Files
oldlinux-files/ftp-archives/tsx-11.mit.edu/1996-10-07/mail-archive/linux-devel/Volume2/digest120
2024-02-19 00:24:15 -05:00

791 lines
30 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: Mon, 5 Sep 94 00:13:04 EDT
Subject: Linux-Development Digest #120
Linux-Development Digest #120, Volume #2 Mon, 5 Sep 94 00:13:04 EDT
Contents:
Kernel "ro" Parameter (Andrew J. Piziali)
my ppp dies (Dorwin Shields)
Re: Got the bastard! [was re:fs corruption] (Beverly J. Brown)
Q:switching VCs under svgalib (Ross J. Reedstrom)
Help with ftp failure (tcpip) (Douglas H. Hirsh)
DOSEMU 0.53 notes (Mark Krischer)
missing headers (Douglas Frank)
Re: Q:switching VCs under svgalib (Ross J. Reedstrom)
Re: Future of Linux (John Saunders)
Re: Wheres blkdev.h?? (compiling 1.1.49) (Mark Lord)
Re: Strange kernel version information (Ramon Martinez)
Re: time speeds up (David Monro)
ipfirewall for linux, Anyone out there care to try it out? (Bob Beck)
Re: IDE Performance enhancement (Pete Deuel)
Re: 1.1.49 trouble (Goncal Badenes)
Re: IDE Performance enhancement (Harry C Pulley)
Re: polled ports (Harry C Pulley)
which kernel with dosemu 0.53 ? (Srini Seetharam)
Re: DOSEMU 0.53 notes (Harry C Pulley)
Re: linux never swaps ? (Uwe Bonnes)
Re: linux never swaps ? (Uwe Bonnes)
Re: Linux - my first impressions (Chris Bitmead)
----------------------------------------------------------------------------
From: andy@tidmmpl.csc.ti.com (Andrew J. Piziali)
Subject: Kernel "ro" Parameter
Date: 1 Sep 1994 17:28:42 GMT
I am in the process of configuring a Linux box to run in a
networked environment. A few days ago I came across an error
message during boot and started investigating it. The details
are listed below:
kernel: 1.1.12
lilo: 0.14
SysVinit: 2.4
Problem:
Use lilo to setup boot partition, mounting root as "ro" (lilo
boots from partition boot sector, not MBR of disk). When
booting, init (SysVinit) reports error that "ro" is an invalid option.
The "ro" flag should not be passed to init, it should be
stripped from the command line by the kernel (or lilo?).
Fix:
Linux source, subdirectory init, file main.c, function
parse_options:
while ((line = next) != NULL) {
NEW--> int PassToInit = 1;
...
}
} else if (!strcmp(line,"ro")) {
root_mountflags |= MS_RDONLY;
NEW--> PassToInit = 0;
} else if (!strcmp(line,"rw")) {
root_mountflags &= ~MS_RDONLY;
NEW--> PassToInit = 0;
} else if (!strcmp(line,"debug"))
...
NEW--> if (PassToInit)
/*
* Then check if it's an environment variable or
* an option.
*/
If this problem has already been fixed, please let me know where I can
get the "official" patch.
Thanks!
--
Andy Piziali |
________------+------________
work: andy@daldd.sc.ti.com / \
home: andy@piziali.lonestar.org *---*
------------------------------
From: parprods@exmidway.ecn.uoknor.edu (Dorwin Shields)
Subject: my ppp dies
Date: 4 Sep 1994 22:52:44 GMT
Can anyone suggests a reason why my ppp dies? I upgraded to the lates 082994 driver because the one in the
kernel (1.1.49) didn't work either. I've used debug to look at what happens -- I get a connection--and it sends
a few requests and gets a few replies for things like mtu--then the linux end sends some requests for ids? maybe,--
lcp requests and then pppd dies with a sighup --- is the kernel killing pppd?--I've also noticed my serial communications
under linux don't seem to be too good--zmodem and term give lots of errors--I'm using a 14.4 modem and term just
told me I was getting 544 CPS can anyone help me?
