometimes but not always. 

Something is strange in linux.. 
-- 
mlord@bnr.ca    Mark Lord       BNR Ottawa,Canada       613-763-7482

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

From: mlord@bnr.ca (Mark Lord)
Subject: Re: ** IDE Multiple Mode Patch **
Date: Thu, 9 Sep 93 20:09:20 GMT

In article <JCBURT.93Sep8101458@gats486.larc.nasa.gov> jcburt@gats486.larc.nasa.gov writes:
>
>Okay, I've got a slightly different problem...I have 2 IDE drives on
>one controller/interface board, one a WD2200 (like the one above -
>200MB WD Caviar drive) the other a WD2340 (340MB WD Caviar drive). I installed
>the patch given and had it display the drive info on bootup. The one
>problem I have is that the WD2200 can only handle 8-sector multiple mode
>while the WD2340 can handle 16-sector multiple mode...is there any way
>to set these individually? perhaps if/when you modify the patch, you 
>could have the ability to set each drive individually...

Coming soon.  Actually, I think a default of 8 might be better for many reasons.
-- 
mlord@bnr.ca    Mark Lord       BNR Ottawa,Canada       613-763-7482

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

From: laurent@brasil.frmug.fr.net (Laurent Chemla)
Subject: Linux on Atari?
Date: 9 Sep 1993 11:37:25 GMT

  Does someone works on such a port? Can we help?

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

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

From: maniac@unlv.edu (Eric J. Schwertfeger)
Subject: Re: ** IDE Multiple Mode Patch **
Date: Thu, 9 Sep 93 21:41:31 GMT

In article <c9108932.747559444@peach.newcastle.edu.au> c9108932@peach.newcastle.edu.au (Simon J Ferrett) writes:
>mlord@bnr.ca (Mark Lord) writes:
>
>>In article <1993Sep7.030025.25084@bmerh85.bnr.ca> mlord@bnr.ca writes:
>>>This patch (unofficial) can be applied to enable "multiple sector mode"
>>>operation in systems with IDE drives.  Virtually all recent IDE drives include
>
>yeah the identifcation biso tells me some stuff about my drive including:
>Model = Maxstor7120AT, firmware=305730
>bfftyope= dualportcache,  b-size=64k, max_mult_sec=16
>dblwordIO= yes,
>
>however when it goes to enable multisector
>hda: enabled 16 sector ...

My VLB IDE controller has a DOS driver that lets me set whether or not

it uses the multiple-mode commands, and it has a list of suggestions.
Included in this list:
        Maxtor 7120AT  Fast (no multi-sector commandss, same for all
                        Maxtor drives listed).
        WD              Turbo (multi-sector), for all drives listed
        Conner          Turbo, for all except CP3000
        Quantum         Fast for all.
        Seagate         Fast for ST3144A, no others listed, and I've
                        heard from one of my beta testers that the
                        ST3390A at least works in Turbo.

Hope this helps out.
-- 
Eric J. Schwertfeger, maniac@cs.unlv.edu

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

From: fnatt@elmer.alaska.edu (Pirate (Anthony Taylor))
Subject: Re: Idea, donate a Pentium machine to Linus, anyone?
Date: Thu, 9 Sep 1993 22:06:59 GMT

In article <26n39a$odv@noc.usfca.edu> callis@noc.usfca.edu (Kim C. Callis) writes:
>From: callis@noc.usfca.edu (Kim C. Callis)
>Subject: Re: Idea, donate a Pentium machine to Linus, anyone?
>Date: 9 Sep 1993 11:15:54 GMT
>Dag Asheim (dash@ifi.uio.no) wrote:
>: >Anyone also comes out with such an idea? It comes into my mind
>: >5 seconds ago so I just present it :-) 
>
>: No, we don't want that.  Then he'll just make the system to slow to
>: use on a 486.  But at least you didn't suggest to give him an Alpha.
>: With a Pentium we would be binary compatible, even if it is a little
>: on the slow side.  :-)
>
>:                                      Dag
>
>: PS:  Actually, if Linus would promise not to forget the days he had
>: a 386, it might be a good idea.  He deserves it!
>
>Of course this would never work, but just think if everyone who has
>Linux going or anyone who used to use Linux donated a single $1 bill...
>We probably would be able to get a brand new, nifty machine for Linus.
>Just think, with a Pentium(tm) based machine, he could start designing
>Linux: The Next Generation (coming soon to a theatre near you!).

No!  PowerPC!  It'll be out next year....  Blows the Pentium out of the 
water.

>
>I do realize that this is wistful thinking. Why would anyone in their
>right mind want to donate a dollar for something that is available all
>over the Internet for free? Nevermind that we are speaking of a mere
>fraction of what it cost to get Un*x from SCO or ATT... Sorry, I just
>thought that something would come out of such a statement.
>--

Send it to the Linus Torvalds fund.  Care of Linus himself.

Better yet, get him a six of beer.  Good stuff.  Watney's RB, Linus? :)

===============================================================
Pirate           (Anthony Taylor)        fnatt@elmer.alaska.edu
===============================================================
You wake up one day and find you won't go back to sleep.  -j.j.


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

From: czeranski@rz.tu-clausthal.de (Joerg Czeranski)
Subject: bug in XFree 1.3 (XDrawArc)
Date: Thu, 9 Sep 1993 21:46:23 GMT

I have found a bug in the XDrawArc-routine of XFree 1.3.
Whenever I draw an arc of some specific size (and position?),
the X server freezes - for 45 seconds on a 386-40, only a short time
on a 486-33.  Even the mouse pointer cannot be moved.
The problem appears on a Hercules card as well as on VGA.

