Linux-Development-Sys Digest #34, Volume #8 Thu, 20 Jul 00 08:13:11 EDT
Contents:
Help with kernel level sockets? (Michael Mesnier)
Re: recompile kernel...?? (A Guy Called Tyketto)
Re: mpeg capture from video capture card (David Konerding)
Re: Database Modeling Tools for Linux? (Harry George)
After compile Kernel NFS file system does't work. (Seung-su Yang)
Re: Question about after upgrading kernel (Wesley Wong)
SC400 bootloader (Jeff Say)
Re: C problem ("Eric Co")
Re: recompile kernel...?? (Tim Moore)
Re: Question about after upgrading kernel ([EMAIL PROTECTED])
Re: Question about after upgrading kernel (Tim Moore)
Re: After compile Kernel NFS file system does't work. (Rob Kroll)
Caldera Wabi does not work in 2.2.16 (Nobody)
Re: (Q) Capturing everything at boot time (Kevin Lacquement)
Problems compiling Kernel 2.0.38 with gcc 2.95.2 (root)
Compiling a 2.0.38 Kernel with gcc 2.95.2 ? (Robert Resch)
Re: How to force a clean reboot from a module? (Scott J. Bertin)
Re: My own distribution (=?iso-8859-1?Q?Rasmus_B=F8g_Hansen?=)
Newbie module problem ("illu")
glibc-2.1.3 problems (Mike Dowling)
Re: glibc-2.1.3 problems (Andreas Jaeger)
Re: Caldera Wabi does not work in 2.2.16 (Uwe Bonnes)
----------------------------------------------------------------------------
From: Michael Mesnier <[EMAIL PROTECTED]>
Subject: Help with kernel level sockets?
Date: Wed, 19 Jul 2000 15:07:44 -0700
I'm running into trouble using sockets from within a module. I have
successfully established a connection using
sock->ops->bind/listen/accept(), but fail on sock->ops->sendmsg/recvmsg
because of a bad address (EFAULT). Here's a snippet of my code that was
obtained by example from similar modules such as nbd.c and nfs:
ptr = kmalloc(1024, GFP_KERNEL);
iov.iov_base = ptr;
iov.iov_len = 1024;
msg.msg_name = NULL;
msg.msg_namelen = 0;
msg.msg_iov = &iov;
msg.msg_iovlen = 1;
msg.msg_control = NULL;
msg.msg_controllen = 0;
msg.msg_namelen = 0;
msg.msg_flags = 0;
if ((err=sock_sendmsg(dupsock, &msg, 1024))<0) {
printk("***ERROR SENDING: %i***\n", err);
}
The error I keep receiving is -EFAULT, which ultimately means the *user*
address that tcp is trying to csum_and_copy_from_user() is not valid.
How can I get around the fact that tcp.c is expecting a user level
buffer to copy data from?
Thanks for any help,
Mike Mesnier
Enterprise Architecture Laboratory
------------------------------
From: A Guy Called Tyketto <[EMAIL PROTECTED]>
Subject: Re: recompile kernel...??
Date: Wed, 19 Jul 2000 22:37:38 GMT
=====BEGIN PGP SIGNED MESSAGE=====
Hash: SHA1
Wesley Wong <[EMAIL PROTECTED]> wrote:
> If I find out that I have a option unchecked in previous make menuconfig,
> that I need it check now, do I need to recompile the whole kernel again?
> (meaning run "make menuconfig dep clean bzImage modules modules_install")
> Is there a fast and convenient way to do it rather than that because
> compiling kernel takes time.... if I just need to reconfigure a little
> part of kernel is there a way to do so? I am asking in general. Thanks for
> teaching this newbie.
This, unfortunately, you will have to do, unless the option you
have checked, can be compiled as a module. If it can, you can give it a
try, but I would advise against it, and recompile the kernel to add in
the support that you need.
BL.
