Linux-Development-Sys Digest #169, Volume #8 Mon, 25 Sep 00 04:13:12 EDT
Contents:
vlock and PAM magic? (James Avery)
Re: Kernel panic: VFS :Unable to mount root (Konrad Mierendorff)
Install RH6.2 Directly to RAID ("PAN")
ioremap not in namespce...? (Sean Patrick McNamee)
Re: ioremap not in namespce...? (Pete Zaitcev)
Compresses root filesystem ("Sumit Birla")
Re: new windowing system (Jan D.)
Re: ioremap not in namespce...? (Sean Patrick McNamee)
Re: new windowing system (Alexander Viro)
Re: Kernel panic: VFS :Unable to mount root (Mike McDonald)
Re: Kernel panic: VFS :Unable to mount root (Mike McDonald)
linking help needed (simple Q?) ([EMAIL PROTECTED])
Re: linking help needed (simple Q?) (Frank Sweetser)
user account ("Peter Huang")
Re: undefined symbols in modules ("St. Otto")
Re: URGENT help required - iproute2+tc Quality of Service ("Andrew King")
Re: undefined symbols in modules (Gilles Riffaud)
Re: Link problem (Magnus Therning)
Re: ioremap not in namespce...? (Arne Driescher)
Re: gdb ("Z")
----------------------------------------------------------------------------
From: James Avery <[EMAIL PROTECTED]>
Subject: vlock and PAM magic?
Date: Sun, 24 Sep 2000 18:03:28 +0200
Hi,
For some reason or other, PAMified /usr/bin/vlock does not have to be suid
root to work - using pam_unix_auth.so. This means, that in some sneaky
way, the PAM module is able to read the shadowed passwords in the context
of a process belonging to a regular user.
How on earth is that possible? They're just normal library function calls
and executed with the normal users priviledges and memory space?
Thank's in advance.
--
Med venlig hilsen,
James Avery <[EMAIL PROTECTED]>
------------------------------
From: Konrad Mierendorff <[EMAIL PROTECTED]>
Subject: Re: Kernel panic: VFS :Unable to mount root
Date: Sun, 24 Sep 2000 21:08:08 +0200
This afternoon the holy penguin came to me and opened my eyes, for I had
not recognized that the ext2-support was compiled as a modules. Duh.
------------------------------
From: "PAN" <[EMAIL PROTECTED]>
Subject: Install RH6.2 Directly to RAID
Date: Sun, 24 Sep 2000 21:29:16 +0200
I want to install the RedHat 6.2 directly to a logical drive of a Raid
Array.
Problem is - the RH62's install disk and cd-rom does not have a built in
support for
this model of raid adapter, thaugh after install, I can use the
driver (.c and .h file), compile kernel and it works.
I want (must!) install redhat 6.2 (kernel 2.2.14-50smp) with the
installation diskette straight onto a Mylex Acceleraid 352 (DAC960).
The kernel itself has a built in DAC960.o but an older version of it, which
does not recognize the new 352 (u160-2ch)
I have the new file, and also builded a module DAC960.o (for 2.2.17 thaugh,
is that ok?)
Currently, I installed RH on auxilary drive (IDE) and upgraded kernel to
2.2.17 with the new DAC960 and
I can now see the logical drive of the disk array.
But - i have to get rid of the IDE after i made a successful installation on
the bootable logical drive of the RAID.
>From what i understood - i need to modify the installation diskette with the
new module. is that correct?
Should I also modify the vmlinuz on the disk (its 2.2.14-50) to go with the
module? (built for 2.2.17 or 2.2.16)
if this is the way, I tried that, gunzipped, looped and mounted to the
initrd thats on the disk, gunzipped the modules
file, and saw another modules file, didnt know what to do with it so i put
the dac960.o becides it, modified the
modules.info file, gzipped modules, unmounted, and gzipped initrd. but now i
cant fit the size of initrd in the disk.
i dont know if thats enough to do thaugh, since maybe modifications are to
be done to the vmlinuz too. also i dont know if i did this whole thing
right.
Maybe you know about a way to copy the rh-cdrom to hd, configure it there,
and install on the raid?
P.S. yep, heard about RH7, and i know it has 2.2.17 and thats what i need,
but customer won't go for it.
