Linux-Development-Sys Digest #551, Volume #8 Fri, 9 Mar 01 09:13:12 EST
Contents:
Re: 2.4.2 printing? (Spyros Tsiolis)
Re: How to change the source IP address in UDP Socket ("Michael Mueller")
Re: How to change the source IP address in UDP Socket ("jacob navia")
Re: finding memory mapped devices ("Lee Ho")
syscall with number > 191 in the 2.2.17 kernel (root)
Re: PLEASE HELP! How to read the structure of a floppy disk at low (Frank Ranner)
syscall with number > 191 in the 2.2.17 kernel (root)
Re: does anybody successfully update rpm 3 to rpm 4 in redhat 6.22? (Frank Ranner)
syscall with number > 191 in the 2.2.17 kernel (Julien Cavoizy)
syscall problem (Mikael Chambon)
Re: How to change the source IP address in UDP Socket ([EMAIL PROTECTED])
Re: New OS want (Heiko Klein)
strace ("Bjoern Brencher")
Re: strace (Josef Moellers)
Re: Kernel 2.4.2: aic7xxx trouble (Ralf Kleineisel)
Re: Processor ID (Kasper Dupont)
----------------------------------------------------------------------------
From: Spyros Tsiolis <[EMAIL PROTECTED]>
Subject: Re: 2.4.2 printing?
Date: Fri, 09 Mar 2001 09:07:55 +0000
http://www.cups.org
s.
Harmon Seaver wrote:
> Is anyone able to print with 2.4.2?
>
> --
> Harmon Seaver, MLIS
> CyberShamanics
> (920) 233-5820 [EMAIL PROTECTED]
------------------------------
From: "Michael Mueller" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: How to change the source IP address in UDP Socket
Date: Fri, 9 Mar 2001 10:18:23 +0100
Hi Matt,
you wrote:
> You will open a raw IP socket, which when you send to, will send to raw
IP:
> no transport layer protocol. You still are opening an IP socket, hence
> the type of the socket being AF_INET.
> The way to open a socket for sending "RAW" link-layer frames is somewhat
> different. Try "man packet".
Have a look on using the IP_HDRINCL socket option. This is the usual way
taken to fake the IP header. The nice point about it are:
- you do not need to calculate the checksum but thats done from the IP stack
and
- it is portable.
Malware
PS: Sorry Matt, again hitted the brain-dead "answer"-Button of Outlook
sending mail instead of a news message.
------------------------------
From: "jacob navia" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: How to change the source IP address in UDP Socket
Date: Fri, 9 Mar 2001 11:27:51 -0800
So, somebody wants to build software for making nasty/illegal things, and
there are immediately people here that will tell him how to do this with all
the needed details.
How can people here be so naive?
Or is it linux THE system to build hacks such as this one?
<[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
>
> I'm writing a UDP routing program which requires sending out UDP
> packets with multiple IP address other than the host address. I
> understanding this is not supported in Berkeley socket API. Is there
> anyway to do this in Linux?
>
> Thanks so much.
>
> w
>
------------------------------
From: "Lee Ho" <[EMAIL PROTECTED]>
Subject: Re: finding memory mapped devices
Date: Fri, 09 Mar 2001 10:44:49 GMT
AFAIK, kernel doesn't not maintain i/o memory region.
Kernel offers no function for reserving memory region like other
resource, (eg, request_region() for i/o port).
*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*
Lee, Ho. Software Engineer, Embedded Linux Dep, LinuxOne
Mail : [EMAIL PROTECTED] (work), [EMAIL PROTECTED] (personal)
Homepage : http://flyduck.com, http://linuxkernel.to
Lee Jenkins Write:
>I'm writing a program to identify devices associated with one or more of
>the following resources: I/O ports, DMA, IRQ, or I/O memory. The first
>three are easily found by examining /proc/ioports, /proc/dma, and
>/proc/interrupts. But I haven't found an analogous listing for I/O
>memory mapped devices. The closest thing I've found is in /proc/pci --
>but, of course, that list doesn't show mappings for non-PCI devices.
