Files
oldlinux-files/ftp-archives/tsx-11.mit.edu/1996-10-07/mail-archive/linux-devel/Volume2/digest294
2024-02-19 00:24:15 -05:00

666 lines
23 KiB
Plaintext

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: Wed, 12 Oct 94 05:13:09 EDT
Subject: Linux-Development Digest #294
Linux-Development Digest #294, Volume #2 Wed, 12 Oct 94 05:13:09 EDT
Contents:
SERIOUS bug with GreenPC functions (Michael Bischoff)
DMA Time out under 1.1.45- and Voxware 2.90-2 (Brandon Long)
Re: Linux For Mac :( (Kevin K. Lewis)
Re: 1.1.53 - fdformat - IOCTL error still there (Rene COUGNENC)
SOCK_INODE(sk->socket)->i_uid sometimes wrong (Lars Fenneberg)
Physical Memory size (Cheryl Dematteis)
Re: Contract Software development : Driver for PDMA16 I/O board (David - Morris)
Re: Linux on a IBM PS/2 (H. Peter Anvin)
Re: Improving SLIP latency under Linux (John Richardson)
WANTED: Xerion bin or ported source (Greg Bruell)
Re: Filesystem idea (Doug Dejulio)
1.1.45+ doesn't compile: ld -qmagic? (Brian Mancuso)
Re: FTP slowdown under 1.1.52 with hdparm on (Mark D. Roth)
[H] missing "Close" item onSWiM motif menu (david her)
Re: A badly missed feature in gcc (Jeff Kesselman)
Re: Will pcsndrv be included in 1.2? (Rafal Maszkowski)
Re: Orchid CDS-3110 CD-ROM (Eberhard Moenkeberg)
Re: Orchid CDS-3110 CD-ROM (Eberhard Moenkeberg)
Re: Large IDE Drive support (Steve DuChene)
Re: Improving SLIP latency under Linux (John Richardson)
----------------------------------------------------------------------------
From: mbi@math.nat.tu-bs.de (Michael Bischoff)
Subject: SERIOUS bug with GreenPC functions
Date: 11 Oct 1994 22:01:26 GMT
Reply-To: mbi@mo.math.nat.tu-bs.de
Hello,
I just encountered a serious problem with the GreenPC functions of
my new motherboard:
I loose a partition, if I do the following:
1) set HDD Power Down to 1 Minute in the BIOS (SiS 471/ AWARD)
2) Boot Linux, all partitions on /dev/hda
3) mkfs.ext2 /dev/hdb2; mount /dev/hdb2 /mnt; cp .bashrc /mnt
4) work on hda until hdb is powered down (I always wondered how this
works with Linux)
5) shut down the system.
I get the message '/dev/hda: reset multiple mode to 0' when
hdb is powered up again and /dev/hdb2 is unmounted.
6) reboot. The first sectors of /dev/hdb2 will be garbage.
My configuration: Linux-1.1.49, 2 Conner HDs (IDE), drive hdb is a Conner
CP 30204. Both drives are initialized to 16 sectors multiple mode.
Motherboard: Sis 471, Award BIOS.
Can this be a hardware fault, or is it a software problem?
I don't know if this a problem with other operating systems too,
since I am using only Linux (and NEVER lost a partition in the last
2 years)
Desperately,
Michael
--
EOI
============================================================================
Michael Bischoff e-mail: mbi@mo.math.nat.tu-bs.de
Abt. Mathematische Optimierung or m.bischoff@tu-bs.de
Inst. Angewandte Mathematik or on.bischoff@zib-berlin.de
Pockelsstrasse 14 Tel. +49-531-391-7555, Fax: +49-531-391-4577
38106 Braunschweig Germany
------------------------------
From: Brandon Long <blong@uiuc.edu>
Subject: DMA Time out under 1.1.45- and Voxware 2.90-2
Date: 9 Oct 1994 01:22:16 GMT
Reply-To: blong@uiuc.edu
With kernel version greater than 1.1.45, all with Voxware 2.90-2,
(486/33, libc4.5.26, 16mb, PAS16)
I've been getting the following message after approx. 3 seconds of play:
Sound: DMA timed out - IRQ/DRQ config error?