- --
Brad Littlejohn | Email: [EMAIL PROTECTED]
Unix Systems Administrator, | [EMAIL PROTECTED]
Web + NewsMaster, BOFH.. Smeghead! :) | http://www.wizard.com/~tyketto
PGP: 1024D/E319F0BF 6980 AAD6 7329 E9E6 D569 F620 C819 199A E319 F0BF
=====BEGIN PGP SIGNATURE=====
Version: GnuPG v1.0.2 (FreeBSD)
Comment: For info see http://www.gnupg.org
iD8DBQE5di2fyBkZmuMZ8L8RAsW2AKDPfapuVBAHRAV14sUJuUNY1aZKkQCfWDfc
lULVmx/bguqbI0Smx63E6dM=
=LCHM
=====END PGP SIGNATURE=====
------------------------------
From: [EMAIL PROTECTED] (David Konerding)
Crossposted-To: comp.os.linux.misc,comp.os.linux.hardware
Subject: Re: mpeg capture from video capture card
Date: 19 Jul 2000 23:11:59 GMT
On Sun, 16 Jul 2000 19:31:30 -0700, blowfish <[EMAIL PROTECTED]>
wrote:
>Thierry wrote:
>>
>> Hi,
>>
>> Do you know an existing program (with sources) wich capture video from a
>> video capture card (bt chips) and save it in mpeg or quicktime format.
>> Without using Xwindows.
>>
>> Thanks a lot.
>>
>> thierry
>
>broadcast2000.
>
bttvgrab. It's excellent. Logs to screen via curses or dumb terminal.
Contains everything you need to save MPEG format files with audio!!
------------------------------
From: Harry George <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Database Modeling Tools for Linux?
Date: 19 Jul 2000 16:22:00 -0700
I've been using Dia 0.85 for quite a few UML models. Works great,
with a few little inconveniences.
1. We need a way to add more handles to the boxes (maybe there is a
way and I haven't found it yet).
2. Aggregates (diamond at the end of a line) sometimes don't hook
properly to the object (they might even point backwards), so you
have to juggle them around.
3. Adding points to polylines and splines is not obvious.
4. I'd like python scripting, but just getting SVG out is good enough
for now. (export...manipulate...import)
5. I can't get it installed on an SGI because I can't get the full
gnome infrastructure instaled -- I wish Dia was just gtk.
"Frank V. Castellucci" <[EMAIL PROTECTED]> writes:
> Thomas Gagne wrote:
> >
> > Does anyone know of any modeling tools like Embarcadero's ER/Studio or CA's
> > ER/Win for Linux. I hate having to flip-flop between Linux and Windows just
> > to use a modeling tool. It's enough to make me look at VMWare...
> >
> > --
> > For Open Source Middleware Visit http://home.netcom.com/~tgagne
>
> Try:
>
> Dia
> ArgoUML (I know, but in a fix...)
>
> --
> Frank V. Castellucci
> http://corelinux.sourceforge.net
> OOA/OOD/C++ Standards and Guidelines for Linux
> http://PythPat.sourceforge.net
> Pythons Pattern Package
--
Harry George
[EMAIL PROTECTED]
------------------------------
From: Seung-su Yang <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help
Subject: After compile Kernel NFS file system does't work.
Date: Thu, 20 Jul 2000 07:38:30 -0400
Hi.
I have redhat 6.1 with 2.2.14 kernel which come with the distribution.
I compiled the kernel for using PSET.
When I choose NFS as an module, the system could not mount the NFS
filesystem.
It says ***unknown file system type ***.
But if I choose NFS include in kernel, it work's correctly. (I used
"make xconfig" for configuring the kernel and just turn on SMP and
choose "M" for NFS. for module. I choose "Y" for NFS to include the NFS
in kernel)
I already check the module of NFS with the command "lsmod" when I booted
kernel with NFS as a module.
Tt says that NFS module loaded.
Also, the previous kernel which comes with RedHat 6.1 distribution,
works well with the same configuration.
(I just choose the kernel at boot time and I think the kernel with
distribution also use NFS as an module.)
How could I compile and install the kernel with NFS as an module?
Thanks in advance
------------------------------
From: Wesley Wong <[EMAIL PROTECTED]>
Subject: Re: Question about after upgrading kernel
Date: Thu, 20 Jul 2000 01:30:08 GMT
> >2. After I got the whole thing working (being able to boot with the new
> >image) is it a good idea to remove the /usr/src/linux tree? If not the
>
> No, because the header files are needed. There are symbolic links
> in /usr/include that point there. If you blow away the whole kernel
source
> tree, you won't be able to compile userland things.
But what if I don't even have the source (if I just upgrade using kernel-
2.2.16-3.rpm) before... why does that work , while once compiled the
source, we have to keep it?
