[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

No X after upgrade from lenny to testing with radeon x700



Did a clean install of lenny i386 onto a laptop with a Radeon X700
(RV4xx chipset). Normally I'd use fglrx but I decided to give the xorg
radeon driver a try to see if it had improved to a usable state yet.

Unfortunately, xrandr still can't handle 2 monitors with this chipset.
It reports that two outputs are connected and even accepts commands to
switch to a spanned desktop without error - it just doesn't obey them
and stays in clone mode.

I decided to upgrade to testing (avoided unstable after hearing a few
cautionary stories on the list) and now X crashes on startup with:

drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is -1, (No such device or address)
drmOpenDevice: open result is -1, (No such device or address)
drmOpenDevice: Open failed

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c91ce]
1: [0xb7f6c400]
2: /usr/lib/xorg/modules/drivers//radeon_drv.so(RADEONDRIGetVersion
+0x114) [0xb7b1abb4]
3: /usr/lib/xorg/modules/drivers//radeon_drv.so [0xb7ae7e53]
4: /usr/lib/xorg/modules/drivers//radeon_drv.so(RADEONPreInit+0xf2f)
[0xb7aeaa9f]
5: /usr/bin/X(InitOutput+0xa0f) [0x80ab3ff]
6: /usr/bin/X(main+0x2b1) [0x8074591]
7: /lib/i686/cmov/libc.so.6(__libc_start_main+0xe5) [0xb7cfb775]
8: /usr/bin/X [0x8073a81]

Fatal server error:
Caught signal 11.  Server aborting

Before it gets to this stage, the log suggests that the kernel module
and DRI extension are loaded OK. I've looked at there is no
device /dev/dri/card0 but I've got the impression that this is something
X should create dynamically. Am I wrong in this assumption - do I
somehow need to create it myself? Switching to fglrx causes this device
to appear, but the fglrx package in testing has a version difference
between the driver and the kernel module meaning it's not an ideal
solution.

Any suggestions on how to solve this missing /dev/dri/card0 with the
radeon driver?


Reply to: