Linux-Development-Sys Digest #129, Volume #8 Thu, 7 Sep 00 18:13:13 EDT
Contents:
ext3 in 2.4.0-test kernels (Stefan Boresch)
Re: buffer_dirty - what's the @#$%? ("Ian Dichkovsky")
Re: PPP/BRIDGE(?)/ROUTING ? (James Carlson)
i'd like some information (Bart De Schuymer)
Re: i'd like some information (Frank Sweetser)
Re: ext3 in 2.4.0-test kernels (Daniel Robert Franklin)
interrupt HD access, how ? (ZIP-Prob.) (Dirk Schenkewitz)
Re: i'd like some information (Szabolcs Csetey)
Re: Curious: compiling test8 series: "warning... token" (Robert Lynch)
Load/unload modules programatically ("Chris")
VFS: Determining FS type (Graeme McLaughlin)
mmap implementation for PCI device ("Edward A. Hildum")
Re: mmap implementation for PCI device (Grant Edwards)
Ftp library ("Patrick Collet")
Re: mmap implementation for PCI device ("Edward A. Hildum")
Re: mmap implementation for PCI device (Grant Edwards)
Re: buffer_dirty - what's the @#$%? ([EMAIL PROTECTED])
help with init (Marble Head)
----------------------------------------------------------------------------
From: Stefan Boresch <[EMAIL PROTECTED]>
Subject: ext3 in 2.4.0-test kernels
Date: 07 Sep 2000 11:30:50 +0200
How do I get
ext3 filesystem support in the 2.4.0-test kernels. Is it not
already there (I believe remembering something like this), but
I find no flag for configuring it. I downloaded the ext3-0.0.2f
package from kernel org, but that seems to contain only diffs
agains a RH 2.2.16 kernel and a 2.2.17-pre something. I would
guess (haven't tried) that those patches are useless for 2.4.0.
Thanks for helping me out -- probably I am just looking in the wrong
places...
Stefan
------------------------------
From: "Ian Dichkovsky" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc,comp.os.linux.questions
Subject: Re: buffer_dirty - what's the @#$%?
Date: Thu, 7 Sep 2000 14:07:54 +0300
> > So, what it mean - buffer_dirty ?
>
> Caching. It happens in Windows if you speed Windows up (but Windows
> defaults to turn off write caching). When you write a file to floppy all
I copy file _FROM_ floppy to hard
mount /mnt/fd; cp /mnt/fd/* /prj; umount /mnt/fd
Bye!
------------------------------
From: James Carlson <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,alt.os.linux.slackware
Subject: Re: PPP/BRIDGE(?)/ROUTING ?
Date: 07 Sep 2000 07:51:55 -0400
David Ronis <[EMAIL PROTECTED]> writes:
> The little I've read about the experimental bridging software sounds
> like it might answer my problem, however, it only discusses the case
> of multiple eth? devices. Will it work for the configuration I show
> above? If not, is there another way to do this? [e.g., using some
> sort of IP number translation]?
The *right* way to do this is to route. You set up a separate subnet
for your home machine's eth0, give the work machine a static route
pointing to your home machine as the next hop for that subnet, and
give all of the other machines on your home LAN default routes
pointing to your home machine.
Failing that, you *might* be able to make this work using the same
subnet in both locations. To do this, you'll have to set up published
ARP entries on the work machine for all of your home machines and
install host routes for each pointing to your home machine as the next
hop (or, for Linux, just pointing to the PPP interface). The opposite
will need to be done on your home machine -- install ARP entries and
host routes for each work machine. Doing this certainly creates a
real mess and I think really has little to recommend it (though I know
of some authors who do recommend this).
Routing is good. Don't try to bridge everything.
--
James Carlson, Internet Engineering <[EMAIL PROTECTED]>
SUN Microsystems / 1 Network Drive 71.234W Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757 42.497N Fax +1 781 442 1677
Second Edition now available - http://people.ne.mediaone.net/carlson/ppp
------------------------------
From: [EMAIL PROTECTED] (Bart De Schuymer)
Subject: i'd like some information
Date: Thu, 07 Sep 2000 12:30:20 GMT
Hello all,
I study computer science at the university of Ghent (Belgium)
and am starting my final year.
I am preparing a thesis in which i have to do the follwing:
I have to incorporate the keep-state implementation in the OpenBSD
unix-like OS in the kernel firewall of linux (Redhat).
This means that I have to make linux' packet filter dynamic (now it is
static).
I believe this will be very interesting for me, but I was wondering
how you became capable of kernel hacking.
There must be books or sites available that can help me with this
right?
