linux 2.6.0-pre11, XFree fails to start

2003-12-18 Thread Éric Brunet
Hello all,

I have an intermitent bug on my computer (intel PIV, shuttle motherboard,
integrated i845 video card): sometimes, at boot time, XFree (4.3.0,
from redhat 9 package) fails to start. The screen goes black (or, more
precisely, darkgrey) as the video mode is changed, and stays there
forever. Keyboard is not responding (not even caps-lock or sysrq), and I
have to reboot the computer. There is nothing in the log, the only
information are the last lines written by the XServer in
/var/log/XFree86.0.log:

(II) I810(0): xf86BindGARTMemory: bind key 5 at 0x067f8000 (pgoffset 26616)
(II) I810(0): xf86BindGARTMemory: bind key 6 at 0x066f8000 (pgoffset 26360)
(II) I810(0): xf86BindGARTMemory: bind key 7 at 0x05278000 (pgoffset 21112)
(II) I810(0): Before: SWF1 is 0x0108
(II) I810(0): After: SWF1 is 0x0108

On a normal boot, those lines are followed by:

(II) I810(0): Display plane A is enabled.
(II) I810(0): Display plane B is disabled.
(II) I810(0): PIPEACONF is 0x8000

There is nothing peculiar in the logs of the kernel.

As I said, the bug is intermitent. I should say it happens every three or
four boots. Before 2.6.0-pre11, I have been using 2.6.0-pre8 and -pre9,
and I don't remember this problem happening. Or if it did occur, it was
much less frequent. Maybe I should go back to 2.6.0-pre9 to be sure.

I really don't know how to debug a hard lock like this, or how to find
usefull information.

Every interesting information on my computer (config files, lspci, log of
XFree, etc.) is available on http://perso.nerim.net/~tudia/bug-reports/.

Éric Brunet

___
Devel mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/devel


Re: [Xpert]i810/i830 CVS Xserver 4.2 kdm logout hangs

2002-07-10 Thread Éric Brunet

On Tue, Jul 09, 2002 at 05:05:20PM +0900, Doug du Boulay wrote:
 I am using the CVS 4.2.x xfree86 xserver, last updated 
 about 1 month ago. Basically 2 out of 3 times when 
 I logout of kde/kdm my sony Vaio laptop (PCG-R505X/PD)
 hangs, completely. No Ctrl-alt-backspace, no ctrl-alt-F1, 
 no gentle reboot from short press of the power button.
 Basically the only escape is to hold the power button down for 
 5 seconds and the machine powers down, losing unflushed
 file buffers etc, trashing inodes and whathave you.
 All a bit inconvenient really. 
 
 basically  at logout from kde, the screen turns blank,
 sometimes the mouse cursor remains active for a couple
 of seconds on the blanked screen, before freezing in place
 as the whole machine locks up.

I just want to say that I have the same kind of problems with
a toshiba S3000-X11 (the graphic card is a i830MG) with XFree-4.2
(official binaries) on a redhat 4.2 with different custom kernels.
 
 I believe it is a total lockup because the machine 
 no longer responds to pings, telnet, ftp,ssh iether.

The difference is that it happens rarerly (once every 20 or 30 times,
maybe even less), and that I have never tried to check if the laptop was
responding or not by ethernet.

Éric Brunet
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert



Re: [Xpert]830mg config

2002-04-30 Thread Éric Brunet

On Mon, Apr 29, 2002 at 11:20:01AM +1000, Manuel M. T. Chakravarty wrote:
  For your information, I have a toshiba S3000-X11 with an i830MG board.
  Using the official binaries for XFree86 4.2 and the latest kernel found
  on rawhide, it works beautifuly. I just had to declare 32Mo of videoram
  in my XF86Config, and I have 1024x768 true colour, and the 3D (tuxracer)
  working.
 
 This works only if the BIOS is configured to preset 8MB of
 (stolen) video ram, but not all laptop vendors do that.

