[off topics] NVidia, Laptop and Mandrake 9.0

MB · Post #21by MB » 08.04.2003, 11:59

Christophe,
Here is what I have:
(II) LoadModule: "glx"
(II) Loading /usr/X11R6/lib/modules/extensions/libglx.so
(II) Module glx: vendor="NVIDIA Corporation"
compiled for 4.0.2, module version = 1.0.4349
Module class: XFree86 Server Extension
ABI class: XFree86 Server Extension, version 0.1
(II) Loading extension GLX
(II) LoadModule: "nvidia"
(II) Loading /usr/X11R6/lib/modules/drivers/nvidia_drv.o
(II) Module nvidia: vendor="NVIDIA Corporation"
compiled for 4.0.2, module version = 1.0.4349
Module class: XFree86 Video Driver

It looks very similar to yours. Then what I get with xpdyinfo:

ine michel]$ xdpyinfo
name of display: :0.0
version number: 11.0
vendor string: Mandrake Linux (XFree86 4.2.1, patch level 3mdk)
vendor release number: 40201000
XFree86 version: 4.2.1
maximum request size: 4194300 bytes
motion buffer size: 256
bitmap unit, bit order, padding: 32, LSBFirst, 32
image byte order: LSBFirst
number of supported pixmap formats: 7
supported pixmap formats:
depth 1, bits_per_pixel 1, scanline_pad 32
depth 4, bits_per_pixel 8, scanline_pad 32
depth 8, bits_per_pixel 8, scanline_pad 32
depth 15, bits_per_pixel 16, scanline_pad 32
depth 16, bits_per_pixel 16, scanline_pad 32
depth 24, bits_per_pixel 32, scanline_pad 32
depth 32, bits_per_pixel 32, scanline_pad 32
keycode range: minimum 8, maximum 255
focus: window 0x2400009, revert to Parent
number of extensions: 31
BIG-REQUESTS
DOUBLE-BUFFER
DPMS
Extended-Visual-Information
FontCache
GLX
LBX
MIT-SCREEN-SAVER
MIT-SHM
MIT-SUNDRY-NONSTANDARD
NV-CONTROL
NV-GLX
NV-GLX
NV-GLX
NVIDIA-GLX
RENDER
SECURITY
SHAPE
SYNC
TOG-CUP
XC-APPGROUP
XC-MISC
XFree86-Bigfont
XFree86-DGA
XFree86-Misc
XFree86-VidModeExtension
XInputExtension
XKEYBOARD
XTEST
XVideo
XVideo-MotionCompensation
default screen number: 0
number of screens: 1

screen #0:
dimensions: 1024x768 pixels (292x222 millimeters)
resolution: 89x88 dots per inch
depths (7): 15, 1, 4, 8, 16, 24, 32
root window id: 0x56
depth of root window: 15 planes
number of colormaps: minimum 1, maximum 1
default colormap: 0x20
default number of colormap cells: 32
preallocated pixels: black 0, white 32767
options: backing-store NO, save-unders NO
largest cursor: 64x64
current input event mask: 0xda4031
KeyPressMask EnterWindowMask LeaveWindowMask KeymapStateMask StructureNotifyMask SubstructureNotifyMask SubstructureRedirectMask PropertyChangeMask ColormapChangeMask number of visuals: 2
default visual id: 0x21
visual:
visual id: 0x21
class: TrueColor
depth: 15 planes
available colormap entries: 32 per subfield
red, green, blue masks: 0x7c00, 0x3e0, 0x1f
significant bits in color specification: 8 bits
visual:
visual id: 0x22
class: DirectColor
depth: 15 planes
available colormap entries: 32 per subfield
red, green, blue masks: 0x7c00, 0x3e0, 0x1f
significant bits in color specification: 8 bits
[michel@line michel]$

MB

jamarsa · Post #22by **jamarsa** » 08.04.2003, 12:25

Ha!! got the same problem as you... I wondered why your xdpyinfo showed less resources than mine, and then modified XF86Config-4 to go down on DefaultColorDepth, from 24 to 8. And voila!! The same problem, kde window with stuck background, and SIGSEV when I try to close the window. It seems you have 15? bits per pixel. Try to get 24 or 32, if you can.

