Linux-Development-Sys Digest #580, Volume #8 Wed, 21 Mar 01 06:13:07 EST
Contents:
Re: large pings with DF bits ([EMAIL PROTECTED])
Re: Linux and IS-IS (h c wispell ii)
Re: what type of data does /dev/mouse pass for alternate pointing device? ("Josh K.")
Re: copy_to_user() question (Kevin Buhr)
Get paid to be online... no surfing! (R. Somebody)
Re: How2 save process memory image so it can resume later ("Nick Long")
Re: copy_to_user() question ("Jim Fischer")
2.4.2, loopback, and lockups (Kent A Vander Velden)
unable to mount floppy? ("Sean Xu")
sock_release in 2.4 (satishk)
can we update the task_struture of parent ?? (satishk)
kernel problem (Dipl. Ing. Werner =?iso-8859-1?Q?K=FChnert?=)
How to create a sendmail user via http request ("Gary Cheng")
user-mode linux ("Stephanie")
Re: How to create a sendmail user via http request ("Nils O. Sel�sdal")
mouse device driver programming (Aminudin Khalid)
Re: unable to mount floppy? (Mike Dowling)
Re: How to use C++ to develope the linux driver module? ("O.Petzold")
Re: Apology for serial porting problems ("Gianluca Puggelli (Gianluca)")
Re: How to use C++ to develope the linux driver module? (Erik de Castro Lopo)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED]
Subject: Re: large pings with DF bits
Date: Tue, 20 Mar 2001 22:10:16 -0000
On Tue, 20 Mar 2001 11:47:14 +0000 (UTC) sreangsu acharyya
<[EMAIL PROTECTED]> wrote:
| I want to ping a host with an unfragmented packet of size around ( 500
| - 900 ) bytes. Linux's ping doesnt have an option to set the DF bit. So
| if i modify the source code of ping with appropriate setsockopt(), will
| that suffice.
|
| I ask this specifically because in the /usr/src/linux/ipv4/icmp.c the
| function icmp_send( ) seems to be checking for sizes in excess of 576
| and dropping the extra baggage. ( I am pretty much a newbie, havent
| clocked more than a hour eye-balling the kernel code...so I might be
| wrong ).
|
| Does this piece of code get involved, when an userlevel program sends
| ICMP packets ? I can always send with ping -s, but have a feeling that
| these get passed along after fragmentation. BTW the MTU is larger than
| the size of the packet I want to send.
You mean /usr/src/linux/net/ipv4/icmp.c ?
If you are referring to:
room = rt->u.dst.pmtu;
if (room > 576)
room = 576;
then I believe this applies only to ICMP packets returned as a
result of an error, such as an unreachable host. The idea here
is to be sure the error packets can get back in the worst case
MTU/MRU. That won't always be the case since you can set MTU
to a value even smaller than 576, though it is sure to fire up
a heated debate as to whether such an administrative configuration
conforms to the RFCs (implementation restrictions seems to be what
is clear in the RFCs). Still, if you have to truncate a packet
with the intent to make sure it gets where it's going, and don't
have a discovered MTU handy, 576 is a reasonable assumption.
This code does not appear to affect ping request or ping reply.
I just verified this by doing:
ping -s 1472 someotherhostofmine
and watched the packets going through with tcpdump -x (tcpdump
apparently has no option to display packet length, so one has to
dump the whole packet and count octets). I saw packets of 1500
octets.
The code does not appear to care if DF is set, so you should be
able to send a packet with a size up to your own MTU, and you
should get back packets indicating drops on routers in between
where you exceed the MTU, if they implement PMTU discovery right.
--
=================================================================
| Phil Howard - KA9WGN | Dallas | http://linuxhomepage.com/ |
| [EMAIL PROTECTED] | Texas, USA | http://phil.ipal.org/ |
=================================================================
------------------------------
From: h c wispell ii <[EMAIL PROTECTED]>
Subject: Re: Linux and IS-IS
Date: Tue, 20 Mar 2001 16:14:13 -0600
open socket with
protocol = htons(ETH_P_ALL);
socket(PF_PACKET, SOCK_RAW, protocol)
and create your own pdus
clay
talmon wrote:
>
> Hi
>
> I need to support the IS-IS routing protocol on Linux, but to handle only IP
> data. Since there is no support for ISO CLNP in Linux:
> How would I go about handling the IS-IS protocol in Linux?