> Does the kernal keep a list of I/O memory mapped devices? If so,
>how is it accessed?
------------------------------
From: root <[EMAIL PROTECTED]>
Subject: syscall with number > 191 in the 2.2.17 kernel
Date: Fri, 09 Mar 2001 12:02:17 +0100
hi,
I am working into the kernel 2.2.17 and i notice that there are syscall
in asm/unistd.h which number are not written in the entry.S file.
Is it normal ?
A friend of mine use the 2.2.18 kernel and don't have these sys_call ...
strange no ?
thanks for your answer.
Julien Cavoizy
Cronos
======Unistd.h======
...
/* #define __NR_ugetrlimit 191 SuS compliant getrlimit */
#define __NR_mmap2 192
#define __NR_truncate64 193
#define __NR_ftruncate64 194
#define __NR_stat64 195
#define __NR_lstat64 196
#define __NR_fstat64 197
=======entry.S=========
.long SYMBOL_NAME(sys_sendfile)
.long SYMBOL_NAME(sys_ni_syscall) /* streams1 */
.long SYMBOL_NAME(sys_ni_syscall) /* streams2 */
.long SYMBOL_NAME(sys_vfork) /* 190 */
/*
* NOTE!! This doesn't have to be exact - we just have
* to make sure we have _enough_ of the "sys_ni_syscall"
* entries. Don't panic if you notice that this hasn't
* been shrunk every time we add a new system call.
*/
.rept NR_syscalls-190
.long SYMBOL_NAME(sys_ni_syscall)
.endr
=======================
------------------------------
From: Frank Ranner <[EMAIL PROTECTED]>
Subject: Re: PLEASE HELP! How to read the structure of a floppy disk at low
Date: Fri, 09 Mar 2001 21:58:21 +1100
John wrote:
>
> Hi,
> Please help! I need some information about the structure of a floppy and
> how to read it at a low level.
How about:
The fdutils package contains utilities for configuring and debugging the
Linux floppy driver, for formatting extra capacity disks (up to 1992K on
a high density disk), and for sending raw commands to the floppy
controller.
Fdutils includes the following items:
* superformat: formats high capacity disks of (up to 1992k for high
density disks or up to 3984k for extra density disks)
* fdmount: automatically mounts/unmounts disks when they are
inserted/removed.
* xdfcopy: formats, reads and writes OS/2's XDF disks.
* MAKEFLOPPIES: creates the floppy devices in /dev
* getfdprm: prints the current disk geometry (number of sectors,
track and heads etc)
* setfdprm: sets the current disk geometry
* fdrawcmd: sends raw commands to the floppy driver
* floppycontrol: configure the floppy driver
* General documentation about the floppy driver
get it from http://fdutils.linux.lu/
Regards, Frank Ranner
------------------------------
From: root <[EMAIL PROTECTED]>
Subject: syscall with number > 191 in the 2.2.17 kernel
Date: Fri, 09 Mar 2001 12:03:57 +0100
hi,
I am working into the kernel 2.2.17 and i notice that there are syscall
in asm/unistd.h which number are not written in the entry.S file.
Is it normal ?
A friend of mine use the 2.2.18 kernel and don't have these sys_call ...
strange no ?
thanks for your answer.
Julien Cavoizy
Cronos
======Unistd.h======
...
/* #define __NR_ugetrlimit 191 SuS compliant getrlimit */
#define __NR_mmap2 192
#define __NR_truncate64 193
#define __NR_ftruncate64 194
#define __NR_stat64 195
#define __NR_lstat64 196
#define __NR_fstat64 197
=======entry.S=========
.long SYMBOL_NAME(sys_sendfile)
.long SYMBOL_NAME(sys_ni_syscall) /* streams1 */
.long SYMBOL_NAME(sys_ni_syscall) /* streams2 */
.long SYMBOL_NAME(sys_vfork) /* 190 */
/*
* NOTE!! This doesn't have to be exact - we just have
* to make sure we have _enough_ of the "sys_ni_syscall"
* entries. Don't panic if you notice that this hasn't
* been shrunk every time we add a new system call.