Please reply to this post, as i desperatly need to bring the system up in 5
hours.
Thanks,
[EMAIL PROTECTED]
------------------------------
From: Sean Patrick McNamee <[EMAIL PROTECTED]>
Subject: ioremap not in namespce...?
Date: Sun, 24 Sep 2000 19:25:00 GMT
Hello. I am trying to write a device driver for a PCI card I helped
develop. I am just starting with it, using O'Reilly's book, and the
IO_MAPPING file in /usr/src/linux/Documentation.
Anyway, I have read that I should map the PCI registers by calling
ioremap(); The problem is that when I try to load my skeleton driver, I
get an error indiciating ioremap is unresolved.
I thought ioremap was part of the core kernel. Do I need to load an
additional kernel module??
Thanks for any help,
Sean McNamee
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Pete Zaitcev)
Subject: Re: ioremap not in namespce...?
Date: Sun, 24 Sep 2000 19:45:28 GMT
> Anyway, I have read that I should map the PCI registers by calling
> ioremap(); The problem is that when I try to load my skeleton driver, I
> get an error indiciating ioremap is unresolved.
> Thanks for any help,
> Sean McNamee
> [EMAIL PROTECTED]
You are doing the right thing. The most common case
of your problem is a kernel compiled with CONFIG_MODVERSIONS.
If you capture the output of the "make modules" you will
see that all supplied modules are compiled with an "-include"
statement that imports necessary defines for CONFIG_MODVERSIONS.
You need to include that file with "-include" or rebuild your
kernel without CONFIG_MODVERSIONS.
--Pete
------------------------------
From: "Sumit Birla" <[EMAIL PROTECTED]>
Subject: Compresses root filesystem
Date: Sun, 24 Sep 2000 20:50:40 GMT
Does anyone know how to instruct the kernel to load a compressed root
filesystem into a ramdisk? The compressed root fs resides on a hard drive
and I would like it to run from /dev/ram0 so that I can spin down the HD
once the system has booted up. I have been able to get this to work when I
boot from a floppy, but no luck getting it to work when booting from a hard
drive.
------------------------------
From: Jan D. <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,comp.windows.x
Subject: Re: new windowing system
Date: Sun, 24 Sep 2000 21:48:57 GMT
In article <[EMAIL PROTECTED]>,
Christopher Browne <[EMAIL PROTECTED]> wrote:
> Hmmm... I thought CDE was essentially an HP 'innovation' that got
> redeployed as an Open Group thing. Or did the committee run rampant
> afterwards?
Not by the looks of things. HP had VUE (Visual User Environment or
something like that) before CDE existed. It looked exactly as CDE
does now, pixel by pixel. I think CDE was just a name change.
Jan D.
------------------------------
From: Sean Patrick McNamee <[EMAIL PROTECTED]>
Subject: Re: ioremap not in namespce...?
Date: Sun, 24 Sep 2000 22:19:21 GMT
Pete Zaitcev wrote:
> > Anyway, I have read that I should map the PCI registers by calling
> > ioremap(); The problem is that when I try to load my skeleton driver, I
> > get an error indiciating ioremap is unresolved.
>
> > Thanks for any help,
> > Sean McNamee
> > [EMAIL PROTECTED]
>
> You are doing the right thing. The most common case
> of your problem is a kernel compiled with CONFIG_MODVERSIONS.
> If you capture the output of the "make modules" you will
> see that all supplied modules are compiled with an "-include"
> statement that imports necessary defines for CONFIG_MODVERSIONS.
> You need to include that file with "-include" or rebuild your
> kernel without CONFIG_MODVERSIONS.
>
> --Pete
Hmmm.... I tried re-compiling the kernel with CONFIG_MODVERSIONS = 0 via
make menuconfig, verified that it was #undef'd in
linxlude/config/modversions.h, and did the
compile the kernel thang, and also tried #inlcuding the linux/version.h in
my own module... neither worked.
Also, shouldn't I be able to see ioremap in the ksyms output??? Methinks my
kernel takes the 'short bus' to school....