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: Jeff Say <[EMAIL PROTECTED]>
Subject: SC400 bootloader
Date: Wed, 19 Jul 2000 22:26:49 -0400
I have sc400 (486) based embedded PC AT board,4M Flash and 16M DRAM SBC.
I do not
want to use any BIOS to load Linux. Board supports headless operation
(no keyboard, no display, no floppy/hard disk) - only serial port can be
used as a console). I need a bootloader to boot Linux. I am trying to
find a bootloader which graps binary linux image and copies to DRAM,
then boots from there.
does anyone knows how to that? if you interested in this project, I will
pay your consultation fee!!
serious guys only!!
TIA,
Jeff Say
------------------------------
From: "Eric Co" <[EMAIL PROTECTED]>
Subject: Re: C problem
Date: Thu, 20 Jul 2000 10:20:18 +0800
thanx henry,
and is there any alternative function to do it as the strtok do?
thanx again,
Best regards,
eric co
?????????????????????????????????????????
? Living is easy with eyes closed, ?
? misunderstanding all you see. ?
?????????????????????????????????????????
"Harold Henry" <[EMAIL PROTECTED]> wrote in message
news:8l43a5$jpv$[EMAIL PROTECTED]...
> [Posted and mailed]
>
> Do a man on strtok
>
> BUGS
> Never use this function. This function modifies its first
> argument. The identity of the delimiting character is
> lost. This function cannot be used on constant strings.
------------------------------
From: Tim Moore <[EMAIL PROTECTED]>
Subject: Re: recompile kernel...??
Date: Thu, 20 Jul 2000 03:12:49 GMT
> that I need it check now, do I need to recompile the whole kernel again?
> (meaning run "make menuconfig dep clean bzImage modules modules_install")
'make clean' is the part that removes all the .o files from the last
compile.
--
timothymoore
bigfoot
com
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Question about after upgrading kernel
Date: Thu, 20 Jul 2000 03:14:07 GMT
Wesley Wong <[EMAIL PROTECTED]> wrote:
> But what if I don't even have the source (if I just upgrade using kernel-
> 2.2.16-3.rpm) before... why does that work , while once compiled the
> source, we have to keep it?
If you built it yourself, you should keep the source tree around
(doing a make clean to clean it out). If you used the rpms, there's a
seperate kernel-headers one which contains just the kernel headers for
a given version.
--
Matt Gauthier <[EMAIL PROTECTED]>
------------------------------
From: Tim Moore <[EMAIL PROTECTED]>
Subject: Re: Question about after upgrading kernel
Date: Thu, 20 Jul 2000 03:17:34 GMT
http://www.linuxdoc.org/HOWTO/Kernel-HOWTO.html
There is a small base level of knowledge required.
--
timothymoore
bigfoot
com
------------------------------
Crossposted-To: comp.os.linux.help
Subject: Re: After compile Kernel NFS file system does't work.
From: [EMAIL PROTECTED] (Rob Kroll)
Date: Thu, 20 Jul 2000 03:54:37 GMT
[EMAIL PROTECTED] (Seung-su Yang) wrote in <[EMAIL PROTECTED]>:
>Hi.
>
>I have redhat 6.1 with 2.2.14 kernel which come with the distribution.
>
>I compiled the kernel for using PSET.
>
>When I choose NFS as an module, the system could not mount the NFS
>filesystem.
>
Did you remember to activate the Kernel Module Loader? By default (under
make menuconfig, at least) it isn't active, and the system will not be able
to load the modules.
Try compiling in the Kernel Module Loader (Under Loadable Module Support),
and see if it works.
If not, then why not compile in the NFS module? You know that it works that
way, and using bzImage instead of zImage, the size difference is
negligible.
HTH
-RK
rkroll(at)ottawa(dot)com
------------------------------
From: [EMAIL PROTECTED] (Nobody)
Subject: Caldera Wabi does not work in 2.2.16
Date: Thu, 20 Jul 2000 05:22:24 GMT
Caldera Wabi 2.2 gives a "exception D in WIN_INST" when I try to use
it under kernel 2.2.16.
------------------------------
From: Kevin Lacquement <[EMAIL PROTECTED]>
Subject: Re: (Q) Capturing everything at boot time
Date: Thu, 20 Jul 2000 05:35:27 GMT
Timothy Murphy wrote:
>
> dmesg doesn't capture everything that flashes by at boot-time,
> eg inconsistencies between modules.