/etc/XF86Config-4:

Code: Select all

Section "Screen"
    Identifier "screen1"
    Device "device1"
    Monitor "monitor1"
    DefaultColorDepth 24

    Subsection "Display"
        Depth 8
        Modes "1024x768" "800x600" "640x480"
    EndSubsection

    Subsection "Display"
        Depth 15
        Modes "1024x768" "800x600" "640x480"
    EndSubsection

    Subsection "Display"
        Depth 16
        Modes "1024x768" "800x600" "640x480"
    EndSubsection

    Subsection "Display"
        Depth 24

MB · Post #23by MB » 08.04.2003, 13:55

Bravo!! jamarsa you were right. I just change the depth to 24 (I did not yet try 32)
A big step again. celestia start and shows something.
Nice cooperative help with you and Christophe.Again thanks
I also had to change the permissions in /dev/nvidiactl. But that is well documented in the Nvidia doc.
Still one problem I do not see the textures!
MB

MB · Post #24by MB » 08.04.2003, 14:20

Sorry! I did not realised that there was no texture in the rpm! it ok now, celestia works with KDE, very good.
Now I have to understand my celestia compilation problemes

MB.

jamarsa · Post #25by **jamarsa** » 08.04.2003, 14:24

I'm not very experienced in the textures field, but posting your OpenGL info in the 'Help' menu of celestia would help a bit... Also it could be a matter of paths, since i had to correct several capitalization issues in the directories of addons (I don't know about the RPM you installed). Check the directories of celestia (mine are in /usr/local/share/celestia).

jamarsa · Post #26by **jamarsa** » 08.04.2003, 14:25

Oops!! Just a bit too late ...

MB · Post #27by MB » 08.04.2003, 15:27

Just for INFO;
After all this stuff: Horrible end !!!!!:
Impossible to restart Linux.
I had to start with the safe??? boot and correct the /etc/security/console.perms and remove the
<xconsole> 0600 <dri> line.
But contrarily to what is claimed the Nvidia readme I did not get any warning before.

Other minor Pb:
I. I have to log as su to start celestia now.
As user, nothing start and No warning at the console.

II. As expected same error messages (given in the other thread at the compile ("make") of celestia.
One Problem after the other.
Bye
MB

jamarsa · Post #28by **jamarsa** » 08.04.2003, 15:29

After thinking a while about the compiling problem...

What compilers do you have installed?

# g++ --version
# gcc --version

I wonder if configure fails if you don't install g++, since I always install it!! :lol:

jamarsa · Post #29by **jamarsa** » 08.04.2003, 15:53

About the booting problem:

Do you mean you were unable to boot Linux itself

, or the graphics console? :wink:

Did you change the /dev/nvidia* permissions?

# chmod 0666 /dev/nvidia*
# chown root /dev/nvidia*

This happened to me when I was testing the 8-bit mode...

MB · Post #30by MB » 08.04.2003, 16:07

First question:
I should confess that I did not really choose the complier: here is the (assumed) relevant line of the configure procedure:

checking for gcc... gcc
checking for C compiler default output... a.out
checking whether the C compiler works... yes
checking whether we are cross compiling... no
checking for suffix of executables...
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for style of include used by make... GNU
checking dependency style of gcc... gcc3
checking for g++... g++
checking whether we are using the GNU C++ compiler... yes
checking whether g++ accepts -g... yes
checking dependency style of g++... gcc3
checking how to run the C preprocessor... gcc -E
checking how to run the C++ preprocessor... g++ -E
checking for a BSD-compatible install... /usr/bin/install -c -p
checking for ld used by GCC... /usr/bin/ld

