Mailinglist Archive: radeonhd (408 mails)

[Matthias Hopf <mhopf@xxxxxxx>]

On Jun 05, 09 07:36:02 -0400, Scott Edick wrote:
> Ah, the problem appears to be an incompatibility with my KVM switch.  I've 

Great that you found that yourself. :-)

> used this switch without trouble for years on lots of different HW and SW.  
> This same HW with Debian Lenny works the same with or without the switch.  

You can explicitly enable the output if you add a monitor section to
your xorg.conf with [[ Option "Enable" ]] added, and referenced by
[[ Option "monitor-VGA_1" "identifier" ]] (using the correct identifier)
in the Device section.

> == using VGA ===
> rhd_conntest: v1.2.5, dist of git branch master, commit 4be5f715
> Found card: (null) - (null)
> Checking connectors on 0x9610, 0x1458, 0xD000  (@01:05:00):
>   Load Detection: RHD_OUTPUT_DACA
>   HotPlug: RHD_HPD_NONE
>   DDC: RHD_DDC_0
>
> == using VGA via KVM ===
> rhd_conntest: v1.2.5, dist of git branch master, commit 4be5f715
> Found card: (null) - (null)
> Checking connectors on 0x9610, 0x1458, 0xD000  (@01:05:00):
>   Load Detection: RHD_OUTPUT_NONE
>   HotPlug: RHD_HPD_NONE
>   DDC: RHD_DDC_0

Ok, the load detection doesn't work. I assume that the KVM switch
doesn't have correct 75 Ohm termination, and the load detection is
picky...

If you want you can play a bit with the value 0x1e6 for register
DACA_FORCE_DATA in rhd_dac.c.

Another idea: Please run the Xserver with -logverbose 7, with the
monitor attached and with the KVM attached. Let's see what DACSense()
found, maybe not all of R, G, and B are terminated correctly...

Matthias

-- 
Matthias Hopf <mhopf@xxxxxxx>      __        __   __
Maxfeldstr. 5 / 90409 Nuernberg   (_   | |  (_   |__          mat@xxxxxxxxx
Phone +49-911-74053-715           __)  |_|  __)  |__  R & D   www.mshopf.de
-- 
To unsubscribe, e-mail: radeonhd+unsubscribe@xxxxxxxxxxxx
For additional commands, e-mail: radeonhd+help@xxxxxxxxxxxx