Re: sk98lin doesn't work
Hi, Bernd
Thanks for the scripts. They worked perfectly well with the sk98lin
install.sh script. And finally I got the sk98lin.ko module.
After 'modprobe sk98lin', I can see the NIC with 'ifconfig -a'.
But when I do 'ifconfig eth1 up', the system hang up.
I did a reboot by pressing the reset button.
Even worse, I got a 'kernel panic' caused by bringing up this NIC.
I am contacting the NIC driver developer.
One more question, I know debian has several ports,
Intel x86 / IA-32 (i386),
IA-64 (ia64),
AMD64, etc.
Which port should I use?
I have two 64bits CPUs (Intel(R) Xeon(TM) CPU 3.00GHz).
the kernel I am using is 2.6.8-9-em64t-p4-smp.
I chosed Intel x86/IA-32 without thinking.
Now I doubt whether my choice was right.
Thanks,
Xiaolin
On Fri, Dec 24, 2004 at 01:09:30AM +0100, Bernd Schubert wrote:
Hello Xiaolin,
Thanks for your gcc trick. I tried just now. It worked, but I still got
errors when compiling:
+++
arch/x86_64/ia32/vsyscall-sigreturn.S: Assembler messages:
arch/x86_64/ia32/vsyscall-sigreturn.S:16: Error: suffix or operands
invalid for `pop'
make[1]: *** [arch/x86_64/ia32/vsyscall-sysenter.o] error 1
make: *** [arch/x86_64/ia32] error 2
+++
More ideas?
Oh dear, I'm afraid you will have to patch every tool that is also required
for compiling. The messages above are probably comming from the assembler.
Here's what my college did for that reason on our cluster:
/usr/bin/as:
#!/bin/sh
#
# This is an as wrapper. It forces as --64 for default assemblage.
#
# Based on the gcc wrapper, /usr/bin/gcc .
# Frank Otto, 2004-10-13
arch=`uname -m`
AS=/usr/bin/as.binutils
args=${1+$@}
case ${arch} in
x86_64)
if ( echo $args | grep -e '--32' /dev/null ) ; then
exec ${AS} ${args}
elif ( echo $args | grep -e '--64' /dev/null ) ; then
exec ${AS} ${args}
else
exec ${AS} --64 ${args}
fi
;;
*)
exec ${AS} ${args}
;;
esac
This required you rename /usr/bin/as to /usr/bin/as.binutils, but of course,
you could also easily do it as above with gcc, so putting it into the PATH
And then here, a wrapper for ld:
#!/bin/sh
# wrapper script for ld:
# look in the 64-bit library directories too, pretty please
#
# by TC-Admin, 11/2004
exec /usr/bin/ld $@ -L/usr/local/lib64 -L/usr/lib64 -L/lib64
This we put to /usr/local/bin, which is in our path befor /usr/bin. I know
we are a bit inconsistent.
The next 17 days I won't read this list, so if you have further
problems/questions to this topic, please mail me directly.
Cheers,
Bernd
PS: As the above as replacement script contains the name of my college, and
as
he wrote this script and I don't want to remove his name. I also don't know
if he likes it if I publish his name on a newsgroup, so send this directly to
you.
--
Bernd Schubert
Physikalisch Chemisches Institut / Theoretische Chemie
Universit?t Heidelberg
INF 229
69120 Heidelberg
e-mail: [EMAIL PROTECTED]
--
**
Re: mkinitrd linking issue.
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Chris Wakefield wrote: | Hi Harri. | | here is the output to: | | bash -x /usr/sbin/mkinitrd -o /tmp/initrd.img-2.6.10-rc3 2.6.10-rc3 log | [snip] | + xldd /sbin/modprobe | + i=/sbin/modprobe | + set +e | + '[' yes ']' | ++ LD_LIBRARY_PATH= | ++ LD_ASSUME_KERNEL=2.4 | ++ ldd /sbin/modprobe | /bin/bash: error while loading shared libraries: libdl.so.2: cannot open | shared object file: No such file or directory | + x= Would you mind to run /usr/bin/ldd /sbin/modprobe on the command line and send me the output? I get this on my PC: % /usr/bin/ldd /sbin/modprobe ~libc.so.6 = /lib/libc.so.6 (0x002a9566d000) ~/lib/ld-linux-x86-64.so.2 = /lib/ld-linux-x86-64.so.2 (0x002a95556000) Regards Harri -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.5 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFBy8t6UTlbRTxpHjcRAra4AJ4uXBo1c4yEgvtCe46YH9RzJxSoUACfXBxL oAIsrFxKNi2zuI3OP3amjcw= =Eqqs -END PGP SIGNATURE-
Re: mkinitrd linking issue.