/dev/dsp: I/O error
So, I started playing with my IRQ settings, figuring I just forgot whats
on what. No go. So I tried only using the pas16, without the soundblaster
emulator. Again, no go. So i booted an older kernel I had on my system
(1.1.16) and sound worked just fine. I tried 1.1.52 with the same sound
setup (dma/irq assigments, and only pas16) and it still errored out.
1.1.16 had voxware 2.5a (according to /dev/sndstat).
Obviously a lot has changed since 1.1.16-1.1.52 and 2.5a-2.90. Has the
PAS support been broken?
Thanks
Brandon
--
Brandon Long (N9WUC) "I think, therefore, I am confused." -- RAW
Computer Engineering Run Linux 1.1.xxx It's that Easy.
University of Illinois blong@uiuc.edu http://www.cen.uiuc.edu/~bl6456/
Don't worry, these aren't even my views.
------------------------------
From: lewikk@grasshopper.aud.alcatel.com (Kevin K. Lewis)
Subject: Re: Linux For Mac :(
Date: 09 Oct 1994 01:16:46 GMT
In article <376imr$92k@martha.utk.edu> williams@martha.utcc.utk.edu (Myles Williams) writes:
What's this? I thought I read in a newspaper that the Power Mac was going to
be an open system, fully documented by Apple.
Will they never learn?
It may be that Apple is opening the Mac, but Apple is a Hamlet of the
computer world, fickle 'til the end.
Also, to Apple "open" may mean that they'll license parts of their OS
to people so that they can build on it (thus ensuring Apple a cut).
I'm always suspicious when Apple makes any claims of openness or
freedom.
--
Kevin K. Lewis | My opinions may be unreasonable
lewikk@aud.alcatel.com | but such is the voice of inspiration
------------------------------
From: rene@renux.frmug.fr.net (Rene COUGNENC)
Subject: Re: 1.1.53 - fdformat - IOCTL error still there
Date: 11 Oct 1994 01:20:02 GMT
Reply-To: cougnenc@blaise.ibp.fr (Rene COUGNENC)
Ce brave Vincent Fatica ecrit:
> Very shortly after patch52 came out, a few persons observed an IOCTL error
> when using fdformat. I figured it would be better in 1.1.53,
For me 1.1.53 does not boot at all.
The kernel hangs after the CSLIP message, probably when detecting
the ethernet card (ne 2000).
Back to pl52...
--
linux linux linux linux -[ cougnenc@renux.frmug.fr.net ]- linux linux linux
------------------------------
From: lf@gimli.comlink.de (Lars Fenneberg)
Subject: SOCK_INODE(sk->socket)->i_uid sometimes wrong
Date: 9 Oct 1994 22:10:26 +0100
Hi!
This happens with Linux 1.1.49 and 1.1.51.
I am an author of an ip accounting software and i have
noticed that the value of SOCK_INODE(sk->socket)->i_uid
is sometimes wrong. This can be seen via netstat (which states
the wrong user) or via wrong accounting information.
The problem only seems to be triggered if there is a lot
of traffic on a specific socket (for example an ftp-data
connection) _and_ netstat respectively my accounting programm
is called. I have put some printks in the relevant places
which print out the user id and the pointer to the inode.
This pointer gets corrupted as it seems. It's value
suddenly changes if netstat or my accounting programm is
called. How can this happen? The only thing, that netstat and
my programm do, is that they read the value of
SOCK_INODE(sk->socket)->i_uid (netstat via access to /proc/net/tcp).
Any help is greatly appreciated!
Regards,
Lars.
--
Lars Fenneberg, lf@gimli.comlink.de, Tel.: 040/5244224
ftz Hamburg eV, mailbox: CL-HH - 040 2500890/2509692/2509520
------------------------------
From: cdematt@corsa.ucr.EDU (Cheryl Dematteis)
Subject: Physical Memory size
Date: 12 Oct 1994 01:53:22 GMT
Is there a system call which returns the number of pages of physical memory a
system has ?
