Very Computer

Hello!
I have a problem with getting to run 3D:
My configuration: MSI K7N2 / Connect3D Radeon 9000pro / MDK 9.1

The drivers from the ATI-Site don't work. Trying configuring "DRI"
doesn't - glxinfo gives me that direct redering is off.

The only clues I have is in the logs below.

I've heared somewhere that agpgart on nforce2 is not supported for
non-nVidia cards, and that DRI doesn't need agpgart anymore to
activate 3D-functions in radeon-cards except in R300-Chips.
Well, that's only what I heared around the net... can anyone
explain this matter more detailed or has any solution for this?

XFree86.conf-4:

Section "Module"
    (..snip..)
    Load "glx" # 3D layer
    Load "dri" # direct rendering
EndSection

Section "DRI"
    Mode 0666
EndSection

Section "Device"
    Identifier "device1"
    VendorName "ATI"
    BoardName "ATI Radeon"
    Driver "radeon"
    Screen 0
    BusID "PCI:2:0:0"
    Option "DPMS"
# Changed by Emre
    Option "AGPMode" "4"
    Option "AGPFastWrite" "on"
    Option "NoAccel" "off"
    Option "EnablePageFlip" "on"
# Emre End
EndSection

Changing "radeon" to "ati" due to
http://www.xfree86.org/4.3.0/DRI10.html#47 had the same result.

var/log/messages:
kernel: [drm:radeon_unlock] *ERROR* Process 25060 using kernel context 0

dmesg:
agpgart: Maximum main memory to use for agp memory: 439M
agpgart: unsupported bridge
agpgart: no supported devices found.
Linux agpgart interface v0.99 (c) Jeff Hartmann [drm:radeon_unlock]
*ERROR* Process 2113 using kernel context 0

/var/log/X11/XFree.0.log

(==) RADEON(0): Write-combining range (0xd4000000,0x4000000)
drmOpenDevice: minor is 0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is -1, (No such device)
drmOpenDevice: open result is -1, (No such device)
drmOpenDevice: Open failed
drmOpenDevice: minor is 0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is -1, (No such device)
drmOpenDevice: open result is -1, (No such device)
drmOpenDevice: Open failed
[drm] failed to load kernel module "agpgart"
drmOpenDevice: minor is 0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 7, (OK)
drmGetBusid returned ''
(II) RADEON(0): [drm] loaded kernel module for "radeon" driver
(II) RADEON(0): [drm] created "radeon" driver at busid "PCI:2:0:0"
(II) RADEON(0): [drm] added 8192 byte SAREA at 0xe482c000
(II) RADEON(0): [drm] mapped SAREA 0xe482c000 to 0x40014000
(II) RADEON(0): [drm] framebuffer handle = 0xd4000000
(II) RADEON(0): [drm] added 1 reserved context for kernel
(WW) RADEON(0): [agp] AGP not available

thanx in advance-
Emre Kutbay

Hello,

Quote:> I have a problem with getting to run 3D:
> My configuration: MSI K7N2 / Connect3D Radeon 9000pro / MDK 9.1

My configuration is EPOX 8RDA / Connect3D Radeon 9100 / Debian GNU/Linux

Quote:> The drivers from the ATI-Site don't work. Trying configuring "DRI"
> doesn't - glxinfo gives me that direct redering is off.

Yep, nforces' agpgart isn't supported cause a leak of docs. Say thanks to
nVidia or something. (correct me, if I'm wrong..)

Quote:> I've heared somewhere that agpgart on nforce2 is not supported for
> non-nVidia cards, and that DRI doesn't need agpgart anymore to
> activate 3D-functions in radeon-cards except in R300-Chips.
> Well, that's only what I heared around the net... can anyone
> explain this matter more detailed or has any solution for this?

Hmm, I would like to know more about this.. Do you have any URLs or
something about that?

--
- tpr
Teemu Rytilahti
tpr (at) d5k . net

Hello Teemu,

[...dri without agp...]

Quote:>Hmm, I would like to know more about this.. Do you have any URLs or
>something about that?