*/
.rept NR_syscalls-190
.long SYMBOL_NAME(sys_ni_syscall)
.endr
=======================
------------------------------
From: Frank Ranner <[EMAIL PROTECTED]>
Subject: Re: does anybody successfully update rpm 3 to rpm 4 in redhat 6.22?
Date: Fri, 09 Mar 2001 22:08:25 +1100
The answer is to install rpm 3.0.5 which understands rpm4 packages, but
still uses
the rpm 3 database. I did the same as you and found all the exisitng
packages 'gone'
but all that really happened is that rpm 4 uses a separate set of files
for its db.
Once I reverted to rpm 3, my package db was OK again. Since then I have
installed some
rpm4 packages that I have downloaded, subject to prereqs of course.
Regards, Frank Ranner
Tim Roberts wrote:
>
> "hushui" <[EMAIL PROTECTED]> wrote:
>
> >Since it is rpm ver 4 in redhat 7.0,and many package is packaged by rpm 4
> >and they can not be installed in rpm 3 (redaht 6.2). But I don't like
> >redhat 7.0 ,because there is so many problem in gcc of redhat 7.0.
> >How to update rpm 3 to rpm 4??
>
> I hope somebody has an answer to this. I installed RPM version 4 because
> one of the packages I wanted was in version 4 format. I foolishly assumed
> the RPM database would be carried forward. It wasn't. Now, RPM 4 knows
> nothing about what is already installed, so I always have to use --nodeps
> when I install, which kinda defeats the purpose.
>
> If anyone can tell me how to import the RPM 3 database into RPM 4, post
> mortem, I would be eternally grateful.
> --
> - Tim Roberts, [EMAIL PROTECTED]
> Providenza & Boekelheide, Inc.
------------------------------
From: Julien Cavoizy <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: syscall with number > 191 in the 2.2.17 kernel
Date: Fri, 09 Mar 2001 12:23:03 +0100
hi,
I am working into the kernel 2.2.17 and i notice that there are syscall
in asm/unistd.h which number are not written in the entry.S file.
Is it normal ?
A friend of mine use the 2.2.18 kernel and don't have these sys_call ...
strange no ?
thanks for your answer.
Julien Cavoizy
Cronos
======Unistd.h======
...
/* #define __NR_ugetrlimit 191 SuS compliant getrlimit
*/
#define __NR_mmap2 192
#define __NR_truncate64 193
#define __NR_ftruncate64 194
#define __NR_stat64 195
#define __NR_lstat64 196
#define __NR_fstat64 197
=======entry.S=========
.long SYMBOL_NAME(sys_sendfile)
.long SYMBOL_NAME(sys_ni_syscall) /* streams1 */
.long SYMBOL_NAME(sys_ni_syscall) /* streams2 */
.long SYMBOL_NAME(sys_vfork) /* 190 */
/*
* NOTE!! This doesn't have to be exact - we just have
* to make sure we have _enough_ of the "sys_ni_syscall"
* entries. Don't panic if you notice that this hasn't
* been shrunk every time we add a new system call.
*/
.rept NR_syscalls-190
.long SYMBOL_NAME(sys_ni_syscall)
.endr
=======================
------------------------------
From: Mikael Chambon <[EMAIL PROTECTED]>
Subject: syscall problem
Date: Fri, 09 Mar 2001 12:24:14 +0100
Hi,
I 've just put my new system call to a kernel 2.2.18 and I've got some
problem with it now.