In Solaris, the system call sysconf(_SC_PHYS_PAGES) gives me the number I'm
looking for. Linux (1.0.9) does not recognize _SC_PHYS_PAGES.
Thanks for the help.
--
Cheryl K. DeMatteis cdematt@cs.ucr.edu
Graduate Student (909)787-3432
Department of Computer Science
University of California
Riverside, CA 92521
------------------------------
From: dwm@shell.portal.com (David - Morris)
Subject: Re: Contract Software development : Driver for PDMA16 I/O board
Date: 12 Oct 1994 05:59:46 GMT
dwm@shell.portal.com (David - Morris) writes:
>I can't believe you want someone outside of australia for the work,
>but if you do, continue reading.
BLUSH !!! 'tis late at night and the r and f are so close
together... to all whose cyber space I have wasted, please
accept my humble appologies and serious chagrin. This reply
of course should never have been sent as a followup!
Dave Morris
------------------------------
From: hpa@ahab.eecs.nwu.edu (H. Peter Anvin)
Subject: Re: Linux on a IBM PS/2
Reply-To: hpa@nwu.edu (H. Peter Anvin)
Date: Mon, 10 Oct 1994 12:41:04 GMT
Followup to: <37a122$lkn@nova.umd.edu>
By author: cirigara@nova.umd.edu (Carlos Irigaray)
In newsgroup: comp.os.linux.development
>
> Hi, does anyone know how to create a "boot disk" and a "root disk" as in
> the Slackware distribution?
> I'm using Slackware 2.0.1 and my runnning kernel is 1.1.52 (I've compiled
> it). What I need is to make those diskettes from my system because then I
> should be able to have my IBM PS/2 booting. (the new kernel support the
> MCA architecture!)
>
> Thanks for the help!
>
The root disk doesn't change -- just use the standard one.
Making a Slackware bootdisk is a bit tricky; an easier way to get a
bootable disk is just to dd it to a floppy disk (dd if=file
of=/dev/fd0 bs=9216), or getting SYSLINUX
from eecs.nwu.edu:/pub/linux/syslinux/syslx101.zip.
/hpa
--
INTERNET: hpa@nwu.edu --- Allah'u'abha ---
IBM MAIL: I0050052 at IBMMAIL HAM RADIO: N9ITP or SM4TKN
FIDONET: 1:115/511 or 1:115/512 STORMNET: 181:294/1 or 181:294/101
Have you hugged your Swede today?
------------------------------
From: jrichard@cs.uml.edu (John Richardson)
Subject: Re: Improving SLIP latency under Linux
Date: 9 Oct 1994 01:38:49 GMT
In article <1994Oct8.053207.22798@unlv.edu>,
Frank Lofaro <ftlofaro@unlv.edu> wrote:
>Hmm, I thought Nagle's algorithm was mostly to cluster together
>outgoing data with ACK and window size change data (2 or 3 packets
>become 1).
Nagle's algorithm says that there can only be one outstanding small
segment that has not been acked. So however long it takes for
an ack to get back, that is how much data you can get (well,
up to a point I'm sure). The point of Nagle's alg is to get
more data across the link in less packets by reducing the number of
1 byte packets. ACKs usually get picked up on the characters'
echo.
>
>Even if larger packet coalescing occurs, it is not a problem really
>though. It is okay if a screen refresh goes a bit slowly. (not too
>slowly though). What _IS_ critical is the very small packets, like
>tty echo and editor echoing (e.g. in emacs). If that has too high
>a latency, then editing and input becomes uncomfortable.
>
I agree, but of course, now who's going to write this? :)
--
John Richardson
jrichard@cs.uml.edu
jrichar
------------------------------
From: gbruell@wellfleet.com (Greg Bruell)
Subject: WANTED: Xerion bin or ported source
Date: 12 Oct 1994 02:58:20 GMT
I know this has been posted before but I couldn't remember
when. If anyone has a working linux port of xerion binaries
or source I would greatly appreciate them. This seems like
a good thing to put on sunsite. I'm sure it would be
accepted into Incoming and I don't think it would break
any copying agreements. Thanks.
Gregory Bruell
Wellfleet Communications, Inc.
------------------------------
From: ddj+@pitt.edu (Doug Dejulio)
Subject: Re: Filesystem idea
Date: 12 Oct 1994 03:18:01 GMT
In article <1f.10483.1566.0NC9C259@compart.fi>,
Riku Saikkonen <riku.saikkonen@compart.fi> wrote:
>Got this odd idea the other day...
>
>Some 'temporary file' capability for the filesystems might be a good
>thing. For example, with an oft-compiled largish application, it's good
>to keep the .o files in the directory(/-ies). But they fill up disk
>space... What I'm suggesting is some sort of 'temp file' attribute that
>would keep the file on disk until space runs low. Then, when the disk
>space is almost 0, it would start deleting the temp files to free space.
This can be done as a daemon, without modifying any filesystem code.
Write a daemon that reads a configuration file to figure out "low
space" thresholds for each partition, and "temporary file extensions",
possibly also partition dependant. Possibly also put in a "degree of
temporary-ness" for each file extension -- eg. "keep *.o files for a
week, keep *~ files for three days (so they survive weekends), but
nuke #*# files in just one day".
Every night, run a cron job to build a table of existing "temporary
files", with their last modification date.
When a given partition goes below the specified amount of free space,
the daemon "wakes up". It goes through the table of existing files,
looking for the "most deletable" (ie. oldest that have gone beyond
their minimum time threshhold) ones. It checks them to see if they've
changed since last night's cron job, and if not, nukes 'em. Iterate
until you've got enough free space.
Who would want to install such a thing on their systems? Show me
there's enough demand and I may write it.
--
Doug DeJulio
ddj+@pitt.edu
http://www.pitt.edu/~ddj/
------------------------------
From: brianm@csa.bu.edu (Brian Mancuso)
Subject: 1.1.45+ doesn't compile: ld -qmagic?
Date: 12 Oct 1994 03:12:17 GMT
`make` terminates in arch/i386/Makefile with error
ld: unrecognized option `-qmagic'
etc.
This linker arguement was not present in previous kernel versions. I have
binutils-1.0 which should be the latest. -q is not in the manpages of
SunOS, News-OS, Solaris or AIX.
What's up with '-qmagic'? Do I have the right ld?
Thanks for any info, sorry for sqandering bandwidth.
Brian Mancuso
brianm@bu.edu
------------------------------
From: roth@ux4.cso.uiuc.edu (Mark D. Roth)
Subject: Re: FTP slowdown under 1.1.52 with hdparm on
Date: 12 Oct 1994 03:22:46 GMT
gnielsen@clam.rutgers.edu (Garth C. Nielsen) writes:
> Not that I am complaining or anything. I just wanted to state that
>I had a drop in transfer rate on my FTP transfers while using hdparm.
>I am using 1.1.52 and running a SLIP line on a 14.4 modem. Normally I
>get about 1.4 K/sec. But while running it with hdparm -m 32 the
>ftp transfer stopped for a few seconds after each disk write. After
>I set the hdparm -m0 then the ftp ran fine again. Any explainations?
I am not sure that hdparm is as good as it claims to be. I was
fooling around with it, and my results were not encouraging.
I tried setting my multcount to 16, which is the max value for my disk
according the the kernel at bootup. I also tried 2, 4, and 8. My
benchmark was the suggested "dd if=/dev/hda of=/dev/null count=32768
bs=1024". When the multcount was set to anything but 0, system and
user time went down, but wall-clock time went up. This was under
kernels 1.1.51 and 1.1.52.
Here are some figures I got, which I found to be fairly consistant:
multcount user system clock %CPU
0 0.21 20.69 0:46.74 44.7
2 0.34 21.53 0:47.69 45.8
4 0.37 18.75 0:49.94 38.2
8 0.37 11.90 1:10.54 17.3
16 0.29 12.37 1:10.55 17.9
I'd rather have the wall-clock time, personally... :)
--
roth@uiuc.edu | Mark D. Roth | http://www.cen.uiuc.edu/~mr4342/
(GEEK CODE 2.1) GCS d-- H+ s++:- g+ p1>4+ !au a-- w++@ v-(*)
C++>$ UL+>++++ 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?