Dorwin
------------------------------
From: bjb@shore.net (Beverly J. Brown)
Subject: Re: Got the bastard! [was re:fs corruption]
Date: 4 Sep 1994 19:05:09 -0400
Reply-To: bjb@shore.net
In article <344nm1$guk@styx.uwa.edu.au>,
cjcason@yarrow.wt.uwa.edu.au (Christopher Cason) wrote:
[description of fs corruption problem]
>
> Hardware/Software.
>
> The problems occurred on two totally separate machines.
>
> 486DX/50
> 20mB
> Buslogic BT-545S SCSI controller, I/O 330, IRQ 11, DMA 5
> Quantum 1080 SCSI drive.
> Maxtor IDE drive (root).
> SB 1.0 sound card (problem happened with it in or out.)
> Trident 8900 w/512k.
> WD8013 at 280 or 240, IRQ 15 or 7, either way. RAM at d0000.
^^^^^
> Kernel 1.0.8 _or_ 1.1.18, either way. same problem.
>
> 486DX2/66
> 16mB
> Adaptec 1542 SCSI controller, I/O 330, IRQ 15, DMA 5
> Conner 1380 SCSI drive.
> with or without Maxtor IDE drive mapped as root.
> PAS16 clone sound card (problem happened with it in or out.)
> Trident 8900 w/512k.
> WD8013 at 280, IRQ 11. RAM at d0000.
^^^^^^
> Kernel 1.0.8 _or_ 1.1.18, either way. same problem.
>
What is the base address of your SCSI controller BIOS?
Beverly J. Brown
bjb@shore.net
beverly@datacube.com
------------------------------
From: rjr@pharmacy.wisc.edu (Ross J. Reedstrom)
Subject: Q:switching VCs under svgalib
Date: Sun, 4 Sep 1994 18:14:14
Hello Linux hackers -
I know this isn't a Linux development question, but it _is_ a Linux specific
programming question, as svgalib is linux specific -
Anyway - I've been fooling around with Jan Kybic's GhostScript front end
BMV, mainly cleaning up the user interface a bit (personal prefs). However,
this is the only svgalib program I have that won't allow me to switch VCs.
I've checked out the sources for zgv, a graphics viewing prog. and it doen't
seem to be doing anything special in order to support VC switching, and bmv
doesn't seem to be doing anything special that would _disable_ switching, so...
BTW, this is with svgalib 1.12 (shared libvga 1.1.8)
Any help out there?
Thanks,
Ross J. Reedstrom
rjr@pharmacy.wisc.edu
------------------------------
From: dhirsh@zilker.net (Douglas H. Hirsh)
Subject: Help with ftp failure (tcpip)
Date: 4 Sep 1994 23:21:00 GMT
I have linux kernel version 1.1.27 and am running a dedicated ppp dial up
line at the site that I connect to they just changed the terminal server that I connect through and I developed a problem when I try to telnet to the machine there at the site which is a sun machine. I don not have a problem connecting to other ftp sites on the internet just this one. when I connect everything appears to work but if I attempt to ftp file from the sun machine I get an message on the console on my machine that reads
Data wakeup
skb1=00049410 :skb1->h.th->seq = 1596480001: skb->h.th->seq =1596480261
copied_seq - 1596480000 acked_seq = 1596480261
tcp_ack: seq ace6df29 ack 5f285b06
data wakeup
cleaning rbuf for sk=00049a04
sk->rspace = 8192, was 8192
cleaning rbuf for sk-00049a04
sk->rspace = 8192, was 8192
This is just a piece of the code that came accross and I hope it is enought
after this occurs the ftp stops usally with 0 bytes butsometimes with more
and the system never returns or fails the transfer.
I hope that this make sense to some one and that they can help
Doug Hirsh
dhirsh@hirsh.com
dhirsh@zilker.net
------------------------------
From: mkrisch@avalanche.mpce.mq.edu.au (Mark Krischer)
Subject: DOSEMU 0.53 notes
Date: 4 Sep 1994 23:05:50 GMT
i've started playing around with DOSEMU .53 and i am VERY impressed.
the idea of being able to have a real word processor (WP 6.0 DOS) and not have to
be in DOS/Windows is absolutely great!!!
just a couple of notes. things i've noticed and problems i've had
mouse doesn't seem to work to well--do i need to run the DOS driver? when i do
i'm told its already loaded.