> Has anybody done this, maybe with some kind of encapsulation of IS-IS PDUs
> in IP?
> Is there some sort of package out there that I can use?
>
> Any suggestions would be appreciated.
>
> TIA
> Talmon
------------------------------
From: "Josh K." <[EMAIL PROTECTED]>
Subject: Re: what type of data does /dev/mouse pass for alternate pointing device?
Date: Tue, 20 Mar 2001 22:19:38 GMT
Sorry, forgot to say that any libraries in the X window system or third
party libraries are welcome. I am more concerned with getting the pointing
device working under the windowing system, it does not have to work under
the bare console if that is what it takes. I though it would be easier to
write a program that replaced the dev/mouse file with it's own then any
service that accesses this file would get normal looking mouse data but
programs with "enhanced" ability that I may write later could get addition
dimension info from the VR device maybe from another device file that passes
differnet data structures to separate both mouse emulation and VR device
only data?.
Sorry to respond to myself but I wanted to acurately describe this project's
goal.
so what are your thoughts?
Thanks a lot, Josh
:-)
"Josh K." <[EMAIL PROTECTED]> wrote in message
news:P%Ot6.596293$[EMAIL PROTECTED]...
> Hi,
> I wish to started a project using a VR device as a simple mouse
pointer.
> I was currious if anybody had any info regarding either changing gpm, or
> pointing the X window system to another dev file for mouse pointer info.
>
> where can I get info on how linux deals with mice and what type of data is
> passed from the gpm app to the window system reading the dev file? I have
> enough info to take 3D points from the device but I do not know in what
> presentation the standard mouse driver presents the X,Y points. Any help
> would be greatly apprciated.
>
> Also any info on using the parport drivers now coming standard to many
linux
> distros for grabing signal of parallel port pins.
>
> Thanks for any help you can provide, if this projects gets out the
planning
> stage just more cool stuff for the linux community to play with :-)
>
> -Josh
>
>
------------------------------
From: [EMAIL PROTECTED] (Kevin Buhr)
Subject: Re: copy_to_user() question
Date: 20 Mar 2001 16:17:36 -0600
"Jim Fischer" <[EMAIL PROTECTED]> writes:
>
> Thanks for the feedback. I understand your comments about the "zero good,
> non-zero bad" idea on result codes, but the bottom line is the char buffer
> up in the user-layer app is not being modified by the kernel-layer
> copy_to_user() call. <?> Note that the user-layer buffer *is* being modified
> correctly when I use the for() loop & put_user() code. I'm probably just
> overlooking something "obvious."
How are you reading into this buffer from user space? If you're not
using the C "read" function---if you're using "fread", "fscanf", or
"fgets"---you're using C buffered I/O. In this case, the kernel is
copying the data into an internal buffer supplied by the C library,
not your buffer.
If your "put_user" version returns 23 while your "copy_to_user"
returns 0, it will appear that "put_user" is writing to the user-space
buffer but "copy_to_user" isn't. In fact, both are writing to the C
library's internal buffer: in the former case, 23 bytes are copied
from there to your buffer while in the latter case, your buffer
remains unmodified.
Have you verified that a "copy_to_user" version that copies "len"
bytes *and* returns the value "len" still doesn't work?
Kevin <[EMAIL PROTECTED]>
------------------------------
From: R. Somebody <[EMAIL PROTECTED]>
Subject: Get paid to be online... no surfing!
Date: Wed, 21 Mar 2001 01:37:51 GMT
Get paid to be online... no surfing!