Hi Harri. Here it is: $ /usr/bin/ldd /sbin/modprobe libc.so.6 = /lib/libc.so.6 (0x002a9566c000) /lib64/ld-linux-x86-64.so.2 = /lib64/ld-linux-x86-64.so.2 (0x002a95556000) Thanks for your interest and help, Chris W. On December 23, 2004 11:55 pm, Harald Dunkel wrote: Chris Wakefield wrote: | Hi Harri. | | here is the output to: | | bash -x /usr/sbin/mkinitrd -o /tmp/initrd.img-2.6.10-rc3 2.6.10-rc3 | log [snip] | + xldd /sbin/modprobe | + i=/sbin/modprobe | + set +e | + '[' yes ']' | ++ LD_LIBRARY_PATH= | ++ LD_ASSUME_KERNEL=2.4 | ++ ldd /sbin/modprobe | /bin/bash: error while loading shared libraries: libdl.so.2: cannot open | shared object file: No such file or directory | + x= Would you mind to run /usr/bin/ldd /sbin/modprobe on the command line and send me the output? I get this on my PC: % /usr/bin/ldd /sbin/modprobe ~libc.so.6 = /lib/libc.so.6 (0x002a9566d000) ~/lib/ld-linux-x86-64.so.2 = /lib/ld-linux-x86-64.so.2 (0x002a95556000) Regards Harri
Re: Debian netboot on a SunFire v20z
On Friday 24 December 2004 01:09, Michael Madden wrote: Has anyone gotten a netboot install for amd64 working on a Sun SunFire V20z? I'm using the netboot from 12/21/2004. It boots from the cd, detects my Broadcom card, loads the tg3 module, configures the network, obtains an IP address, netmask, gateway, and DNS servers from my DHCP server. It then asks me to select a mirror (I have tried ftp.us.debian.org, mirrors.kernel.org, and numerous other foreign mirrors), and it gets the following error: Read the FAQ. This has a list of mirrors. https://alioth.debian.org/docman/view.php/30192/21/debian-amd64-howto.html Paul
Re: libGL compiled without -fPIC?
On Thu, Dec 23, 2004 at 08:20:40PM -0600, Mark Nipper wrote: While trying to build vlc here at home, I got this: --- gcc -Wsign-compare -Wall -pipe -o libopengl_plugin.so libopengl_plugin_a-opengl.o `top_builddir=../.. ../../vlc-config --libs plugin opengl` /usr/bin/ld: /usr/lib/libGL.a(glapi.o): relocation R_X86_64_32 can not be used when making a shared object; recompile with -fPIC /usr/lib/libGL.a: could not read symbols: Bad value collect2: ld returned 1 exit status make[5]: *** [libopengl_plugin.so] Error 1 This is how it should fail if you're trying to link to a static library. Static libraries should be build _without_ -fPIC, and shared with -fPIC. In the special case you need a static lib to link a shared lib, you have to have a static PIC lib named like that. But there is a shared version of the lib available as /usr/lib/libGL.so (in xlibmesa-gl-dev), why aren't you using that? I should mention I'm trying to debuild the vlc package, not straight from source from upstream. The maintainer left off --enable-ncurses (bug filed already) and I successfully rebuilt the package at work on x86. I was just trying to build the same here at home on x86_64. :) Unfortuantly, x86 does not enforce those -fPIC for shared libraries. amd64 seem to be the one that enforces it most strictly, followed by hppa. And from which sources are you trying to build it? And you installed all the build dependencies? (apt-get build-dep vlc) I did added --enable-ncurses to the configure flags and did a complete build which was succesful. Kurt
Re: Debian netboot on a SunFire v20z
Hi, I have sucessfully installed debian-amd64 on the same hardware but I also was not able to download any packags while the debian-installer was running. Therefor I downloaded the latest netinstall which dows not really needs to download any packages. After installing this way I was able to a deb mirrors on the running system. regards bjoern Michael Madden schrieb: Has anyone gotten a netboot install for amd64 working on a Sun SunFire V20z? I'm using the netboot from 12/21/2004. It boots from the cd, detects my Broadcom card, loads the tg3 module, configures the network, obtains an IP address, netmask, gateway, and DNS servers from my DHCP server. It then asks me to select a mirror (I have tried ftp.us.debian.org, mirrors.kernel.org, and numerous other foreign mirrors), and it gets the following error: The installer failed to download a file from the mirror. This may be a problem with your network, or with the mirror. You can choose to retry the download, select a different mirror, or cancel and choose another installation method. I know the network card and the network are fine. I recently installed Solaris 10 b72 and Ubuntu on this server and connected to the network without a hitch. I've tried retrying, changing numerous ftp/http mirrors, and I always get the same error. Any suggestions? Thanks in advance, Mike
Re: Xorg
On Tue, 21 Dec 2004, lore wrote: A little survey: has anyone tried Xorg instead of X11? Yes. I gave it a shot about a month ago. My PCI Matrox Millennium II was useless on it as it is on the current XFree86 too. I think the mga driver is buggy. I get garbage on the millennium screen. Besides the MGA driver problem I also had all kind of keyboard weirdness, couldn't get the meta keys to work as they do on the XFree86 so I just switched back to XFree. Besides these problems it did compile and run flawlessly. -- Antti Pyykko --- [EMAIL PROTECTED] --- +358 50 539 5438 --. .--- OAMK --- http://www.pyksy.fi/ --- pyksy @IrcNet --' `- The OTEki Information Minister - Linux/Amiga/c64 freak Microsoft Corp is prohibited from redistributing this work.