is there support for the ATI mach32 chip? i can only get 640x480 16 colors in
graphics mode for WP--though the fact i can get graphics mode at all is great! the
screen goes screwy if i try 256 colors. and if i try a different mode, WP tells me
it won't work with my hardware config (it does work in native DOS).
also, what do i need to do to get color in text mode--i got it once and i haven't a
clue what i did, but i never saw it again.
i do get color text in an xterm though. that has LOTS of potential. it does seem
a lot slower though. also delete and bs wouldn't work, but that might have been
my termcaps not set up write in the dosemu.config or something like that.
interesting--MS-UNDELETE won't work, it says i'm on a network drive and must use
delete sentry--that's the one where it just copies deleted files to a hidden dir.
lastly, can someone give me some information on good and SAFE ways to redirect boot to
my DOS partition. i also use double-space, so i'd be interested in knowing how to
gain access to that--the HOWTO is not completely clear and i'd really prefer not to
thrash my hard drive, it's young and i want to break it in slowly.
all in all, great stuff!!
mark
------------------------------
From: frankdn@mv.mv.com (Douglas Frank)
Subject: missing headers
Date: Sun, 4 Sep 1994 23:40:42 GMT
I need some help from someone who has a copy of W.R. Stevens'
"Advanced Programming in the Unix Environemnt" (Addison Wesley
1992, ISBN 0-201-56317-7) - a great book, btw.
This book contains tons of exercises, and you can get the source
from ftp.uu.net (look for published/books/stevens*).
Anyway, many of the example sources include a couple of headers
that aren't in my Linux distribution (Slackware 2.0). They
are: stddef.h & stdarg.h. There are also other headers Stevens
says are common, but I don't have: cpio.h, float.h, langinfo.h,
& search.h.
My question is, was leaving these out a mistake? If not, what
are the substitues? If no substitutes, then why does Voelkerding
think we don't need them?
Thanks
--Doug
frankdn@mv.mv.com
------------------------------
From: ross@localhost (Ross J. Reedstrom)
Subject: Re: Q:switching VCs under svgalib
Date: 5 Sep 1994 00:12:32 GMT
Oops!
Sorry to follow up my own post, but of course, five minutes after posting
I solved the problem, after perusing the svgalib sources a little more
closely.
It turns out the bmv spawns a background process to monitor the keyboard
while the parent is doing the display , and the two processes use SIGUSR1
to communicate - svgalib uses SIGUSR1 and SIGUSR2 to indicate VC aquire and
release! So, bmv's sighandler on SIGUSR1 was getting in the way. I removed
the background reading process ( I'm not sure why it was there anyway) and
voila! VC switching!
Ross J. Reedstrom
rjr@pharmacy.wisc.edu
------------------------------
From: john@odin.apana.org.au (John Saunders)
Subject: Re: Future of Linux
Date: Sat, 3 Sep 1994 04:53:46 GMT
With every body talking the future of Linux, how about this idea.
We use the Mach microkernel as a base and implement Linux on top of it.
The new name will be MacLinux and it comes with a free burger!
--
-- . +----------------------------------------------+
,-._|\ | John SAUNDERS - Home john@odin.apana.org.au |
/ OZ \ | - Work johns@rd.scitec.com.au |
\_.-\__/ | "Mmmmmmmm beer..." - Homer Simpson |
v +----------------------------------------------+
------------------------------
From: mlord@bnr.ca (Mark Lord)
Crossposted-To: comp.os.linux.help
Subject: Re: Wheres blkdev.h?? (compiling 1.1.49)
Date: 5 Sep 1994 00:56:14 GMT
In article <CvL0JI.G2F@dorsai.org> Carlos Dominguez writes:
>I'm trying to compile the latest/greatest kernel in order to
>get support for my 1mb/sec QIC80/floppy controller.
>
>I got the 1.1.45 kernel, applied all the patches sequentially from
>46 to 49 to my 45 source tree, and whenever I do a make dep I always
>get this.
>
>ksyms.c:13: linux/blkdev.h: No such file or directory
Look in /usr/src for blkdev.h.. found it? If so, start over
and use "patch -p0" instead of just "patch".