This is indeed the case. From my dmesg:
 Linux agpgart interface v0.99 (c) Jeff Hartmann
 agpgart: Maximum main memory to use for agp memory: 196M
 agpgart: Detected an Intel 830M Chipset.
 agpgart: detected 8192K stolen memory.
 agpgart: AGP aperture is 128M @ 0xe800
 [drm] AGP 0.99 on Intel i810 @ 0xe800 128MB
 [drm] Initialized i830 1.2.0 20011004 on minor 0

Regards,

Éric Brunet

___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert



[Xpert]Dual Head on i830MG

2002-03-20 Thread Éric Brunet

Hello all,

I have bought a toshiba Satellite 3000 X11 laptop with an intel i830MG
graphic board. With Windows XP, I can use this computer in a dual-head
configuration: the laptop screen and an external screen on the vga port
can display different pictures. I would like to do the same thing with
linux.

I am using the official binaries of XFree 4.2.0, with the linux kernel
2.4.17 patched by redhat available on rawhide. Looking at the log of the
X server, I found the lines

(--) PCI:*(0:2:0) Intel i830M rev 3, Mem @ 0xe800/27, 0xe000/19
(--) PCI: (0:2:1) Intel i830M rev 0, Mem @ 0xf000/27, 0xe008/19

Does this mean that the i830MG board is like two i830M boards ? To test
this, I write in my XF86Config file:

Section ServerLayout
Identifier Anaconda Configured
Screen  0  Screen0 0 0
Screen  1  Screen1 RightOf Screen0
InputDeviceMouse0 CorePointer
InputDeviceKeyboard0 CoreKeyboard
EndSection

[...]

Section Device
Identifier   Intel 830 MG
Driver   i810
VendorName   Intel
BoardName Intel 830
BusID   PCI:0:2:0
EndSection

[...]

Section Device
Identifier   Intel 830 MG 2
Driver   i810
VendorName   Intel
BoardName Intel 830
BusID   PCI:0:2:1
EndSection

(And, of course, Screen0 references Intel 830 MG and Screen1
references Intel 830 MG 2)

Here are some selected lines of the logs:

(**) I810(0): Depth 24, (--) framebuffer bpp 32
(==) I810(0): RGB weight 888
(==) I810(0): Default visual is TrueColor
(WW) I810(0): Bad V_BIOS checksum
(II) I810(0): Primary V_BIOS segment is: 0xc000
(--) I810(0): Chipset: i830
(--) I810(0): Linear framebuffer at 0xE800
(--) I810(0): IO registers at addr 0xE000
(II) I810(0): detected 8192K stolen memory.
(II) I810(0): I810CheckAvailableMemory: 200700k available
(==) I810(0): Will alloc AGP framebuffer: 8192 kByte
(==) I810(0): Using gamma correction (1.0, 1.0, 1.0)
(II) I810(0): Currently active displays on Pipe A:
(II) I810(0): LFP (Local Flat Panel) child device
[...]
(II) I810(1): VESA BIOS detected
(II) I810(1): VESA VBE Version 3.0
(II) I810(1): VESA VBE Total Mem: 8000 kB
(II) I810(1): VESA VBE OEM: Almador Graphics Chip Accelerated VGA BIOS
[...]
(II) I810(0): VESA VBE DDC supported
(II) I810(0): VESA VBE DDC Level 2
(II) I810(0): VESA VBE DDC transfer in appr. 1 sec.
(II) I810(0): VESA VBE DDC read successfully
[...]
(II) Reloading /usr/X11R6/lib/modules/linux/libint10.a
(**) I810(1): Depth 24, (--) framebuffer bpp 32
(==) I810(1): RGB weight 888
(==) I810(1): Default visual is TrueColor
(EE) I810(1): Cannot read V_BIOS
(EE) I810(1): Int10 initialization failed.

And it doesn't work: I only have one screen working. So, what should I do ?
What is this V_BIOS thing ? Am I correct when I assume that the two screens
correspond to the two devices, or has the device on 0:2:1 nothing to do
with a graphic card ? What is this Pipe A in the log (Currently active
displays on Pipe A:) ? Is there a Pipe B that I should somehow
activate ?

I am pretty much in the dark and the documentation about the i810 driver
seems a bit outdated.

I am not sure that all of you would love having my full config file and
full log, but, of course, I can send them inprivate to anyone interested.

Thank you for any help,

Éric Brunet
___
Xpert mailing list
[EMAIL PROTECTED]
http://XFree86.Org/mailman/listinfo/xpert