Thanks again for your help,
Sean McNamee
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Alexander Viro)
Crossposted-To: comp.os.linux.x,comp.windows.x
Subject: Re: new windowing system
Date: 24 Sep 2000 19:00:19 -0400
In article <[EMAIL PROTECTED]>,
Nix <$}xinix{[email protected]> wrote:
>[EMAIL PROTECTED] (Alexander Viro) writes:
>
>> And yes, I realize that cost of running CDE et.al. outweights that -
>> which is a great reason for not touching them with a 10 feet pole...
>
>It seems (thankfully) that Sun agrees with you.
>
>(They're going to GNOME, but at least that has a little bit of flair.
> CDE, well, design-by-committee run rampant. Looks horrible too, even
> if you try to tart it up.)
Ahem. GNOME is "design by committee and guy who doesn't get UNIX". It's not
a flame towards Miguel - just that his design decisions make me very
uncomfortable. Use of CORBA is one of them, general attitude wrt
orthogonality is another, happy bloating the APIs - one more... Try to
read their codebase, just keep the barfbag ready. And no, KDE is not better.
--
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid. Get yourself a better computer" - Dilbert.
------------------------------
Reply-To: [EMAIL PROTECTED]
From: [EMAIL PROTECTED] (Mike McDonald)
Subject: Re: Kernel panic: VFS :Unable to mount root
Date: Mon, 25 Sep 2000 00:47:11 GMT
In article <[EMAIL PROTECTED]>,
Konrad Mierendorff <[EMAIL PROTECTED]> writes:
> This afternoon the holy penguin came to me and opened my eyes, for I had
> not recognized that the ext2-support was compiled as a modules. Duh.
Mine isn't and I still have the same error.
CONFIG_EXT2_FS=y
Mike McDonald
[EMAIL PROTECTED]
------------------------------
Reply-To: [EMAIL PROTECTED]
From: [EMAIL PROTECTED] (Mike McDonald)
Subject: Re: Kernel panic: VFS :Unable to mount root
Date: Mon, 25 Sep 2000 00:48:10 GMT
In article <[EMAIL PROTECTED]>,
Konrad Mierendorff <[EMAIL PROTECTED]> writes:
> rich wrote:
>>
>> Hi all
>>
>> I have RedHat 6.2 installed on a Pentium machine with level 1 IDE RAID
>> (mirroring) partitioned hard drives, and am now trying to build a
>> monolithic kernel.
>>
>> As part of rebuilding the kernel, I uninstalled the Kernel RPM package with
>>
>> rpm -e --nodeps kernel-headers kernel
>>
>> The following message was reported when trying to boot (I can boot
>> successfully off a boot-stiffy):
>>
>> VFS: cannot open root device 09:06
>> Kernel panic: VFS: Unable to mount root fs on 09:06
> No suggestions, but I have a similar problem:
>
> I'm trying to get my Debian 2.2 (i.e. potato) box to work with kernel
> 2.4.0-test8. However, while booting I get a Kernel panic after the
> partition check. This looks as follows:
> -------------------
> Partitions check:
> hda: hda1 hda2
> hdc: hdc1 hdc2 hdc3 hdc4
> Kernel panic: VFS: Unable to mount root fs on 16:02
> -------------------
> The 2.2.14 kernel does not complain.
Same problem for me with 2.4.0-test7 except mine's on 3:1 (/dev/hda1).
2.2.16 works fine. I haven't bothered trying to chase it down yet.
Mike McDonald
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Subject: linking help needed (simple Q?)
Date: Mon, 25 Sep 2000 11:06:46 +1000
Hi,
This should be a fairly simple question for some of you gurus out
there...
I am wanting to link in the gpg source with the linux kernel source
(2.2.12)so that I can call gpg routines from within the kernel...
apart from why I am wanting to do that and how bad it is and how many
better ways there are to go about it, I don't know much (read nothing)
about the makefiles and how to go about linking the two things together.
Has anyone got any ideas on how to do this or pointers in the right
direction..
Many thanks..
dan.
------------------------------
From: [EMAIL PROTECTED] (Frank Sweetser)
Subject: Re: linking help needed (simple Q?)
Date: 25 Sep 2000 02:21:09 GMT
[EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
>Hi,
>
>This should be a fairly simple question for some of you gurus out
>there...
>
>I am wanting to link in the gpg source with the linux kernel source
>(2.2.12)so that I can call gpg routines from within the kernel...
