Bug 180007

Created attachment 124168 [details]
log file

Created attachment 124169 [details]
xorg.conf file, modified for ati driver

This card is already listed in radeon.xinf as:

alias pcivideo:v00001002d00004A50sv*sd*bc*sc*i* radeon

This is the second card that has been reported to not autodetect, which has
turned out to already have an entry.  I suspect something might be wrong with
the parsing of the videoalias files.

CC'ing Bill/Jeremy for comment.

The bug has been set to NEEDINFO_REPORTER but i'm not sure what info is required
of me? :-)

Please attach the output of 'kudzu -p -c video'.

Umm that might be a problem, kudzu doesn't currently seem to like my system:

[root@localhost ~]# kudzu -c -p video
Segmentation fault
[root@localhost ~]# kudzu -c -p video --safe
(no output)

It's 'kudzu -p -c video'; you have the args reversed. Still, it shouldn't
segfault - can you attach a gdb backtrace of the segfault (after installing
kudzu-debuginfo)?

When i run kudzu thru gdb i get:

Program received signal SIGSEGV, Segmentation fault.
0x0016abd3 in strdup () from /lib/libc.so.6
(gdb) bt
#0  0x0016abd3 in strdup () from /lib/libc.so.6
#1  0x0805d28e in vbe_get_vbe_info () at vbe.c:165
#2  0x0805a35e in ddcProbe (probeClass=Variable "probeClass" is not available.
) at ddc.c:395
#3  0x08050366 in probeDevices (probeClass=CLASS_UNSPEC, probeBus=-9,
    probeFlags=1) at kudzu.c:806
#4  0x0804d054 in main (argc=Cannot access memory at address 0xffffffff
) at hwconf.c:938
#5  0x00116784 in __libc_start_main () from /lib/libc.so.6
#6  0x0804a491 in _start ()

Just to make sure i wasn't wasting anyone's time i yum upgrade'd to todays
recent tree and tried to reconfig using system-config-display again, and got the
same result still:
Section "Device"                                                               
                                             
        Identifier  "Videocard0"                                               
                                             
        Driver      "vesa"                                                     
                                             
        VendorName  "Videocard vendor"                                         
                                             
        BoardName   "ATI Technologies Inc R420 [X800XT-PE] (Secondary)"        
                                             
EndSection

As it was noted on the mailing list, it seems the (Secondary) thing confuses it,
it has a different pci_id, and there's no driver for it .. and it looks like its
trying to configure it, and not the main part, ie tries to configure 01:00.1 and
not 01:00.0 from lspci:
01:00.0 VGA compatible controller: ATI Technologies Inc R420 JP [Radeon X800XT]
01:00.1 Display controller: ATI Technologies Inc R420 [X800XT-PE] (Secondary)

Ugh, that segfault implies garbage in the ddc return - not much that can be done
there.

(In reply to comment #4)
> The bug has been set to NEEDINFO_REPORTER but i'm not sure what info is required
> of me? :-)

Sorry, lately I've been doing that by accident instead of NEEDINFO_ENG. ;)


(In reply to comment #9)
> Ugh, that segfault implies garbage in the ddc return - not much that can be
> done there.

Sanity checking to avoid SEGV? ;o)

Hard to sanity check the pointer without dereferencing it.

Changing component to 'kudzu'

Closing as CANTFIX - garbage in, garbage out. :/

As for configuring the second device, that's handled in bug 176978.