Top service of its kind. No commitments. No hassles. No mandatory
purchases. No mandatory "offers" to take part in.
http://www.desktopdollars.com/default.asp?id=dwodefey
Desktopdollars
------------------------------
Reply-To: "Nick Long" <[EMAIL PROTECTED]>
From: "Nick Long" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.unix
Subject: Re: How2 save process memory image so it can resume later
Date: Wed, 21 Mar 2001 02:52:50 GMT
You need to look for a UNIX OS that supports checkpoint restart. SGI IRIX
and Cray are ones that supports this. I am not sure if other UNIX flavors
such as Linux do.
Check into IRIX Hibernator II for more information.
Also, I have heard of a project called "EPCKPT" which is a project in
development to do this with Linux. Go to
http://www.cs.rutgers.edu/~edpin/epckpt/ for mor information on this.
Nick Long
www.unixpertise.com
"Yotam Medini" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>
> Assume an application wants to save its status,
> stack and dynamic memory. For simplicity let's assume
> all file descriptors and sockets are closed.
>
> Are there standard or common (free?) libraries with an API, that
> can save such process image to a file which later
> can be executed to continue the program
> at the saving point.
>
> /* Pls send a copy of your reponse to [EMAIL PROTECTED] */
>
> thanks -- yotam
>
------------------------------
From: "Jim Fischer" <[EMAIL PROTECTED]>
Subject: Re: copy_to_user() question
Date: Tue, 20 Mar 2001 19:17:09 -0800
"Kevin Buhr" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> "Jim Fischer" <[EMAIL PROTECTED]> writes:
> >
> > Thanks for the feedback. I understand your comments about the "zero
good,
> > non-zero bad" idea on result codes, but the bottom line is the char
buffer
> > up in the user-layer app is not being modified by the kernel-layer
> > copy_to_user() call. <?> Note that the user-layer buffer *is* being
modified
> > correctly when I use the for() loop & put_user() code. I'm probably just
> > overlooking something "obvious."
Sigh... My brain hurts. It just occurred to me that the copy_to_user()
version is always returning 0 (zero) when in fact it should be returning the
number of bytes copied from kernel space to user space. Doh! And of course,
this is *exactly* what you told me about two or three messages ago, but for
whatever reason it didn't sink in. [Excuses: It's finals week here; I'm not
getting enough sleep; too much caffine; too many hours in front of the
computer; mind is spinning after reading IEEE journal articles on modeling
self-similar network traffic using multifractal wavelet transforms
<big-yawn>, etc., etc., ...]
Anyway, the code is working now. Thanks for the help, Kevin (et al.)
Jim
------------------------------
From: [EMAIL PROTECTED] (Kent A Vander Velden)
Subject: 2.4.2, loopback, and lockups
Date: 21 Mar 2001 04:42:08 GMT
Since updating to 2.4.x, I have been experiencing lockups whenever I
use the lookback device for mounting iso images. This on an SMP
machine if that makes any difference. I have not experimented much
with this, but wondered if someone had already discovered a solution.
Thanks
--
Kent Vander Velden
[EMAIL PROTECTED]
------------------------------
From: "Sean Xu" <[EMAIL PROTECTED]>
Subject: unable to mount floppy?
Date: Wed, 21 Mar 2001 06:12:31 GMT
I have installed RedHat Linux 7.0(kernel 2.2.16-22). It works fine before.
But after installing a few rpm packges. I got the message "wrong major or
minor number for/dev/fd0...." while trying to mount the floppy drive. What
is causing the problem?
Is there any easy way to fix it without re-installing the whole OS!
BTW: I don't even have a network connection on that computer so that i
cannot even back up my files.
Thanks in advance!
Sean
------------------------------
From: satishk <[EMAIL PROTECTED]>
Subject: sock_release in 2.4
Date: Wed, 21 Mar 2001 12:19:00 +0530
Hello,
I have a kernel extension where we call sock_release to release the
socket. Unforntunately it's not able to close the socket. Is there any
known problem with sock_release in 2.4 ? or is the interface for socket
release changed???
Help is really appreciated.
Thanx
Regards
Sat
------------------------------
From: satishk <[EMAIL PROTECTED]>
Subject: can we update the task_struture of parent ??