As there is a printk inside, at boot and all the time my syscall is
called each time that I executed a program
I don't know what is going on, here is what I did:
Here is my system call
==== systest.c ====
#include <kernel.h>
asmlinkage int sys_systest(int val)
{
printk("systest called with %d\n", val);
return (0);
}
==== end systest.c ====
I went in arch/i386/kernel/entry.S (in your kernel tree) and add new entry:
.long SYMBOL_NAME(sys_systest) /* 191 */
and change .rept NR_syscalls-190 to .rept NR_syscalls-191
And then I went in include/asm/unistd.h (in your kernel tree) and add a new entry:
#define __NR_systest 191
And when I rebooted on my new kernel I've got the problem,
Did I miss something ??
Thanls for all.
--
Mikael Chambon
[EMAIL PROTECTED] || [EMAIL PROTECTED]
ICQ 10249913 || http://www.cronos.org
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.development.apps
Subject: Re: How to change the source IP address in UDP Socket
Date: Fri, 9 Mar 2001 11:56:08 +0100
In comp.os.linux.development.system jacob navia <[EMAIL PROTECTED]> wrote:
> So, somebody wants to build software for making nasty/illegal things, and
> there are immediately people here that will tell him how to do this with all
> the needed details.
> How can people here be so naive?
> Or is it linux THE system to build hacks such as this one?
It is much better for anyone to know that one can arbitrarily publish
any IP number and ethernet adress on his network card, so that
security does not rely on misconceptions. You may be sure that hackers
know that very well, only naive people falling in the trap. Also
these "hacks" can be done on other machines, particularly Suns.
You can even routinely change ethernet adress on the prom of most cards.
There exist programs on D. Becker site to do that, and it is standard
on Suns. Fortunately this allows to still continue running programs
which check the hardware adress when you change machine.
Also the "hack" is mentioned in man ifconfig in Linux:
hw class address
Set the hardware address of this interface, if the
device driver supports this operation. The keyword
must be followed by the name of the hardware class
and the printable ASCII equivalent of the hardware
address. Hardware classes currently supported
include ether (Ethernet), ax25 (AMPR AX.25), ARCnet
and netrom (AMPR NET/ROM).
--
Michel Talon
------------------------------
From: Heiko Klein <[EMAIL PROTECTED]>
Crossposted-To: com.sys.hp.hpux,comp.sys.sun.admin,comp.os.ms-windows.misc
Subject: Re: New OS want
Date: 09 Mar 2001 13:14:21 +0100
Freelancer <[EMAIL PROTECTED]> writes:
>
> I am an UNIX system admin. for many years, experiences in SUN Solaris,
> HP-UX, SGI Iris, Redhat Linux, etc. No experiences in Windows NT.
> Also Windows '95/98/2000 user. Not a Mac user.
> All current OS's are take long time to boot and shutdown.
> Sometimes I can't shutdown in Windows '95/98/2000, I have to
> turn the power off and get the blue screen in next time boot up.
>
> In the 80's SUN came up the diskless client. Later we also have
> X-Terminal
> (X-server). But those can only work in LAN, UNIX environment.
> Today Windows NT shared software on LAN in MS-Windows environment.
An they needed a server, which boots a long time ...
> I am looking for a OS which is NOT a computer OS.
> 0. Intel base PC.
> 1. takes less than 30 seconds to boot and shutdown, just like TV, VCR,
> DVD player, can turn of the power any time without corrupt the
> disk.
>
> 2. Only can do
> a.Watch TV, play DVD/CD/VCD(video/audio), games.
> b. Web browser (may be can do email, may be not).
> c. Can connect with LAN (Ethernet), Cable (broadband), Dial-up (PPP,
> DSL),
> Fiber Optic (future use).
For games, on PC hardware, you need Winows!
> 3. Can use any select software or application programs currently run in
> Windows,
> UNIX (May be JavaOS). I don't want to develop new app software for
> new
> OS. Avoid
> Steve Job's (CEO of Apple) mistake.