ASUS K8N-E DELUXE
Hello debian-adm64 I am trying to install amd64 on my ASUS K8N-E DELUXE: K8 2800+, 1GB-RAM, 120GB-IDE I have tried to use Sarge amd64 from 14 dec, and everything seems to work o.k.(it didn't find my netcard, so I used a 8139-pci-card) When it tries to reboot the very first time, GRUB hangs. Has anyone successfully installed Debuían-AMD64 on a ASUS K8N-E DELUXE ?? best regards [EMAIL PROTECTED]
Re: udev problems
I have been experiencing a similar problem.. I cannot install udev without losing x as described. I just did netinstall with sid from alioth and x is fine until udev is installed (ie. with gnome-desktop-environment). Then I get the problems described by Kunjan. I did aptitude purge udev, which took out gnome-desktop-environment with it and now my x display is working again. The only thing i could note from the install was that I chose to have my devices managed by discover when that option was presented. Season's Greetings, Norv On Thu, 2004-12-23 at 02:27 -0600, Kunjan Shah wrote: shoot i been replying to you only all the time and this thread never made it to the list. so now it does. :( On Thu, 23 Dec 2004 02:25:53 -0600, Kunjan Shah [EMAIL PROTECTED] wrote: ok hmm i booted with a live cd and did cd /mnt/hda8/dev/ mknod -m 600 console c 5 1 mknod -m 666 null c 1 3 and booted with devfs=nomount and still my mtab looks the same:( and i have udevd running and /proc/filesystems shows nodev devfs yup my brain's fried. On Wed, 22 Dec 2004 22:33:05 -0600, Kunjan Shah [EMAIL PROTECTED] wrote: the thing is i think udev and devfs are running together... coz i got a /.dev in root... yeah my /home is on a seperate partition. i wouldnt mind reinstalling its a fresh system and i have no data on it. AMD64 u know i knew it wasnt going to be smooth sailing ;) here's my fstab (or u can go here http://kunjan.net/fstab ) # /etc/fstab: static file system information. # # file system mount point type options dump pass proc/proc procdefaults0 0 /dev/hda8 / ext3defaults,errors=remount-ro 0 1 /dev/hda7 /boot ext3defaults0 2 /dev/hda9 /home ext3defaults0 2 /dev/hda10 noneswapsw 0 0 /dev/hdc/media/cdrom0 iso9660 ro,user,noauto 0 0 and here's grep output CONFIG_DEVFS_FS=y # CONFIG_DEVFS_MOUNT is not set # CONFIG_DEVFS_DEBUG is not set i can reinstall if its required but if i can fix it i can learn a few new things ;) this is a fresh install so its cool!! Thanks a lot! Regards, Kunjan On Wed, 22 Dec 2004 12:14:52 -0600, Ron Johnson [EMAIL PROTECTED] wrote: On Wed, 2004-12-22 at 11:45 -0600, Kunjan Shah wrote: here's my mtab /dev/ide/host0/bus0/target/lun0/part7 /boot ext3 rw 0 0 /dev/ide/host0/bus0/target/lun0/part8 / ext3 rw 0 0 /dev/ide/host0/bus0/target/lun0/part9 /home ext3 rw 0 0 devpts /dev/pts devpts rw,gid=5,mode=620 0 0 none /dev tmpfs rw,size=5M,mode=0755 0 0 proc /proc proc rw 0 0 sysfs /sys sysfs rw 0 0 tmpfs /dev/shm tmpfs rw 0 0 usbfs /proc/bus/usb usbfs rw 0 0 Well, you're definitely running devfs... :( /dev/ide/host0/bus0/target/lun0/partX this i need to change to /dev/hdaX right? *You* won't. Linux will, when you disable devfs. How the heck devfs got there in the 1st place is what I want to know. what else is wrong there? Hmmm. It's not wrong, its just different. (I sound s Politically Correct, don't I?) I need you to do these 4 things: - Please *attach* your /etc/fstab. - Please tell me that your /home is on a separate partition. - Run this command grep DEVFS /boot/config-$(uname -r) - Steel yourself for the probability that you'll have to reinstall Thanks a lot Regards