That's part of the Information I got as I read 100s of discussions here
and there party in german.
Unfortunately that's the only URL pointing to this I bookmarked:
http://www.xfree86.org/4.3.0/radeon.4.html

I made 3D work as I found an option there:

Option "ForcePCIMode" "on"

Then I deleted all the option that looked like they belong to agp,
so this is the Device-Section:

Section "Device"
    Identifier "device1"
    VendorName "ATI"
    BoardName "ATI Radeon"
    Driver "radeon"
    Screen 0
    BusID "PCI:2:0:0"
    Option "DPMS"
#   Option "AGPMode" "4"
#   Option "AGPFastWrite" "on"
    Option "ForcePCIMode" "on"
    Option "NoAccel" "off"
    Option "EnablePageFlip" "on"
EndSection

glxgears gives me the same fps, but armagetron now works more fluid and
tuxracer too. glxinfo also gives me that direct rendering is on. But:

After 3-5 minutes my monitor goes into standby-mode, my keyboard is
blocked (ctrl-alt-del does not work), so the only way to get out is the
reset-button.
Have not found out what happened there, /var/log/messages gives me this:

kernel: Linux agpgart interface v0.99 (c) Jeff Hartmann
kernel: agpgart:Maximum main memory to use for agp memory: 439M
kernel: [drm] Loading R200 Microcode
xinetd[2222]: warning: can't get client address: Transport endpoint
is not connected

My assumption: Maybe agpgart is a module, and there is a way not to load
it at all... I haven't checked it yet... I hope I don't have to recompile
the kernel, because I don't know, which patches are on the kernel made by
mandrake.

Please tell me, if you experience the same and have found a soltuion-

Emre

Hello,

Quote:> Option "ForcePCIMode" "on"

Turned this on, like you did.

Quote:> After 3-5 minutes my monitor goes into standby-mode, my keyboard is
> blocked (ctrl-alt-del does not work), so the only way to get out is the
> reset-button.

My monitor goes into standby immediately after startx/startup and the only
way out is reset. So I'm having the same problem, but immediately after
startx.

Quote:> Have not found out what happened there, /var/log/messages gives me this:
> kernel: Linux agpgart interface v0.99 (c) Jeff Hartmann
> kernel: agpgart:Maximum main memory to use for agp memory: 439M
> kernel: [drm] Loading R200 Microcode
> xinetd[2222]: warning: can't get client address: Transport endpoint
> is not connected

I have nothing strange in /var/log/messages but I 'tail -n 50
/var/log/kdm.log' gives me this:
(EE) RADEON(0): GetBuffer timed out, resetting engine...
(EE) RADEON(0): RADEONCPGetBuffer: CP reset -1020
(EE) RADEON(0): RADEONCPGetBuffer: CP start -1020
(EE) RADEON(0): RADEONCPGetBuffer: CP GetBuffer -1020
(EE) RADEON(0): GetBuffer timed out, resetting engine...
(EE) RADEON(0): RADEONCPGetBuffer: CP reset -1020
[<--cut--> A lot of those.. <--cut-->] and then in the end:
X: warning; /dev/dri has unusual mode (not 755) or is not a directory.

Quote:> My assumption: Maybe agpgart is a module, and there is a way not to load
> it at all... I haven't checked it yet... I hope I don't have to recompile
> the kernel, because I don't know, which patches are on the kernel made by
> mandrake.

Hmm, if you mean agpgart as kernel module, I can't load it:
tuli:/var/log# insmod agpgart
Using /lib/modules/2.4.20/kernel/drivers/char/agp/agpgart.o
/lib/modules/2.4.20/kernel/drivers/char/agp/agpgart.o: init_module: No such
device
Hint: insmod errors can be caused by incorrect module parameters, including
invalid IO or IRQ parameters.

Quote:> Please tell me, if you experience the same and have found a soltuion-

Hope someone (we) will find solution for this. Inform me about your
experiences and so. Thanks.

--
- tpr
Teemu Rytilahti
tpr (at) d5k . net

Hi Teemu,

Quote:> Hope someone (we) will find solution for this. Inform me about your
> experiences and so. Thanks.

I've asked a few people in a discussion-forum

http://www.nforcershq.com/forum/viewtopic.php?t=10611&sid=a8dee89769f46
7ea6b1784444afa64a3

I haven't tried the refresh-rate thing, but I don't believe that that's
the problem because the keyboard gets  blocked too. Even tried to find a
solution on irc with no result.

This seems to be interesting:
http://www.nvnews.net/vbulletin/showthread.php?s=7f962637096a1f078f65df
62c233e953&threadid=9396

I don't have any time for this anymore, so I'll wait nvidia to support a
nvagp module. At least there is hope ;)

cu-
Emre