[MPlayer-users] Re: mplayer > 1.x doesn'trecognize myradeon cardwith vidix

[justin case]

> it is opening this libdha library but this is what
> actually performs the pci scan used by the vidix drivers.
> library not even loaded normally with my default vo option.
>
> > do you have any more ideas?
>
> yes from your strace log
>
> > open("/usr/local/lib/mplayer/vidix/radeon_vid.o", O_RDONLY) = 8
> > read(8,
> "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\1\0\3\0\1\0\0\0\0\0\0\0"..., 1024)
> >\
>
> Look above this is not a shared object. has extension of .o should be .so
> (compare to the other vidix drivers in log)
> This file is possibly just the output of compiling a single .c
> file or object.
>
> you can use the file command to verify this
> file /usr/local/lib/mplayer/vidix/radeon_vid.so
> output should show shared object
> ELF 32-bit LSB shared object, Intel 80386, version 1 (SYSV), not stripped
>
> I would first make sure to copy the file vidix/drivers/radeon_vid.so
> to /usr/local/lib/mplayer/vidix
>
> and while you are at it might as well delete every other file  in
> /usr/local/lib/mplayer/vidix
> it is not required to load them and probe for non existant cards
> when you know you have a radeon 7500.


the radeon_vid.o was indeed created by a make in the vidix driver directory
and then i copied everything over.
deleting everything in the directory /usr/local/lib/mplayer/vidix/ and then
doing a make install in my 1.X mplayer source directory copied the correct
radeon_vid.so into mplayer /usr/local/lib/mplayer/vidix but still it refuses
to find the chip.
make install in my 0.9 sources dierectory leaves me with a working vidix
mplayer again.
./mplayer -vo vesa:vidix in my 1.X sources starts mplayer, even detects the
chip, since it is using the 0.9 radeon_vid.so but i get a black screen,
although the sound keeps playing.
finally using the -nocolorkey option made the picture appear.

so now i have a alsa capable mplayer with a (strangely working)
vidix-driver.

thx for all the support.

/jc