This is a short (and dirty) program, that demonstrates the bug:

=================== bite here ===================
#include <X11/Xlib.h>

int main()
{
    Display *display;
    int screen, xid;
    display = XOpenDisplay("");
    screen = DefaultScreen(display);
    xid = XCreateWindow(display, RootWindow(display, screen),
        0, 0, 300, 170, 0, DefaultDepth(display, screen),
        InputOutput, CopyFromParent, 0, 0);
    XDrawArc(display, xid, XCreateGC(display, xid, 0, 0),
        -409, -1113, 3576, 2358, 0, 64 * 360);
    XFlush(display);
    return 0;
}
=================== bite here ===================

If is too fast, just add a for(i=0;i<20;i++) around the
XDrawArc :-).

BTW, I have the slight feeling, that the circle is not drawn exactly
with the specified width/height but maybe one pixel too wide.
(But I will have to test around a bit further.)

joerg


--
Joerg Czeranski                EMail czeranski@rz.tu-clausthal.de
Osteroeder Strasse 55          SMTP  injc@[139.174.2.10]
D38678 Clausthal-Zellerfeld    Voice (at work)  +49-5323-72-3896
Germany                        Voice (at home)  +49-5323-78858

To obtain PGP public key, finger injc@sun.rz.tu-clausthal.de, or email me.

At any given site for any given application or feature, there's someone
who knows more about it than the support staff.  Finding that person is
the first step to take to diagnose any given problem.

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

From: pjb@cco.caltech.edu (Paul J. Brewer)
Subject: Re: ** IDE Multiple Mode Patch **
Date: 9 Sep 1993 23:42:28 GMT

        Gives me about 25% improvement on a system wide "du" command.
My system consists of two linux partitions on a 345 meg new Maxtor
and two dos partitions mounted read only on a one year old 130 meg seagate.
The computer is a 386/40, with a cheap no-name $15 IDE card.

        Because the drive info for the maxtor said it could handle 32 sector
multimode, I changed the define in the patched hd.c from 16 to 32 (just the one
define at the top). Is that all I needed to do? It didnt seem to make much
difference going from 16 to 32.

time output for root "du" command, no other user activity. 

before patch

real       51.0
user        1.2
sys        22.2


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

patch enabled for 16 sectors

real       40.0
user        1.1
sys        21.4

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

patch enabled for 32 sectors

real       39.0
user        0.9
sys        20.9

=========================
kernel startup message follows
(with 32 sector patch enabled)

Console: colour EGA+ 80x25, 8 virtual consoles
Serial driver version 3.96 with no serial options enabled
tty00 at 0x03f8 (irq = 4) is a 16450
tty01 at 0x02f8 (irq = 3) is a 16450
tty02 at 0x03e8 (irq = 4) is a 16550A
lp_init: no lp devices found
Microsoft BusMouse detected and installed.
Memory: 7152k/8192k available (460k kernel code, 384k reserved, 196k data)
Floppy drive(s): fd0 is 1.2M, fd1 is 1.44M
floppy: FDC version 0x90
SLIP: version 0.7.5 (4 channels): OK
dl0: D-Link pocket adapter: probe failed at 0x378.
Calibrating delay loop.. ok - 7.29 BogoMips (tm)
Partition check:

+-------------------------------------------------------------------+
hda:  Drive identification info:
 Model=st3144AT, FwRev=REV184.0, SerialNo=00KL3823160000000000
 Config={ sHard NotMFM HdSw>15uSec Fixed DTR>5Mbs RotSpdTol>.5% }
 Default c/h/s=1001/15/17, TrkSize=9792, SectSize=576, ECCbytes=4
 BuffType=DualPortCache, BuffSize=32KB, MaxMultSect=0
 DblWordIO=no, LBA=no, DMA=no, tPIO=slow, tDMA=slow
 (maybe): Current c/h/s=0/0/0, TotSect=0, LBAsect=0
 (maybe): MultSect=0, DMA-1w=0000, DMA-mw=0000
+-------------------------------------------------------------------+

  hda: hda1 hda2 < hda5 >

+-------------------------------------------------------------------+
hdb:  Drive identification info:
 Model=Maxtor 7345 AT, FwRev=BADH0956, SerialNo=C2059M7S
 Config={ sHard NotMFM HdSw>15uSec Fixed DTR>5Mbs FmtGapReq }
 Default c/h/s=790/15/57, TrkSize=36366, SectSize=638, ECCbytes=11
 BuffType=DualPortCache, BuffSize=64KB, MaxMultSect=32
 DblWordIO=yes, LBA=no, DMA=no, tPIO=fast, tDMA=fast
 (valid): Current c/h/s=790/15/57, TotSect=1316618250, LBAsect=0
 (valid): MultSect=0, DMA-1w=0000, DMA-mw=0000
+-------------------------------------------------------------------+

  hdb: Enabled 32-sector multiple mode
  hdb: hdb1 < > hdb2 hdb3 hdb4
EXT2-fs warning: mounting unchecked file system, running e2fsck is recommended
[EXT II FS 0.3c, 93/08/05, bs=1024, fs=1024, gc=16, bpg=8192, ipg=1936]
VFS: Mounted root (ext2 filesystem).
Adding Swap: 8544k swap-space
[EXT II FS 0.3c, 93/08/05, bs=1024, fs=1024, gc=5, bpg=8192, ipg=1760]
VFS: Warning: time using old stat() call. Recompile your binary.
VFS: Warning: time using old stat() call. Recompile your binary.
VFS: Warning: time using old stat() call. Recompile your binary.

Paul Brewer
pjb@cco.caltech.edu




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


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