Kunjan On Wed, 22 Dec 2004 11:08:12 -0600, Ron Johnson [EMAIL PROTECTED] wrote: On Wed, 2004-12-22 at 10:57 -0600, Kunjan Shah wrote: Well if i add devfs=nomount then the system doesnt boot. the Kernel says cant open /dev/console. and /dev/null. I tried creating static entries but since they are already in /dev i cant create them either so if i want to boot the system i have to add devfs=mount to the kernel args. Interesting. What does cat /etc/mtab|sort look like? funny thing is i already removed devfsd package arrrgh. Indeed. -- - Ron Johnson, Jr. Jefferson, LA USA PGP Key ID 8834C06B I prefer encrypted mail. I wish the USA could get out of the UN. But a forum where governments can talk is too useful. The next best thing is to only pay a fraction of our dues. Or find a better forum. -- - Kunjan Shah http://kunjan.net -- - Kunjan Shah http://kunjan.net -- - Kunjan Shah http://kunjan.net
Re: udev problems
On Tue, 28 Dec 2004 02:54:57 +1100, Norval Watson [EMAIL PROTECTED] wrote: I have been experiencing a similar problem.. I cannot install udev without losing x as described. I just did netinstall with sid from alioth and x is fine until udev is installed (ie. with gnome-desktop-environment). Then I get the problems described by Kunjan. I did aptitude purge udev, which took out gnome-desktop-environment with it and now my x display is working again. The only thing i could note from the install was that I chose to have my devices managed by discover when that option was presented. Season's Greetings, Norv Merry Christmas to all! Do you have a nividia display card? if you do then you might have installed a custom kernel. and for some reason even if u dont have the devfsd installed it will still load up devfs what i did was reinstall everything and i didnt select any pacakge from base-config instead i bailed out and installed udev first and then fired up aptitude and installed other stuff. how ever i am having some problems getting nvidia drivers running again i dont need 3d accel but the nv driver currupts the text mode display when you get out of x. and gnome-volume-manager needs udev... so thats the only reason i need udev. Regards, Kunjan. On Thu, 2004-12-23 at 02:27 -0600, Kunjan Shah wrote: shoot i been replying to you only all the time and this thread never made it to the list. so now it does. :( On Thu, 23 Dec 2004 02:25:53 -0600, Kunjan Shah [EMAIL PROTECTED] wrote: ok hmm i booted with a live cd and did cd /mnt/hda8/dev/ mknod -m 600 console c 5 1 mknod -m 666 null c 1 3 and booted with devfs=nomount and still my mtab looks the same:( and i have udevd running and /proc/filesystems shows nodev devfs yup my brain's fried. On Wed, 22 Dec 2004 22:33:05 -0600, Kunjan Shah [EMAIL PROTECTED] wrote: the thing is i think udev and devfs are running together... coz i got a /.dev in root... yeah my /home is on a seperate partition. i wouldnt mind reinstalling its a fresh system and i have no data on it. AMD64 u know i knew it wasnt going to be smooth sailing ;) here's my fstab (or u can go here http://kunjan.net/fstab ) # /etc/fstab: static file system information. # # file system mount point type options dump pass proc/proc procdefaults0 0 /dev/hda8 / ext3defaults,errors=remount-ro 0 1 /dev/hda7 /boot ext3defaults0 2 /dev/hda9 /home ext3defaults0 2 /dev/hda10 noneswapsw 0 0 /dev/hdc/media/cdrom0 iso9660 ro,user,noauto 0 0 and here's grep output CONFIG_DEVFS_FS=y # CONFIG_DEVFS_MOUNT is not set # CONFIG_DEVFS_DEBUG is not set i can reinstall if its required but if i can fix it i can