> Is there any simple way -- a kind of kernel "script" --
> which will save everything that appears on the screen at boot-up?
>
> --
cat /var/log/bootlog
------------------------------
From: root <[EMAIL PROTECTED]>
Subject: Problems compiling Kernel 2.0.38 with gcc 2.95.2
Date: Thu, 20 Jul 2000 10:12:47 +0200
Hi!
I want to compile a 2.0.38 Kernel with the 2.92.2 Version of the GCC.
The reason: I need the kernel for a small system with not much ram. The
functionality of the 2.0.38 Kernel is enough - also the size. The 2.95.2
Compiler makes much more and better optimizations and a smaller code so
that there will be mor free memory and the system will run faster.
The target-System has only 8 MBytes of ram and a 4 MB Ramdisk -> 4 MB
Ram.
So every saved byte would be great.
------------------------------
From: Robert Resch <[EMAIL PROTECTED]>
Subject: Compiling a 2.0.38 Kernel with gcc 2.95.2 ?
Date: Thu, 20 Jul 2000 10:51:19 +0200
Hi!
I want to compile a 2.0.38 Kernel with the 2.92.2 Version of the GCC.
The reason: I need the kernel for a small system with not much ram. The
functionality of the 2.0.38 Kernel is enough - also the size. The 2.95.2
Compiler makes much more and better optimizations and a smaller code so
that there will be mor free memory and the system will run faster.
The target-System has only 8 MBytes of ram and a 4 MB Ramdisk -> 4 MB
Ram.
So every saved byte would be great.
------------------------------
From: [EMAIL PROTECTED] (Scott J. Bertin)
Subject: Re: How to force a clean reboot from a module?
Date: 20 Jul 2000 09:52:08 GMT
Reply-To: [EMAIL PROTECTED]
[Posted and mailed]
In article <cZjd5.6922$[EMAIL PROTECTED]>,
"Norm Dresner" <[EMAIL PROTECTED]> writes:
> I'm in the middle of developing a device driver/module for a board and I
> occasionally run into situations in which the driver knows that its data
> structures are fouled up and (probably) can't be removed with rmmod -f.
>
> Is there some way for a driver to force a safe/clean reboot which includes
> the normal sync-mechanism and unmounting of file systems?
Even if there is, I would strongly urge you not to do it. You don't know
what else might be running on the machine. I'd hate for a week long
simulation aborted 5 minutes before completion because your driver thought
there was a problem. Just log an error message and fail all accesses to
the device after that point.
Scott J. Bertin
[EMAIL PROTECTED]
------------------------------
From: =?iso-8859-1?Q?Rasmus_B=F8g_Hansen?= <[EMAIL PROTECTED]>
Subject: Re: My own distribution
Date: Thu, 20 Jul 2000 12:26:25 +0200
On Tue, 18 Jul 2000, Kevin Lacquement wrote:
> Mike Dowling wrote:
> > <SNIP>
> >
> > In a sense, I have always had my own distribution. I.e., I have never
> > used a distribution. The advantage for me is obvious; I know where
> > everything is and how it is configured.
> <SNIP>
>
> I've been thinking of doing this. One question, though. How do you get
> started without a distribution to compile stuff on?
>
> Cheers,
> Kevin
The RedHat kickstart is another option. Install once - make a kickstart -
then there would not be so many configuration options. But of course that
is still a RedHat compiled system...
Rasmus B. Hansen
------------------------------
From: "illu" <[EMAIL PROTECTED]>
Subject: Newbie module problem
Date: Thu, 20 Jul 2000 13:22:25 +0200
Hi
I have tried to work with the simple module from oreilly book...
#define MODULE
#include <linux/module.h>
int init_module(void) {printk("<1>Hello, world\n"); return 0;}
void cleanup_module(void) {printk("<1>Goodbye cruel world\n"; } }
compiling goes fine
When i try to load the module with
insmod hello.o
i get
hello.o: unresolved symbol printk
What the .... that means?? Something wrong with the libraries perhaps?
I have gcc 2.95.3.