Not gonna happen - at least, not without rewriting signifigant portions
of gpg. gpg, being a user space application, expects certain thing, such
as having the various userspace libs such as libc, stdin and stout, not
having to deal with hardware or interrupts, virtual memory, a filesystem,
etc.
--
Frank Sweetser rasmusin at wpi.edu, fs at suave.net | $ x 10
Full-time WPI Network Tech, Part time Linux/Perl guy |
"It's Lois! she's in trouble!"
"Wow, Superman, did you look through that building?"
"Well, kinda.. it's glass."
------------------------------
From: "Peter Huang" <[EMAIL PROTECTED]>
Subject: user account
Date: Sun, 24 Sep 2000 21:00:09 -0700
Since I develop my device driver module in the root account, what do I need
to do to enable the users to access the device driver. I had already change
the file permission in the /dev directory but that didn't do a thing. Does
any one have any other ideas?
Peter
------------------------------
From: "St. Otto" <[EMAIL PROTECTED]>
Subject: Re: undefined symbols in modules
Date: Mon, 25 Sep 2000 07:58:41 +0200
Gilles Riffaud wrote:
>
> I have a project involving modules.
> When I compile my modules using "gcc -c -O module.c -DMODULE \
> -D__KERNEL__ -o module.o" and then use insmod to
> insert the module, I get several undefined symbols.
> Why do I get the undefined symbols, and how can I resolve
> them?
>
> Gilles
>
> These are the errors that insmod gives.
>
> acr.o: unresolved symbol open
> acr.o: unresolved symbol read
> acr.o: unresolved symbol select
> acr.o: unresolved symbol __xstat
> acr.o: unresolved symbol close
> acr.o: unresolved symbol write
> acr.o: unresolved symbol fcntl
> acr.o: unresolved symbol ioctl
> acr.o: unresolved symbol gettimeofday
Check the manuals for required libraries for open(), read()...
Append them to the linkers command line with -L.
St. Otto
------------------------------
From: "Andrew King" <[EMAIL PROTECTED]>
Subject: Re: URGENT help required - iproute2+tc Quality of Service
Crossposted-To: comp.os.linux.networking
Date: Mon, 25 Sep 2000 07:04:20 GMT
Steve, you have not created a root cbq qdisc to add the classes to.
you need something like:
tc qdisc add dev eth0 root handle 10: cbq bandwidth 10Mbit avpkt 1000
as the first line. This creates a cbq qdisc and also a cbq class.
Otherwise, you are trying to attach classes to nothing! hope this
helps. (this will create the 10:0 class that you are trying to add classes to).
Andrew King.
------------------------------
From: Gilles Riffaud <[EMAIL PROTECTED]>
Subject: Re: undefined symbols in modules
Date: Mon, 25 Sep 2000 09:14:24 +0200
yes, but how can I send data to a serial port from kernel module whitout these
functions: open,read, write,close
Karl Heyes wrote:
> In article <[EMAIL PROTECTED]>,
> Gilles Riffaud <[EMAIL PROTECTED]> wrote:
> > I have a project involving modules.
> > When I compile my modules using "gcc -c -O module.c -DMODULE \
> > -D__KERNEL__ -o module.o" and then use insmod to
> > insert the module, I get several undefined symbols.
> > Why do I get the undefined symbols, and how can I resolve
> > them?
> >
> > Gilles
> >
> > These are the errors that insmod gives.
> >
> > acr.o: unresolved symbol open
> > acr.o: unresolved symbol read
> > acr.o: unresolved symbol select
> > acr.o: unresolved symbol __xstat
> > acr.o: unresolved symbol close
> > acr.o: unresolved symbol write
> > acr.o: unresolved symbol fcntl
> > acr.o: unresolved symbol ioctl
> > acr.o: unresolved symbol gettimeofday
> >
>
> This looks more like a C program than a kernel module. Remember there isn't
> a libc winthin the kernel.
>
> karl.
------------------------------
From: Magnus Therning <[EMAIL PROTECTED]>
Subject: Re: Link problem
Date: 25 Sep 2000 09:18:46 +0200
Lee Kang Won <[EMAIL PROTECTED]> writes:
> Hi!
