Mailinglist Archive: radeonhd (223 mails)

[David Morrison <dave@xxxxxxx>]

I have some small success to report.  I had been using

cd $SRC/drm
./autogen.sh --prefix=/usr
make
make install
cd linux-core
make DRM_MODULES="radeon"
make install
depmod -a

to install the drm module.  On my Ubuntu 8.04 x86_64 system that ends up 
putting drm.ko and radeon.ko in

/lib/modules/2.6.24-17-generic/extra

Those are never seen and I kept loading old modules from elsewhere. 
When I actually followed the directions in drm/README and copied the 
*.ko files to

/lib/modules/2.6.24-17-generic/kernel/drivers/char/drm

Then they were properly loaded and I get this in Xorg.0.log (code still 
has my extra instrumentation):

drmOpenByBusid: Searching for BusID pci:0000:01:05.0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: major device number is 226
drmOpenDevice: remove (/dev/dri/card0) result is -1, (No such file or 
directory)
drmOpenDevice: mknod (/dev/dri/card0) result is 0, (OK)
drmOpenDevice: chown (/dev/dri/card0) result is 0, (OK)
drmOpenDevice: chmod (/dev/dri/card0) result is 0, (OK)
drmOpenDevice: stat (/dev/dri/card0) result is 0, (OK)
drmOpenDevice: open result is 7, (OK)
drmOpenByBusid: drmOpenMinor returns 7
drmOpenByBusid: drmGetBusid reports pci:0000:01:05.0
(II) RADEONHD(0): Found libdrm 1.3.0.
(II) RADEONHD(0): Found radeon drm 1.29.0.

I see that drm/linux-core/Makefile is invoking a Makefile from the 
installed kernel itself to build and install the modules.  At least on 
Ubuntu there seems to be some skew between the assumptions in 
drm/linux-core and the Makefile in the kernel tree.

Regards,
Dave
--
To unsubscribe, e-mail: radeonhd+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: radeonhd+help@xxxxxxxxxxxx