--
mlord@bnr.ca Mark Lord BNR Ottawa,Canada 613-763-7482
------------------------------
Crossposted-To: comp.os.linux.help
From: rampa@encomix.com (Ramon Martinez)
Subject: Re: Strange kernel version information
Date: Sun, 4 Sep 1994 23:28:47 GMT
:>Just to avoid misunderstandings; yes, I did a complete rebuild between each
:>patch (make dep && make clean && make zlilo), and got new /vmlinuz images
:>from each build. Makes it even more strange, yes ? :-)
What your lilo.conf file says?
-ramon-
------------------------------
From: davidm@syd.dms.CSIRO.AU (David Monro)
Subject: Re: time speeds up
Date: Mon, 5 Sep 1994 00:41:45 GMT
Yeah, I get this too. I always assumed it was a hardware problem, since
I don't run the machine under DOS, but recently my mother has been using
the machine under DOS and it doesn't seem to happen there. It was about
45mins fast after about 5 hours use the other night too - gave me a real
fright. The problem seems to have got worse a few kernels ago.
I manually fix the clock (and yes, I do write it back to the CMOS clock
as well) whenever I boot the machine.
Nothing special about the machine -
486DX33, ISA, OPTI chipset, 8MB RAM
2xIDE controllers, 3 drives (Segate 130Mb, Conner 170Mb, WD 540Mb).
1xfloppy controller, 2 drives (1.2, 1.44)
4x16540 serial
2xparallel
SB16 sound
3c501 ethernet
Running Linux v1.1.30, patched for 2xIDE controllers with performance
patches.
Anybody figure this one?
David
------------------------------
From: beck@cs.ualberta.ca (Bob Beck)
Subject: ipfirewall for linux, Anyone out there care to try it out?
Date: 5 Sep 1994 01:16:32 GMT
Any experienced people out there willing to test out
ipfirewall? Ipfirewall is a package with a kernel modification and an
application to allow you to do packet level filtering/firewalling with
your linux box. I've completed a port of this which was originally
written for BSD/386 by Daniel Boulet. I've been running it on my
machine for about a month now. For the moment, I consider this an
alpha release, although it has proven stable where I have tried
it. Installation involves applying a patch to a 1.1.45+ source tree
and compling it and the ipfirewall application. If you're not familiar
with this or your networking setup, please wait for a later release.
Primary-site: sunsite.unc.edu /pub/Linux/kernel/patches/network
42 kB ipfirewall0.1.tar.gz
(well, that's what the lsm said, currently it's in pub/incoming.)
Alternate-site: ftp.forsci.ualberta.ca /pub/linux
42 kB ipfirewall0.1.tar.gz
If you try, please let me know how you do.
Enjoy,
-Bob Beck
------------------------------
From: deuelpm@craft.camp.clarkson.edu (Pete Deuel)
Subject: Re: IDE Performance enhancement
Date: Mon, 5 Sep 1994 01:49:33 GMT
>About IDE multiple-sector: the rate of a "dd" off the disk doubles when
>setting 8-sector transfer on the Quantum disk in my machine at work.
>I did not time the difference in compilation speed, but again it will
>probably not make so much difference.
I have to think that, if there is to be any noticible speed difference, you
really must stick to disk-intensive applications. I would guess kernel
compilation is not THE thing to test this with.
Off the top of my head, I'd suggest this: compile the kernel with and without
hd enhancements. In each state (out of X-windows, to be scientific), mount an
empty filesystem and pipe yes into a file (yes > /mnt/tempfile) until the fs
is full--the amount of time it takes to fill up the fs should change if there
is any improvement to be had.
I'd try this myself, but I'd like to get everything else working right now!
If someone tries this (at your OWN risk! (: ), please post the results.
Pete
=====================================================
"I'm just a lab mouse on stilts..."
deuelpm@craft.camp.clarkson.edu
=====================================================
------------------------------
From: badenes@imec.be (Goncal Badenes)
Subject: Re: 1.1.49 trouble
Date: Fri, 2 Sep 1994 07:52:13 GMT
JL Gomez (kitana!sysop@caprica.com) wrote:
:> My 1.1.49 kernel panics when doing heavy I/O.
:> Here is relevant syslog entries:
:> Aug 30 22:05:29 kitana kernel: Unable to handle kernel NULL pointer
:> dereference at virtual address c0000834
:> Aug 30 22:05:29 kitana kernel: current->tss.cr3 = 00654000, Xr3 = 00654000
:> Aug 30 22:05:29 kitana kernel: *pde = 00102027
:> Aug 30 22:05:29 kitana kernel: *pte = 00000027
:> <INN stuff deleted>
:> Aug 31 14:38:46 kitana kernel: Kernel panic: EXT2-fs panic (device 8/19):
:> ext2_read_inode: unable to read i-node block
:> I've been reading threads about heavy file I/O being flakey. I've had this
:> problem since 1.1.39.
:> --
:> sysop@kitana.org
I have also experienced that problem with 1.1.40 -> .49
My machine is a 486DX33, 16MB, with Adaptec 1542B, Conner 540MB,
Micropolis 2.1GB, Toshiba 4101 CD-ROM and Tandberg TDC3600 streamer. In
my case the error sometimes occurs even if there is no real heavy file
I/O. I was susptecting my hard disk was defective (is a new one).
--
Goncal Badenes Tel: +32 16 281529
IMEC Fax: +32 16 281214
Advanced Semiconductor Processing
Micropatterning Group
Kapeldreef 75 e-mail: badenes@imec.be
B-3001 Leuven (Belgium) goncal@cnm.es
------------------------------
From: hpulley@uoguelph.ca (Harry C Pulley)
Subject: Re: IDE Performance enhancement
Date: 5 Sep 1994 03:00:47 GMT
Pete Deuel (deuelpm@craft.camp.clarkson.edu) wrote:
: >About IDE multiple-sector: the rate of a "dd" off the disk doubles when
: >setting 8-sector transfer on the Quantum disk in my machine at work.
: >I did not time the difference in compilation speed, but again it will
: >probably not make so much difference.
: I have to think that, if there is to be any noticible speed difference, you
: really must stick to disk-intensive applications. I would guess kernel
: compilation is not THE thing to test this with.
: Off the top of my head, I'd suggest this: compile the kernel with and without
: hd enhancements. In each state (out of X-windows, to be scientific), mount an
: empty filesystem and pipe yes into a file (yes > /mnt/tempfile) until the fs
: is full--the amount of time it takes to fill up the fs should change if there
: is any improvement to be had.
: I'd try this myself, but I'd like to get everything else working right now!
: If someone tries this (at your OWN risk! (: ), please post the results.
I use iozone. I got a 50% speed up when I went from 1.0 to 1.1.45 with
multcount 16, read-ahead 16, etc. Going to VLB with my IDE drives yielded
another 35% improvement (on top of the 50% improvement from 1.1.45).
I think that iozone does a pretty good job. Creating a new filesystem will
give you completely reproducable results, this is true. But echoing yes until
it fills up may be overkill.
Even iozone isn't really great for anything but testing big silly sequential
writes and reads. Timing the startup time of X is pretty good ;-)
Harry
--
<:-{} hpulley@uoguelph.ca |This message released|It takes all kinds,
\ Harry C. Pulley, IV |to the PUBLIC DOMAIN.|and to each his own.
==================================+=====================|This thought in mind,
Stay away from the DOS side, Luke!|Un*x don't play that.|I walk alone.
------------------------------
From: hpulley@uoguelph.ca (Harry C Pulley)
Subject: Re: polled ports
Date: 5 Sep 1994 03:06:18 GMT
Rob Janssen (rob@pe1chl.ampr.org) wrote:
: In <34chrk$1me@nermal.cs.uoguelph.ca> hpulley@uoguelph.ca (Harry C Pulley) writes:
: >Thanks to people here I can now run with polled ports. But, even with just a
: >1200baud mouse connection it doesn't work very well. Under X it is barely
: >usable but under dosemu it is totally useless.
: I told you that
Yes, and you were very correct, unfortunately...
: >Are there any known improvements to the polling code? 1200 baud shouldn't be
: >too fast...
: It polls 100 times per second, the timer-interrupt rate. So 1200 baud
: is already slightly too fast for this mode.
: It could be usable with 16550A chips, but they have other problems when
: used with a mouse. Therefore they are disabled below 2400.
Well, it just bugs me that I can reboot my system to Coherent and use the mouse
on COM2 and a terminal on polled COM4 at 9600 bps or higher (165450) without
any trouble. Running my terminal polled under Linux yields a speed which
looks, coincidentally, like it is slower than 1200 bps. Coherent changes its
polling rate depending on the bps rate of the port. This doesn't work when you
really go up in speed but it will definitely allow a 386 or even 286 to handle
9600 baud OK.
I am looking at the serial polling code now. If I come up with anything, I'll
post patches for people to try.
: It is better to listen to the rest of the world, and somehow move your
: IRQs or use a board which can share IRQs.
I know this might work. But, I don't want to wreck even a $5 board if I can
help it and I know that a lot of people don't want to either. Linux should be
able to poll a port at a reasonable rate. It's not too much to ask for.
Harry
--
<:-{} hpulley@uoguelph.ca |This message released|It takes all kinds,
\ Harry C. Pulley, IV |to the PUBLIC DOMAIN.|and to each his own.
==================================+=====================|This thought in mind,
Stay away from the DOS side, Luke!|Un*x don't play that.|I walk alone.
------------------------------
Subject: which kernel with dosemu 0.53 ?
From: srini@tisl.ukans.edu (Srini Seetharam)
Date: 4 Sep 94 20:35:31 CDT
Reply-To: srini@tisl.ukans.edu (Srini Seetharam)
I downloaded the dosemu0.53 pre alpha release.
I would like to know which kernel it is compatible with.
preferably aftre 1.1.40 or so.
I also need compatibility with ncr 53c810 scsi,
PPP and PCI bus.
please bear with me , it takes me 2 hrs to compile a
kernel on my slow 386.
I currenly run on a 386-20 but plan to move to a 486
based machine with the ncr and pci bus.
Thank you
--
srini
------------------------------
From: hpulley@uoguelph.ca (Harry C Pulley)
Subject: Re: DOSEMU 0.53 notes
Date: 5 Sep 1994 03:14:42 GMT
Mark Krischer (mkrisch@avalanche.mpce.mq.edu.au) wrote:
: i've started playing around with DOSEMU .53 and i am VERY impressed.
: the idea of being able to have a real word processor (WP 6.0 DOS) and not have to
: be in DOS/Windows is absolutely great!!!
I too enjoy running WP 6.0 for DOS under Linux.
: just a couple of notes. things i've noticed and problems i've had
: mouse doesn't seem to work to well--do i need to run the DOS driver? when i do
: i'm told its already loaded.
Hmm. I load a mouse driver and it work OK as long as /dev/mouse is interrupt
driven rather than polled. Have you checked the config.sys and autoexec.bat
files to make sure it isn't being loaded twice somehow?
: is there support for the ATI mach32 chip? i can only get 640x480 16 colors in
: graphics mode for WP--though the fact i can get graphics mode at all is great! the
: screen goes screwy if i try 256 colors. and if i try a different mode, WP tells me
: it won't work with my hardware config (it does work in native DOS).
Hmmm. I don't know about this but my Cirrus Logic 5428 allows me to run at
1024x768x256. What is your video line from dosemu.conf? Mine is just:
video { vga console graphics }
: also, what do i need to do to get color in text mode--i got it once and i haven't a
: clue what i did, but i never saw it again.
This may be connected to your getting it to work in an xterm. You need a
different video config, I think, for things to work well on the console as
opposed to in an xterm. For an xterm I leave out all video lines while for the
console I have the video line listed above. Are you sure you didn't change it
to run under X and have not changed it back?
: a lot slower though. also delete and bs wouldn't work, but that might have been
: my termcaps not set up write in the dosemu.config or something like that.
You can get delete and backspace to work in an xterm. I didn't do anything
special but they work for me.
I find it is somewhat slower but WP6.0, DOSEdit, Turbo C++, etc. seem to run
pretty well. You can't expect emulation to run flat out!
: lastly, can someone give me some information on good and SAFE ways to redirect boot to
: my DOS partition. i also use double-space, so i'd be interested in knowing how to
: gain access to that--the HOWTO is not completely clear and i'd really prefer not to
: thrash my hard drive, it's young and i want to break it in slowly.
: all in all, great stuff!!
If you don't use lilo on your hard drive to boot (like me), all I did was
change the disk line from the "disk { image..." thing to:
disk { wholedisk "/dev/hda" }
It boots right off of it and loads Stacker just fine.