------------------------------
From: davidher@netcom.com (david her)
Subject: [H] missing "Close" item onSWiM motif menu
Date: Wed, 12 Oct 1994 05:28:23 GMT
I have SWiM 1.2.4 on Slackware 2.0. everything seems work fine, but the
window default menu (run mwm) isn't supposely has items 'Restore' 'Minimize'
'Maximize' 'Size' 'Move' 'Lower' 'Close' ? Mine dose NOT has'Restore''Lower'or
'Close'. It is very troublesome whenever to close a window. Do I need to
fix somthing to make them shown ? Please help ! Thanks in advance.
regular menu My window menu (Alt+F4, Alt+F5, Alt+F3 don't function)
+-----------------+ +-----------------+
| Restore Alt+F5 | | |
| Move Alt+F7 | | Move Alt+F7 |
| Size Alt+F8 | | Size Alt+F8 |
| Minimize Alt+F9 | | Minimize Alt+F9 |
| Maximize Alt+F10| | Maximize Alt+F10|
| Lower Alt+F3 | + ----------------+
| Close Alt+F4 |
+-----------------+
------------------------------
From: jeffpk@netcom.com (Jeff Kesselman)
Subject: Re: A badly missed feature in gcc
Date: Sun, 9 Oct 1994 08:43:27 GMT
In article <373vcn$1da@nz12.rz.uni-karlsruhe.de>,
Thomas Koenig <Thomas.Koenig@ciw.uni-karlsruhe.de> wrote:
>David Taylor (ddt@idcube.idsoftware.com) wrote in comp.os.linux.development,
> article <9409231051.AA08511@idcube.idsoftware.com>:
>>I wish gcc for Linux could handle // comments.
>
>I don't ;-)
>
>This would break perfectly correct C code, like
>
> a = b//* Comment here */ c;
>--
And this is the cannonical case of why c++ is NOT a true super-set of
ANSI c. (Thanks, I'm going to save the exampel for the next time I have
THAT argument.)
Question, though, if gcc is fed a c++ file, will it accept that (a .C
extension, or one of the other gcc/g++ blessed c++ extensions)?
Seems to me it should. The answer to the poster woudl then be 'Youa re
trying to write c++ code. Stop telling the compiler its C code, and all
will be fine...'
Jeff Kesselman
------------------------------
From: rzm@dain.oso.chalmers.se (Rafal Maszkowski)
Subject: Re: Will pcsndrv be included in 1.2?
Date: 11 Oct 1994 19:32:20 GMT
Bryan Wright (bryan@elvis.phys.virginia.edu) wrote:
> Hi Folks,
> Just a quick question: Does anyone know if pcsndrv will be included
> in kernel version 1.2 when it comes out?
I think that it won't. But it is supposed to be in 1.3.x somewhere.
R.
--
Rafal Maszkowski rzm@oso.chalmers.se http://www.mat.uni.torun.pl/~rzm
Opinia publiczna powinna byc zaalarmowana swoim nieistnieniem - St. J. Lec
------------------------------
Date: Tue, 11 Oct 1994 05:50:35 +0100
From: Eberhard_Moenkeberg@p27.rollo.central.de (Eberhard Moenkeberg)
Subject: Re: Orchid CDS-3110 CD-ROM
Hello Jeff Kesselman and all others,
on 08.10.94 Jeff Kesselman wrote to All in USENET.COMP.OS.LINUX.DEVELOPMENT:
>> I've recently purchased a pentium computer from Comtrade that has a brand
>> new Orchid CDS-3100 CD-ROM on it.
JK> I've got bad news for you, I'm afraid. I have one of these too, which
JK> I've instaleld in various op systems and at thsi poitn know ALOT about...