Cool way of thinking. You will need all libraries for any OS and some
kind of emulator for them, without loading it from a disk.
> 3. can NOT write to disk, can NOT download to disk, only can write to
> RW-DVD drive (avoid virus) for download. Also can NOT execute
> any file from DVD drive, read only for video/audio, image, data
> files.
You can have viruses on the RW-DVD. But think about changable disks.
> 4. Can NOT do software upgrade online in any site, only can do software
> from
> few trust destination sites.
Thats only configuration. AFAIK, every OS but windows is configured
this way by default
A fast and 'multimedia' based OS is BeOS. Booting in less than 20s it
is closest to your wishes.
Heiko
------------------------------
From: "Bjoern Brencher" <[EMAIL PROTECTED]>
Subject: strace
Date: Thu, 8 Mar 2001 14:10:39 +0100
Hello,
i want to start a script file with strace so that i can trace all processes
which are started by the script!
But when i start the script with:
strace ./scirptname
i always get an errormessage:
execve("./scriptname", ["./scriptname"], [/* 55 vars */]) =0
strace: exec: Exec format error
If i use strace with the executables which are started by the script
everything goes fine!
Someone knows help?
------------------------------
From: Josef Moellers <[EMAIL PROTECTED]>
Subject: Re: strace
Date: Fri, 09 Mar 2001 14:47:49 +0100
Bjoern Brencher wrote:
> =
> Hello,
> =
> i want to start a script file with strace so that i can trace all proce=
sses
> which are started by the script!
> =
> But when i start the script with:
> =
> strace ./scirptname
> =
> i always get an errormessage:
> =
> execve("./scriptname", ["./scriptname"], [/* 55 vars */]) =3D0
> strace: exec: Exec format error
> =
> If i use strace with the executables which are started by the script
> everything goes fine!
> =
> Someone knows help?
What does the first line of your script look like?
It should look like this:
#! /bin/bash -
Also, you should use the "-f" flag:
man strace
Josef
-- =
Josef M=F6llers (Pinguinpfleger bei FSC)
If failure had no penalty success would not be a prize
-- T. Pratchett
------------------------------
From: Ralf Kleineisel <[EMAIL PROTECTED]>
Subject: Re: Kernel 2.4.2: aic7xxx trouble
Date: Fri, 09 Mar 2001 14:50:15 +0100
Roy Batty wrote:
> I'm having problems using Kernel 2.4.2 on a machine with
> an Adaptec 2940-U2W Controller. When the kernel mounts
> the root-drive (a reiserfs-partition on a ST118273LW-drive)
> I get SCSI-timeouts which end in a bus reset.
I tried 2.4.2 here, got problems with aic7xxx, too. It wouldn't boot, it
stops with a 'Unable to mount root fs on...'. Then I switched to the old
aix7xxx driver, this one works fine.
--
Ralf Kleineisel
------------------------------
From: Kasper Dupont <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.hardware
Subject: Re: Processor ID
Date: Fri, 09 Mar 2001 14:07:05 +0000
Arne Driescher wrote:
>
> Kasper Dupont wrote:
> >
> > Arne Driescher wrote:
> > >
> > [...]
> > >
> > > Lets say that it is much more up to you
> > > what MAC a program "sees" than it is for CPUids.
> > > (hardwareid.sourceforge.net)
> >
> > Couldn't you do the same with CPUids?
> >
> > I presume queries traps into kernel mode.
> >
> As far as I understand Intels specs CPUID will not
> trap. This seems logical to me because the returned
> values of CPUID depend on what you put into EAX
> and different CPUs support different values for EAX.
> So, illegal values for EAX simply return undefined
> results and don't trap. Only processors not supporting
> CPUID at all will raise an ill. op. exception.
>
> -Arne
Too me that sounds like a bad design, a program
runing in user space should not have access to
that information without the kernels knowledge.
--
Kasper Dupont
------------------------------
** 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
******************************