The only other problem I have noticed is that mem does not show any EMS memory
even though I ask for it. Am I the only one for which EMS doesn't show up?
Harry
--
<:-{} hpulley@uoguelph.ca |This message released|It takes all kinds,
\ Harry C. Pulley, IV |to the PUBLIC DOMAIN.|and to each his own.
==================================+=====================|This thought in mind,
Stay away from the DOS side, Luke!|Un*x don't play that.|I walk alone.
------------------------------
From: bon@lte.e-technik.uni-erlangen.de (Uwe Bonnes)
Subject: Re: linux never swaps ?
Date: Sun, 4 Sep 1994 23:44:55 GMT
Larry Doolittle (doolitt@recycle.cebaf.gov) wrote:
> I have monkeyed around with a thrash program to force a machine into
> swap (actually for testing a SCSI driver). Here is one such attempt,
> it needs some work but should do the job for Tim. Personally, I have
> no trouble making my machine swap. I only have 8M RAM.
> (many line of code deleted)
Did you try a "tail -20 /dev/zero"?
Short after bootup this stalls my system (486-66DX2, 16Meg, PCI) for about
an hour. On a system running some time, the command fails after a minute or
so with the notice " virtual memory exhausted".
Anybody has the same experience?
--
Uwe Bonnes bon@lte.e-technik.uni-erlangen.de
---
Uwe Bonnes bon@lte.e-technik.uni-erlangen.de
------------------------------
From: bon@lte.e-technik.uni-erlangen.de (Uwe Bonnes)
Subject: Re: linux never swaps ?
Date: Sun, 4 Sep 1994 23:50:35 GMT
Larry Doolittle (doolitt@recycle.cebaf.gov) wrote:
> I have monkeyed around with a thrash program to force a machine into
> swap (actually for testing a SCSI driver). Here is one such attempt,
> it needs some work but should do the job for Tim. Personally, I have
> no trouble making my machine swap. I only have 8M RAM.
> (many line of code deleted)
Did you try a "tail -20 /dev/zero"?
Short after bootup this stalls my system (486-66DX2, 16Meg, PCI) for about
an hour. On a system running some time, the command fails after a minute or
so with the notice " virtual memory exhausted".
Anybody has the same experience?
--
Uwe Bonnes bon@lte.e-technik.uni-erlangen.de
---
Uwe Bonnes bon@lte.e-technik.uni-erlangen.de
------------------------------
From: chrisb@wombat.cssc-syd.tansu.com.au (Chris Bitmead)
Subject: Re: Linux - my first impressions
Date: 5 Sep 94 12:12:22
In article <HUMBERTO.94Sep3062931@annexstein.csm.uc.edu> humberto@annexstein.csm.uc.edu (Humberto Ortiz Zuazaga) writes:
>>>>>> "WL" == Warner Losh <imp@boulder.parcplace.com> writes:
>
> WL> In article <345inv$mo6@bmerha64.bnr.ca>
> WL> Hamish.Macdonald@bnr.ca (Hamish Macdonald) writes:
> >> Rather than containing code to read partition tables and Linux
> >> partitions, lilo figures out which disk blocks the kernel file
> >> is on, and puts the block number data in a place it can find on
> >> bootup.
>
> WL> The lilo way of doing it is a hack. The boot code should know
> WL> enough to know where to find the kernel by name, and not by a
> WL> list of blocks that may change in the future.
>
>I'm surprised Warner, you know this was a design decision. The
>problem is _which_ filesystem should lilo understand (minix, ext,
>ext2, xiafs, umsdos, SysV, FFS)? Answer: none, much simpler code,
>lilo will boot off of anything.
>
>Now that most people have settled on ext2 as a standard, perhaps it
>would be worthwhile to write a new bootloader that understood it (any
>volunteers?). This was not the case when lilo was developed, there
>were likely four "standard" file system types at the time, and that's
>a lot of code to stuff into a boot loader.
Actually I've always wondered how lilo works. I understand it records the
disk block the kernel is on. But what if the kernel isn't continuous? What
is it that knows enough about the file system in order to find *all* the
disk blocks that make up the kernel?
------------------------------
** 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
******************************