Question II:
YES it was linux itself. Usually I am rather cautious and I remove the automatic startx.
Actually it is not yet re-established.
and YES I changed the permissions:
chmod 0666 /dev/nvidia*
chown root /dev/nvidia*

MB

MB · Post #31by MB » 08.04.2003, 16:15

Impossibe to start celestia as user:
I forgot, Just to show you what happens when try it, then as su:

[michel@line michel]$ celestia
[michel@line michel]$ su
Password:
[root@line michel]# celestia
_KDE_IceTransmkdir: Owner of /tmp/.ICE-unix should be set to root
DCOPServer up and running.
nStars: 112521
Parent body 'TYC 5503-946-1' of 'b' not found.
Initializing vertex programs . . .
Loading vertex program: shaders/diffuse.vp
Loading vertex program: shaders/specular.vp
Loading vertex program: shaders/haze.vp
Loading vertex program: shaders/bumpdiffuse.vp
Loading vertex program: shaders/bumphaze.vp
Loading vertex program: shaders/shadowtex.vp
Loading vertex program: shaders/diffuse_texoff.vp
Loading vertex program: shaders/rings.vp
Loading vertex program: shaders/ringshadow.vp
Loading vertex program: shaders/night.vp

then when I exit celestia I get at the console:

[root@line michel]# Mutex destroy failure: Device or resource busy
kdeinit: Fatal IO error: client killed
kdeinit: sending SIGHUP to children.
kdeinit: sending SIGTERM to children.
kdeinit: Exit.

jamarsa · Post #32by **jamarsa** » 08.04.2003, 18:51

I get these messages too (and a lot of other complaints), but only running KDE celestia from GNOME (as non-root user). If I try from KDE, there are no 'Fatal IO' errors.

both users are able to run celestia.

are you able to run another OpenGL program as non-root?
check also the permissions in the celestia data folder.

jamarsa · Post #33by **jamarsa** » 08.04.2003, 23:07

Another test I made:

I replaced the previous versions of gl.h, glx.h, glxtokens.h in /usr/X11R6/include/GL with the ones found in /usr/share/doc/NVIDIA_GLX-1.0, and .... Crash!! boom!!
exactly your compile output. So, you have to recover the previous ones from XFree86-devel-4.2.1-3mdk (Thanks Christophe for the 'rpm -qf' trick , I was doing it in a much uglier way before)

Code: Select all

Nvidia includes:

-r--r--r--    1  root   root 148924 dic  9 21:26 gl.h
-r--r--r--    1  root   root     9827 dic  9 21:26 glx.h
-r--r--r--    1  root   root   10978 dic  9 21:26 glxtokens.h


Good'n'old includes:

-r--r--r--    1  root   root   77761 sep 13 2002 gl.h
-r--r--r--    1  root   root     6817 sep 13 2002 glx.h
-r--r--r--    1  root   root     8369 sep 13 2002 glxtokens.h

Guest · Post #34by **Guest** » 13.04.2003, 09:02

It is again me after a while. Actually I am away and I just got a modem connexion.
As you said, Jamarsa, I have the Nvidia gl*.h includes, I renamed them, (may be useful in case of)
Them I tried to reinstall the "old" ones from the devel rpm package. You also said in your last message
that you use the trick of Christophe (rpm -qf) but that does not install it is just a question. Then I was somewhat lost in the man rpm. I tried to re-install the XFree86-devel-4.2.1-3mdk. But uninstall it means to uninstall a lot of others files (packages?) due to dependences and I don't dare, being afraid to destroy my system!. After all, what I need is just the three .h includes.Is there any simple procedure to get these three files from my Mandrake CD without any harm. I again appreciate any help from you or Christophe.
MB

jamarsa · Post #35by **jamarsa** » 23.04.2003, 10:18

Bad, too late ... (I was on vacation visiting Biarritz, sorry ) :wink:

Celestia Forums

[off topics] NVidia, Laptop and Mandrake 9.0

another approach to the compiling nightmare ...