>
> I'm having some problem porting an application from on Solaris7 to
> Linux. When I build on Solaris, I linked a shared object and it is
> all I have to do, but on Linux, I have to link all the so files that
> is needed to run the app. Well, this may sound like it's a problem
> with makefie, but what I'm worried about is that I have to replace
> the so file(one of many libraries) with the one some third party is
> to provide which reply on many other objects we don't want to care
> about. for example, In my environment I don't have Sybase client
> module, and I compile an app. with dummy funtions which does almost
> nothing but when it is replaced with the real one afterward, it
> should work fine.. of course with other files that I don't have in
> my system. I want to prevent building everytime I supply program
> with other's solutions. How can I make it on Linux?
I just hope I have understood this correctly. There should be no
problem at all. Usage of shared libraries (.so-files) allow you to
upgrade to newer versions.
If you want to use the same make-file under Linux (that is linking
against only one library) then you need to make sure that lib3rlib.so
is linked against the other libraries. It is possible to have
"chained" libraries in Linux.
/M
--
=======================================================================
Magnus Therning Philips Research Laboratories Eindhoven
Building: WL1.1.50 Prof. Holstlaan 4
Phone: +31 40 2745179 5656 AA Eindhoven
Fax: +31 40 2744004 The Netherlands
mailto:[EMAIL PROTECTED]
------------------------------
From: Arne Driescher <[EMAIL PROTECTED]>
Subject: Re: ioremap not in namespce...?
Date: Mon, 25 Sep 2000 09:33:29 +0200
> Hmmm.... I tried re-compiling the kernel with CONFIG_MODVERSIONS = 0 via
> make menuconfig, verified that it was #undef'd in
> linxlude/config/modversions.h, and did the
> compile the kernel thang, and also tried #inlcuding the linux/version.h in
> my own module... neither worked.
> Also, shouldn't I be able to see ioremap in the ksyms output??? Methinks my
> kernel takes the 'short bus' to school....
>
Hmm, strange problem. What kernel version are you useing?
ioremap is not presend in the 2.0.x series. It was called vremap.
Try to include sysdep.h from the driver book. May be that
will fix your problem.
-Arne
P.S. ioremap should be defined in io.h.
------------------------------
From: "Z" <[EMAIL PROTECTED]>
Subject: Re: gdb
Date: Mon, 25 Sep 2000 09:48:04 +0200
Once upon a while "Paul" <[EMAIL PROTECTED]> wrote:
> Forgot to mention - I did compile/build with GCC, not CC as wrongly stated
> below.
>
>
> Paul <[EMAIL PROTECTED]> wrote in message
> news:8qkapj$7o3$[EMAIL PROTECTED]...
>> I am trying to debug my application using GDB with no luck.
>> So I wrote a simple program and tried debugging it - with still no luck.
>> I have read up on GDB on the RedHat website (using Ver 6.2 of RedHat).
>> Anyway I'm trying to display some variables using the "print" command as
> per
>> doco. after a crash.
>>
>> So here is the code: test.c
>>
>> ====================
>> #include <stdio.h>
>>
>> main ( )
>> {
>> int i, count;
>>
>> count = 0;
>> printf("count=%d\n", count);
>> i = 10 / count;
>> printf("i=%d\n", i);
>> }
>> =======================
>>
>> Here is my compiler command:
>>
>> cc testdb.c -g -otestdb
>>
>> =====================
>>
>> When I run it and it crashes, I type in the following to run GDB
>>
>> gdb -e testdb -c core
>>
>> Then at the GDB prompt I want to view the value of the variable "count".
>> I type in:
>>
>> (gdb) print 'testdb.c'::count
>>
>> The response from GDB is:
>>
>> No symbol "testdb.c" in current context
>>
>>
>> Can someone help me and tell me why I can't view the value of any
> variables
>> ???
>>
>> Thank you in advance
>> Paul
>>
>>
>
>
Hello Paul,
what is the state of gdb when you want to lookup count?
You shopuld do the following:
1) gdb progname
2) set breakpoint near to where you think progname crashes
3) run progname
4) step thru program with 'n' or 's'
5) print count (without a filename! because gdb will only
show the value of count if it is available
in the actual context)
hope this is of any help.
Z
--
LISP is worth learning for the profound enlightenment experience you
will have when you finally get it; that experience will make you a
better programmer for the rest of your days. Eric S. Raymond
------------------------------
** 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
******************************