Date: Wed, 21 Mar 2001 12:49:33 +0530
Hello ,
I want to update some cred related info in to my parent task_struct
. I am trying to use say currrent->parent_ptr->groups[1] =
current->groups[1] . It does'nt stick in to parent when i control-d. Is
there any special steps to do that ???
Thanx
Regards
Sat
------------------------------
From: Dipl. Ing. Werner =?iso-8859-1?Q?K=FChnert?=
Subject: kernel problem
Date: Wed, 21 Mar 2001 08:22:28 +0100
Hi all,
Can anyone give me a hint what the reason for the following message can
be :
Kernel panic: Attempted to kill the idle task!
In Swapper task - not syncing.
TIA
Regards
Werner Kuehnert
--
Werner Kuehnert Siemens AG Oesterreich PSE ECT IPN 1
E-Mail: [EMAIL PROTECTED]
------------------------------
From: "Gary Cheng" <[EMAIL PROTECTED]>
Crossposted-To:
alt.comp.linux,alt.comp.mail.qmail,comp.lang.java.programmer,comp.mail.sendmail,comp.os.linux.development.apps
Subject: How to create a sendmail user via http request
Date: Wed, 21 Mar 2001 15:28:45 +0800
Hello,
I'm now developing a webmail system base on sendmail, I do programming using
jsp/servelet and Javamail API, I've a problem that I don't know how to
create a user( either sendmail or linux) when somebody apply mail account,
would someone can help me, it'll be highly appreciated!
Regards,
Gary Cheng
------------------------------
From: "Stephanie" <[EMAIL PROTECTED]>
Subject: user-mode linux
Date: Wed, 21 Mar 2001 07:26:17 GMT
Hi,
I'm trying to run user-mode linux for the first time, and I'm having trouble
setting up the virtual network. The how-to says there's a networking utility
called um_eth_net_util but I couldn't find it anywhere. When I try to set up
eth0 in the virtual machine it crashes.
Anybody know how to get around this?
Thanks,
Stephanie
------------------------------
Reply-To: "Nils O. Sel�sdal" <[EMAIL PROTECTED]>
From: "Nils O. Sel�sdal" <[EMAIL PROTECTED]>
Crossposted-To:
alt.comp.linux,alt.comp.mail.qmail,comp.lang.java.programmer,comp.mail.sendmail,comp.os.linux.development.apps
Subject: Re: How to create a sendmail user via http request
Date: Wed, 21 Mar 2001 08:55:47 +0100
"Gary Cheng" <[EMAIL PROTECTED]> wrote in message
news:999lch$k7a$[EMAIL PROTECTED]...
> Hello,
>
> I'm now developing a webmail system base on sendmail, I do programming
using
> jsp/servelet and Javamail API, I've a problem that I don't know how to
> create a user( either sendmail or linux) when somebody apply mail account,
> would someone can help me, it'll be highly appreciated!
use the Runtime.exec(..) to execute a(native) command or script that adds a
user to the system. (adduser e.g)
------------------------------
From: Aminudin Khalid <[EMAIL PROTECTED]>
Subject: mouse device driver programming
Date: Thu, 22 Mar 2001 00:00:48 +0800
Reply-To: [EMAIL PROTECTED]
==============47F89BA8F88EC66A4B6B322C
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
How to resolve this problem ?
I have created a module for my mouse and I 've tried to load it into the
kernel but it gave unresolved symbols like below .Help me
mouse.o: unresolved symbol free_region
mouse.o: unresolved symbol wakeup_interruptible
Thanks.
==============47F89BA8F88EC66A4B6B322C
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
How to resolve this problem ?
<p>I have created a module for my mouse and I 've tried to load it into
the kernel but it gave unresolved symbols like below .Help me
<p>mouse.o: unresolved symbol free_region
<br>mouse.o: unresolved symbol wakeup_interruptible
<br>
<pre>Thanks.</pre>
</html>
==============47F89BA8F88EC66A4B6B322C==
------------------------------
From: [EMAIL PROTECTED] (Mike Dowling)
Subject: Re: unable to mount floppy?