Also, do I have to talk to someone about what I am going to do. So
people know about this? So, do I need to inform someone or some
organization?
I could actually come up with a hundred questions but I'll save your
time.
Hope I'm at the right place here.
happy programming,
Bart De Schuymer
------------------------------
From: [EMAIL PROTECTED] (Frank Sweetser)
Subject: Re: i'd like some information
Date: 7 Sep 2000 12:45:06 GMT
Bart De Schuymer <[EMAIL PROTECTED]> wrote:
>I believe this will be very interesting for me, but I was wondering how you
>became capable of kernel hacking. There must be books or sites available that
>can help me with this right? Also, do I have to talk to someone about what I
>am going to do. So people know about this? So, do I need to inform someone or
>some organization? I could actually come up with a hundred questions but I'll
>save your time.
There's not much in the way of sites or (current) books - but you also don't
have to notify anyone or fill out any paperwork. just grab a copy of the
source off of ftp://ftp.us.kernel.org (or looking at your headers, probably
ftp://ftp.nl.kernel.org) and start hacking. when you're dont, go ahead and
release it - but make sure that you've read the GPL license (it comes with
the kernel source code) if you start distributing it.
--
Frank Sweetser rasmusin at wpi.edu, fs at suave.net | $ x 5
Full-time WPI Network Tech, Part time Linux/Perl guy |
Before you criticize another man, walk a mile in his shoes. That way,
when you do criticize him, you're a mile away, and you've got his shoes.
------------------------------
Subject: Re: ext3 in 2.4.0-test kernels
From: [EMAIL PROTECTED] (Daniel Robert Franklin)
Date: 7 Sep 2000 22:30:37 +1100
Stefan Boresch <[EMAIL PROTECTED]> writes:
>How do I get
>ext3 filesystem support in the 2.4.0-test kernels. Is it not
>already there (I believe remembering something like this), but
>I find no flag for configuring it. I downloaded the ext3-0.0.2f
>package from kernel org, but that seems to contain only diffs
>agains a RH 2.2.16 kernel and a 2.2.17-pre something. I would
>guess (haven't tried) that those patches are useless for 2.4.0.
It's not there yet. I think they are trying to hammer out some
common-functionality between the various journalled FS projects before
putting ext3/Reiser/XFS etc. in (plus I think XFS needs some more work). I'd
expect something early in 2.5, probably backported to 2.4.x, x > 0.
- Daniel
--
******************************************************************************
* Daniel Franklin - Postgraduate student in Electrical Engineering
* University of Wollongong, NSW, Australia * [EMAIL PROTECTED]
******************************************************************************
------------------------------
From: Dirk Schenkewitz <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware
Subject: interrupt HD access, how ? (ZIP-Prob.)
Date: Thu, 07 Sep 2000 16:46:10 +0200
Hi All,
Is there a way to make accesses to _locally_mounted_ discs interruptible ?
The problem is: when I acces a ZIP drive, it sometimes happens that the drive
goes into a loop of rhythmic stepping and such, over and over, until I pull
and replug its power (It's a parallel ZIP, accessed via ppa). After that,
the ZIP-drive calms down, but now the linux system hangs completely - no
console switching, no shitdown er shutdown via alt-ctrl-del, only hard reset
works.
Even if could not access the (replugged) ZIP-drive, it would be nice to (at
least) get back control of the rest of the system, so I could do a shutdown
- and don't have to wait for 'fsck' to get finished.
I know that nfs-mounts can be made interruptibe - is there a solution for
local HDs ?
(I hope I haven't overlooked something in some docs - if so, yell at me (but
please tell me where to look))
greeting
dirk
--
Dirk Schenkewitz
InterFace AG phone +49 (0)89 / 610 49 - 126
Leipziger Str. 16 fax +49 (0)89 / 610 49 - 85
D-82008 Unterhaching
http://www.InterFace-AG.com mailto:[EMAIL PROTECTED]
------------------------------
From: Szabolcs Csetey <[EMAIL PROTECTED]>
Subject: Re: i'd like some information
Date: Thu, 07 Sep 2000 15:29:37 GMT
[EMAIL PROTECTED] (Frank Sweetser) wrote:
> Bart De Schuymer <[EMAIL PROTECTED]> wrote:
> >I believe this will be very interesting for me, but I was wondering
> >how you became capable of kernel hacking. There must be books or
> >sites available that can help me with this right?
> There's not much in the way of sites or (current) books [...]
cd /usr/src/linux/Documentation/
egrep 'Title:|Name:' kernel-docs.txt | wc -l
57
On the other hand this also seems a decent number of resources to start
with ...