learn a few new things ;) this is a fresh install so its cool!! Thanks a lot! Regards, Kunjan On Wed, 22 Dec 2004 12:14:52 -0600, Ron Johnson [EMAIL PROTECTED] wrote: On Wed, 2004-12-22 at 11:45 -0600, Kunjan Shah wrote: here's my mtab /dev/ide/host0/bus0/target/lun0/part7 /boot ext3 rw 0 0 /dev/ide/host0/bus0/target/lun0/part8 / ext3 rw 0 0 /dev/ide/host0/bus0/target/lun0/part9 /home ext3 rw 0 0 devpts /dev/pts devpts rw,gid=5,mode=620 0 0 none /dev tmpfs rw,size=5M,mode=0755 0 0 proc /proc proc rw 0 0 sysfs /sys sysfs rw 0 0 tmpfs /dev/shm tmpfs rw 0 0 usbfs /proc/bus/usb usbfs rw 0 0 Well, you're definitely running devfs... :( /dev/ide/host0/bus0/target/lun0/partX this i need to change to /dev/hdaX right? *You* won't. Linux will, when you disable devfs. How the heck devfs got there in the 1st place is what I want to know. what else is wrong there? Hmmm. It's not wrong, its just different. (I sound s Politically Correct, don't I?) I need you to do these 4 things: - Please *attach* your /etc/fstab. - Please tell me that your /home is on a separate partition. - Run this command grep DEVFS /boot/config-$(uname -r) - Steel yourself for the probability that you'll have to reinstall Thanks a lot Regards Kunjan On Wed, 22 Dec 2004 11:08:12 -0600, Ron Johnson [EMAIL PROTECTED] wrote: On Wed, 2004-12-22 at 10:57 -0600, Kunjan Shah wrote: Well if i add devfs=nomount then the system doesnt boot. the Kernel says cant open /dev/console. and /dev/null. I tried creating static entries but since they are already in /dev i cant create them either so if i want to boot the system i
Re: libGL compiled without -fPIC?
On 24 Dec 2004, Kurt Roeckx wrote: But there is a shared version of the lib available as /usr/lib/libGL.so (in xlibmesa-gl-dev), why aren't you using that? Yeah, I'd done apt-get build-dep already. And believe me, it wasn't by choice that I wanted it to use the static version over the shared version! :) Anyway, I found the problem. In /usr/lib I had: --- lrwxrwxrwx 1 root root 12 Dec 23 21:47 libGL.so - libGL.so.1.2 lrwxrwxrwx 1 root root 17 Dec 14 16:21 libGL.so.1 - libGL.so.1.0.6629 -rw-r--r-- 1 root root 586K Dec 7 21:14 libGL.so.1.0.6629 but you'll note libGL.so.1.2 is missing! I think this may be an nvidia-glx related bug. I receall having to manually create libGL.so.1.2 myself on my x86 systems some time ago for a similar GL library issue. After adding the link: --- lrwxrwxrwx 1 root root 10 Dec 24 19:18 libGL.so.1.2 - libGL.so.1 everything compiles and is happy. If you could confirm (assuming you don't have the binary nVidia drivers installed) the sanity of this situation pre nvidia-glx, I'll file a bug report as I recall the same thing on x86 too. Interestingly enough, if I do 'dpkg -S /usr/lib/libGL.so.1.2' it gives me: --- diversion by nvidia-glx from: /usr/lib/libGL.so.1.2 diversion by nvidia-glx to: /usr/lib/nvidia/libGL.so.1.2.xlibmesa xlibmesa-gl: /usr/lib/libGL.so.1.2 but it obviously never actually created it. Thanks for the reply. -- Mark Nippere-contacts: Computing and Information Services [EMAIL PROTECTED] Texas AM Universityhttp://ops.tamu.edu/nipsy/ College Station, TX 77843-3142 AIM/Yahoo: texasnipsy ICQ: 66971617 (979)575-3193 MSN: [EMAIL PROTECTED] -BEGIN GEEK CODE BLOCK- Version: 3.1 GG/IT d- s++:+ a- C++$ UBL$ P---+++ L+++$ !E--- W++(--) N+ o K++ w(---) O++ M V(--) PS+++(+) PE(--) Y+ PGP t+ 5 X R tv b+++@ DI+(++) D+ G e h r++ y+(**) --END GEEK CODE BLOCK-- ---begin random quote of the moment--- Whiskey-Tango-Foxtrot, over. -- anonymous end random quote of the moment