------------------------------
From: [EMAIL PROTECTED] (Mike Dowling)
Subject: glibc-2.1.3 problems
Date: 20 Jul 2000 10:58:31 GMT
I configured using
# ../glibc-2.1.3/configure --prefix=/var/tmp/lib --disable-nls
--enable-add-ons=linuxthreads,crypt
(on one line). Then both the commands
# make
and
# make PARALLELMFLAGS=-j3
compiles glibc correctly (using make-3.79.1 and gcc-2.95.2). (I have
two CPUs.) However, I do get zillions of errors messages like
sysdeps/unix/make-syscalls.sh: invalid character 45 in exportstr for
full-config-sysdirs
(The INSTALL file does say that things that look like errors are not
errors, so perhaps that's one.)
However, make check then fails with
gcc -nostdlib -nostartfiles -o /usr/src/build/posix/globtest
-Wl,-dynamic-linker=/var/tmp/lib/lib/ld-linux.so.2
/usr/src/build/csu/crt1.o /usr/src/build/csu/crti.o `gcc
--print-file-name=crtbegin.o` /usr/src/build/posix/globtest.o
-Wl,-rpath-link=/usr/src/build:/usr/src/build/math:/usr/src/build/elf:/usr/src/build/nss:/usr/src/build/nis:/usr/src/build/db2:/usr/src/build/rt:/usr/src/build/resolv:/usr/src/build/linuxthreads
/usr/src/build/libc.so.6 /usr/src/build/libc_nonshared.a -lgcc `gcc
--print-file-name=crtend.o` /usr/src/build/csu/crtn.o
/bin/sh: invalid character 45 in exportstr for full-config-sysdirs
/bin/sh -e globtest.sh /usr/src/build/ /usr/src/build/elf/ \
ld-linux.so.2
globtest.sh: invalid character 45 in exportstr for full-config-sysdirs
- /tmp/globtest-out differ: char 2, line 1
make[2]: *** [do-globtest] Error 1
make[2]: Leaving directory `/usr/src/glibc-2.1.3/posix'
make[1]: *** [posix/tests] Error 2
make[1]: Leaving directory `/usr/src/glibc-2.1.3'
make: *** [check] Error 2
I've therefore installed the libraries in /var/tmp/lib, and set
LD_LIBRARY_PATH to /var/tmp/lib/lib. So far, I have not run into
problems. Can I expect any?
Actually, that is still not the full story. I initially tried to
include ncurses as one of the add-ons, but then "make" failed. I could
compile ncurses separately, and that succeeded. However, even with the
--with-shared parameter to configure, no shared libraries were built.
Cheers,
Mike
--
My email address [EMAIL PROTECTED] above is a valid email address.
It is, in fact, a sendmail alias; the digit 'N' is incremented regularly.
Spammed aliases will be deleted. Currently, mike[29,30]
are valid. If email to mikeN bounces, try mikeN+1.
------------------------------
From: Andreas Jaeger <[EMAIL PROTECTED]>
Subject: Re: glibc-2.1.3 problems
Date: 20 Jul 2000 13:15:36 +0200
>>>>> Mike Dowling writes:
> globtest.sh: invalid character 45 in exportstr for full-config-sysdirs
I've answered your email already as part of your bug report.
Just s/full-config-sysdirs/full_config_sysdirs/
or don't use bash 2.04.
[...]
> Actually, that is still not the full story. I initially tried to
> include ncurses as one of the add-ons, but then "make" failed. I could
> compile ncurses separately, and that succeeded. However, even with the
> --with-shared parameter to configure, no shared libraries were built.
Report this to the people responsible for the ncurses add-on.
Andreas
--
Andreas Jaeger
SuSE Labs [EMAIL PROTECTED]
private [EMAIL PROTECTED]
------------------------------
From: Uwe Bonnes <[EMAIL PROTECTED]>
Subject: Re: Caldera Wabi does not work in 2.2.16
Date: 20 Jul 2000 11:57:38 GMT
Nobody <[EMAIL PROTECTED]> wrote:
: Caldera Wabi 2.2 gives a "exception D in WIN_INST" when I try to use
: it under kernel 2.2.16.
My Wabi is broken by 2.2.16 too.
We should make noise on [EMAIL PROTECTED]
Bye
--
Uwe Bonnes [EMAIL PROTECTED]
Institut fuer Kernphysik Schlossgartenstrasse 9 64289 Darmstadt
========= Tel. 06151 162516 ======== Fax. 06151 164321 ==========
------------------------------
** 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
******************************