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Robert Lynch <[EMAIL PROTECTED]>
Subject: Re: Curious: compiling test8 series: "warning... token"
Date: Thu, 07 Sep 2000 08:57:38 -0700
Reply-To: [EMAIL PROTECTED]
Kasper Dupont wrote:
>
> Robert Lynch wrote:
> >
> > This is subjective, based on idly watching the output of the
> > kernel compile of some of the 2.4.0-test8 series (it might have
> > happened with other, testX series, but maybe I simply wasn't
> > watching at the time, thus didn't see the message), but it seems
> > a new sort of warning pops up frequently:
> >
> > "warning: pasting would not give a valid preprocessing token"
> >
> > It seems to come from defines including a "##", involving a
> > paste, for example:
> >
> > #define BUGMSG2(x,msg,args...) do { BUGLVL(x) printk(msg, ##
> > args); } while (0)
> >
> > Any comments on this, is it incorrect, a hi-falutin' C extension,
> > etc.?
> >
> > Bob L.
> > --
> > Robert Lynch-Berkeley CA [EMAIL PROTECTED]
>
> The gcc manual (File: gcc.info, Node: Macro Varargs)
> says that you must put a space before the ,. That
> might be changed in later versions so the space will
> no longer be needed but still allowed.
>
> #define BUGMSG2(x,msg,args...) do { BUGLVL(x) printk(msg , ##
> args); } while (0)
>
> --
> Kasper Dupont
Hmm; thanks for the observation, but it doesn't seem to do it.
Adding the space:
$ cat error1.h
#define BUGMSG2(x,msg,args...) do { BUGLVL(x) printk(msg , ##
args); } while (0)
BUGMSG2(1,"doobie","feeble","weeble");
$ cpp -E error1.h
# 1 "error1.h"
error1.h:4:37: warning: pasting would not give a valid
preprocessing token
do { BUGLVL(1) printk("doobie" ,"feeble","weeble"); } while (0);
Bob L.
--
Robert Lynch-Berkeley CA [EMAIL PROTECTED]
------------------------------
From: "Chris" <[EMAIL PROTECTED]>
Subject: Load/unload modules programatically
Date: Thu, 7 Sep 2000 10:23:26 -0700
Is there a way to load and unload kernel modules from within a user
application? I am currently using a shell script consisting of insmod,
search /proc/devices for major number, mknod. Then another script using
rmmod to unload the module when the app closes.
What I really want is to have the application manage the loading and
unloading of the module without having to rely on the shell.
Has anyone figured this one out?
Chris
------------------------------
From: Graeme McLaughlin <[EMAIL PROTECTED]>
Subject: VFS: Determining FS type
Date: Thu, 07 Sep 2000 17:35:39 GMT
Hello,
I'm trying to use a "special" filesystem for my root device. Using
initrd, I load the proper module, prior to mounting the root
filesystem.
This filesystem is a wrapper around ext2. The device I am trying to
mount as the root filesystem is actually ext2. This special
filesystem provides extra functionality over ext2. The problem is
that the VFS recognizes it as ext2, and not the filesystem I want.
My /proc/filesystems looks like this:
ext2
nodev proc
iso9660
nodev devpts
nodev myfsnd
myfs
nodev nfs
>From what I understand, the kernel will try each filesystem in the
order of that list. In order for this filesystem to be mounted
properly (by the VFS, at boot time) "myfs" would need to be registered
with the kernel *above* ext2.
Is it possible to do this without a kernel modification? Is there
some sort of parameter that I can pass the kernel specifying which
filesystem type I want the root filesystem to be mounted with?
I need this to work on the 2.2.* kernels.
Thanks for any help,
Graeme
------------------------------
From: "Edward A. Hildum" <[EMAIL PROTECTED]>
Subject: mmap implementation for PCI device
Date: Thu, 07 Sep 2000 10:59:11 -0700
I'm working on a driver for a memory-mapped PCI processor, trying to get
mmap to work. What kind of address do I pass to remap_page_range() for
the physical address of the PCI memory? The value returned in the
pci_dev structure by pci_find_device() doesn't seem to work. I tried
going through the frame buffer drivers, but they all get the base
address from find_devices(), which doesn't seem to exist :-)
Ted Hildum
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Subject: Re: mmap implementation for PCI device
Date: Thu, 07 Sep 2000 19:28:05 GMT
In article <[EMAIL PROTECTED]>, Edward A. Hildum wrote:
>I'm working on a driver for a memory-mapped PCI processor, trying to get
>mmap to work. What kind of address do I pass to remap_page_range() for
>the physical address of the PCI memory? The value returned in the
>pci_dev structure by pci_find_device() doesn't seem to work. I tried
>going through the frame buffer drivers, but they all get the base
>address from find_devices(), which doesn't seem to exist :-)
There's an example driver that does this at:
ftp://ftp.visi.com/users/grante/stuff/demomm.tar.gz
--
Grant Edwards grante Yow! Are you still
at SEXUALLY ACTIVE? Did you
visi.com BRING th' REINFORCEMENTS?