Date: 21 Mar 2001 08:12:01 GMT
On Wed, 21 Mar 2001 06:12:31 GMT, Sean Xu <[EMAIL PROTECTED]> wrote:
>I have installed RedHat Linux 7.0(kernel 2.2.16-22). It works fine before.
>But after installing a few rpm packges. I got the message "wrong major or
>minor number for/dev/fd0...." while trying to mount the floppy drive. What
>is causing the problem?
The message is clear enough; the major number for /dev/fd0 is 2, and the
minor number 0. It is a block device. Check this (ls -l /dev/fd0).
It surprises me immensely that mount should issue such a message; I
would have expected a message to the effect that the superblock could
not be found; I suspect that this is a redhat thingy. You might try to
make access to the floppy without the superfluous file system using tar.
$ tar cvf /dev/fd0 _file_to_be_archived.
See if that works; if it does, then the major and minor numbers are all
right.
>Is there any easy way to fix it without re-installing the whole OS!
$ man mknod
Cheers,
Mike
--
My email address [EMAIL PROTECTED] above is a valid email
address. It is a mail alias. Once spammed, the alias is deleted, and
the integer 'N' incremented. Currently, mike[51,50] are valid. If
email to mikeN bounces, try mikeN+1.
------------------------------
From: "O.Petzold" <[EMAIL PROTECTED]>
Subject: Re: How to use C++ to develope the linux driver module?
Date: Wed, 21 Mar 2001 10:38:58 +0100
"Peter T. Breuer" wrote:
> O.Petzold <[EMAIL PROTECTED]> wrote:
> > "Peter T. Breuer" wrote:
>
> >> <[EMAIL PROTECTED]> wrote:
> >> > Can C++ be used in developing kernel module? and if yes, how
> >>
> >> No (short answer).
> >>
> >> > can I do to fix the problems?
> >>
> >> You can implement C++ memory and exception handling in the kernel.
> >> Etc.
> > do you have the code for exception handling in kernel space ? This is
>
> Do you?
I have a start by ripping it from the gcc-2.95.2 source. I can throw
exceptions,
unfurtuneally I can't catch it. The secret is related to crtstuff, where
the symbols
are initialized. The order seems to be important. I'm not at home at those
low level stage. I started to write I small libkc++, due to the miss time,
it's
not finished yet. Rtti should work as well (never tested). There are some
thinks to be think about - like abort(), to call panic() isn't a solution.
Are you interested in ?
Regards
Olaf
------------------------------
From: "Gianluca Puggelli (Gianluca)" <[EMAIL PROTECTED]>
Subject: Re: Apology for serial porting problems
Date: Wed, 21 Mar 2001 11:19:24 +0000
Reply-To: [EMAIL PROTECTED], [EMAIL PROTECTED]
Josef Allen wrote:
> I would like to apologize if I have offended anyone for putting this
> message up there so many times.
>
> I wish to receive help and not insult anyone
>
> Josef D. Allen
Hello Josef,
I have one link about serial programming on linux/unix:
http://dns.easysw.com/~mike/serial/
enjoy
gianluca
------------------------------
From: Erik de Castro Lopo <[EMAIL PROTECTED]>
Subject: Re: How to use C++ to develope the linux driver module?
Date: Wed, 21 Mar 2001 11:05:57 GMT
O.Petzold wrote:
>
> I have a start by ripping it from the gcc-2.95.2 source. I can throw
> exceptions, unfurtuneally I can't catch it. The secret is related to
> crtstuff, where the symbols are initialized. The order seems to be
> important. I'm not at home at those low level stage. I started to
> write I small libkc++, due to the miss time, it's not finished yet.
> Rtti should work as well (never tested). There are some thinks to be
> think about - like abort(), to call panic() isn't a solution.
Is all this pain worth it? Wouldn't it just be easier to write your
kernel module in C?
> Are you interested in ?
No thanks.
Erik
--
=================================================================
Erik de Castro Lopo [EMAIL PROTECTED] (Yes its valid)
=================================================================
Linux: Because rebooting is for adding new hardware
------------------------------
** 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 by posting to the
comp.os.linux.development.system newsgroup.
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
******************************
