*** GMX Spamverdacht *** Re: [patch] samsung P35 suspend quirk

[Nikolaus Filus]

Danny Kukawka wrote:
> On Mittwoch, 19. März 2008, Nikolaus Filus wrote:
>> Danny Kukawka wrote:
>>> On Mittwoch, 12. März 2008, Nikolaus Filus wrote:
>>>> Hi,
>>>>
>>>> this patch fixes suspend/resume for Samsung P35. Tested directly
>>>> with pm-suspend and pm-suspend $QUIRKS and
>>>>
>>>> dbus-send --system --print-reply --dest="org.freedesktop.Hal"
>>>> /org/freedesktop/Hal/devices/computer
>>>> org.freedesktop.Hal.Device.SystemPowerManagement.Suspend int32:0
>>>>
>>>> System: Ubuntu Hardy (~ Alpha6)
>>>>
>>>> Please note also https://bugs.launchpad.net/bugs/198808
>>> Could you please send output of :
>>>
>>>   lshal -lu /org/freedesktop/Hal/devices/computer
>> Hi,
>>
>> thanks for the response and your attention to the report. I'm currently
>> using Ubuntu Hardy's daily updates so the pasted hal output may be
>> fluctuating. Hope that helps.
> 
> Could you please check if these quirks also work (from X a and tty-console):
> 
>   power_management.quirk.vbe_post = true
>   power_management.quirk.vbemode_restore = true


Hi,

so I tested a lot during the easter holidays and here are my results.

Resume DOESN'T work with any quirk with pure vga console and without radeonfb.
radeonfb is absolutely necessary. On some recent kernels radeonfb is broken and 
needs a patch as described in https://bugs.launchpad.net/bugs/201591

With radeonfb loaded:
	power_management.quirk.vbestate_restore = true
works always. The above proposed quirks
	power_management.quirk.vbe_post = true
	power_management.quirk.vbemode_restore = true
work AFTER Xorg was started at least once (using free radeon driver).
So my assumption is Xorg radeon driver does some initialisation that is missing 
in radeonfb. As this seems like a bug where do I have to report the issue?


Hope that helps.


Nikolaus