JK>
JK> 1) What you REALLY have is a Wearnes CD-ROM and (if you are using the
JK> orchid interface) a Wernes sony act-alike interface. The drive is
JK> ELCTRONICLY compatable with the sony interface BUT
JK>
JK> 2) The Wearnes is NOT compatable on a software level with the sony
JK> drives. It uses a different control protocol. Any driver written for
JK> the sony drives will NOT work with this drive. You need a Wearnes
JK> specific driver.
JK>
JK> 3) As far as I know (and this may be outdated info) nobody has written a
JK> Wearnes driver for Linux.
I will not write a Linux driver for it - but if someone will mail me
the DOS driver, I can easily "disclose" all information a driver
writer needs. I have practice with that, and doing that is legal here. ;-)
Greetings ... Eberhard
------------------------------
Date: Tue, 11 Oct 1994 05:55:12 +0100
From: Eberhard_Moenkeberg@p27.rollo.central.de (Eberhard Moenkeberg)
Subject: Re: Orchid CDS-3110 CD-ROM
Hello Jeff Kesselman and all others,
on 09.10.94 Jeff Kesselman wrote to All in USENET.COMP.OS.LINUX.DEVELOPMENT:
JK> As i have been careful to point otu many times, i have never
JK> nor do I currently work for Yygdrasil.
Why not? ;-)
Yggdrasil has done lots of good things for the Linux community, and
one of the best is the "idea" to hire programmers and to put the result
under the GPL.
Greetings ... Eberhard
------------------------------
From: s0017210@unix1.cc.ysu.edu (Steve DuChene)
Subject: Re: Large IDE Drive support
Date: 8 Oct 1994 21:43:05 -0400
Dave: The difficulties with EIDE drives has been taken care of
in the newer ALPHA kernels (since about 1.1.39?). I have been
running 1.1.51 for about a month now and I have had no problems
with my EIDE harddrive in my Gateway P4D-100 system.
BTW, the return address in your posting with a system name of
bettlebrox doesn't work and caused bounced mail on my system.
If you want replies to your questions in the future you may
want to fix it as some people get irritated by bounced mail
when they are trying to reply to a plea for help.
--
| Steven A. DuChene sduchene@cis.ysu.edu or s0017210@cc.ysu.edu
| Youngstown State University | Computer Science / Math / Mech. Eng.
|-------------------------------------------------------------------
| Friends don't let friends do DOS
------------------------------
From: jrichard@cs.uml.edu (John Richardson)
Subject: Re: Improving SLIP latency under Linux
Date: 9 Oct 1994 01:56:10 GMT
In article <374crv$6mp@gate.noris.de>,
Matthias Urlichs <urlichs@smurf.noris.de> wrote:
>Why would you want to turn off the receive buffer?
I was guessing that the receive buffer was acting as a router that
didn't have TOS since 2k of packets could get queued up in there.
However, I've heard that some other folks without buffered modems
have the same problem, so that's probably not it.
>
>The reasonable thing to do here is to increase the baud rate you use when
>talking to the modem. Remember that with error-corecting modems, every data
>packet (which has no relation to the IP packets it carries) has to arrive
>completely before being forwarded because the checksum must be verified,
>and on a serial line there's no "Oops, discard the last fifteen characters"
>command.
Increase the baud rate? Huh? How? It is already at spd_hi. With
round trip times of 2000 to 3000ms I don't think I'm just waiting for
30 characters... I don't know if you realize this, this thread
was started by someone complaining about the abnormal interactive
responce times /during background ftp transfers/. If you use
SLIP or PPP can you say that you don't get 2-3 sec delays (you
can modify ping to get this) for interactive traffic.
If your slip server has a queue of ftp packets to send you, it should
be thowing your interactive packet at the front of the queue.
Or if you are sending and you have a full buffer of interactive
packets, your interactive packet should go to the front of your
queue.
>
>If you need less latency, then
>1- implement priority queueing
This is done! In 1.1.52, check out net/inet/dev.c and ip.c.
>2- use IP header compression
I don't have this option.
>3- turn off error correction in the modems
I tried this to no avail.
>4- use a smaller MTU
I've tried down to 100.
>
>If you use SLIP instead of PPP, decide on one of -3- and -4-.
>
--
John Richardson
jrichard@cs.uml.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
******************************