------------------------------
From: "Patrick Collet" <[EMAIL PROTECTED]>
Subject: Ftp library
Date: Thu, 7 Sep 2000 21:48:47 +0200
Hi,
I'm looking for an ftp library on Linux. Do you know anyone ? Do you know
which one is the best ?
Thanks for your answer
--
Patrick Collet
France
------------------------------
From: "Edward A. Hildum" <[EMAIL PROTECTED]>
Subject: Re: mmap implementation for PCI device
Date: Thu, 07 Sep 2000 12:12:25 -0700
Many thanks for your quick response. This looks like just the ticket.
Ted Hildum
Grant Edwards wrote:
> In article <[EMAIL PROTECTED]>, Edward A. Hildum wrote:
>
> >I'm working on a driver for a memory-mapped PCI processor, trying to get
> >mmap to work. What kind of address do I pass to remap_page_range() for
> >the physical address of the PCI memory? The value returned in the
> >pci_dev structure by pci_find_device() doesn't seem to work. I tried
> >going through the frame buffer drivers, but they all get the base
> >address from find_devices(), which doesn't seem to exist :-)
>
> There's an example driver that does this at:
>
> ftp://ftp.visi.com/users/grante/stuff/demomm.tar.gz
>
> --
> Grant Edwards grante Yow! Are you still
> at SEXUALLY ACTIVE? Did you
> visi.com BRING th' REINFORCEMENTS?
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Subject: Re: mmap implementation for PCI device
Date: Thu, 07 Sep 2000 21:05:30 GMT
In article <[EMAIL PROTECTED]>, Edward A. Hildum wrote:
>> >I'm working on a driver for a memory-mapped PCI processor, trying to get
>> >mmap to work. What kind of address do I pass to remap_page_range() for
>> >the physical address of the PCI memory?
>> There's an example driver that does this at:
>>
>> ftp://ftp.visi.com/users/grante/stuff/demomm.tar.gz
>Many thanks for your quick response. This looks like just the ticket.
BTW, that example has only been tested w/ 2.2 kernels -- 2.3 or
2.4 may require some adjustments.
--
Grant Edwards grante Yow! Hand me a pair of
at leather pants and a CASIO
visi.com keyboard -- I'm living
for today!
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: buffer_dirty - what's the @#$%?
Crossposted-To: comp.os.linux.hardware,comp.os.linux.misc,comp.os.linux.questions
Date: 7 Sep 2000 17:58:15 -0400
In comp.os.linux.questions Ian Dichkovsky <[EMAIL PROTECTED]> wrote:
>> > So, what it mean - buffer_dirty ?
>>
>> Caching. It happens in Windows if you speed Windows up (but Windows
>> defaults to turn off write caching). When you write a file to floppy all
> I copy file _FROM_ floppy to hard
Well, then the system holds off and does the copy in the background when
it has time. If you remove the floppy before it gets around to
doing/finishing the copy the buffer is "dirty."
sync
forces the write (rather than waiting).
(there is a manual page for sync)
(Windows caches reads and not writes unless you change the defaults.
Linux caches both. You have to be careful on a system which
caches writes.)
------------------------------
From: Marble Head <[EMAIL PROTECTED]>
Subject: help with init
Date: Thu, 07 Sep 2000 17:56:10 -0400
Can anybody tell me why the kernel would panic, saying
"Kernel panic: No init found. Try passing init= option to kernel."
(Don't tell me there's no init. Read on.)
sdb5 used to be a copy of sdb1, but i deleted some files, described
below.
I am able to boot using "root=/dev/sdb1" but when I try "root=/dev/sdb5"
the problem happens.
sdb1 has a full install, debian potato.
sdb5 I want to set up for a minimal boot. contains copies of the
following
directories from sdb1:
/bin delete everything but sh, a copy of (not lnk to) bash.
/dev delete everything but console, fd?, initctl, sda*, sdb*, tty?
/etc delete everything but inittab
/sbin delete nothing
/usr /var /tmp /home were entirely discarded.
When I say "minimal boot," I mean just enough to get sulogin up. In
fact, my
inittab file is as follows:
si::sysinit:/sbin/sulogin
I am sure that init is unharmed in the sbin directory.
------------------------------
** 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: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
