[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Sleep on desktop system?



><snip>
>
>> This is typically the case with all video cards. Almost each 
>> modern video chip requires a complicated initialization sequence
>> involving setting up of the chip's internal clock net, turning
>> on the various PLLs, configuring the SGRAM controller, etc...
>> along with blasting a couple of non documented registers controlling
>> internal chip workarounds ...
>
>Is this the case on laptops (read: 3rd gen tibook) too?

No. On laptops the chip is just unclocked afaik (though the video
ram may be powered off as well).

We use the chip's own power management capabilities to put it in
D2 state (at least with recent chips, older mach64 used an
equivalent mecanism that wasn't driven via the PCI PM).

The problem with the M7 (and partially with the M6 too though
we finally got that one sleeping not too badly) is that for those
modern chips to enter sleep state properly, a whole lot of
"preparation" work must be done to put various parts of the chip
in a state where they can be suspended. Similar work is needed
to bring things back to life. An none of this is documented.

>Assuming yes on the former question, would this work on the TiBook too?

Well... if we had a way to hard reset the chip, that may work as
I would then throw the chip to D3 mode instead of D2 on suspend and
hard reset it & fully restore it's state on resume, but afaik, we
don't have control over the reset line. (Well, we probably do via
a non documented uninorth register, but I don't know what it is, and
it seems ATI doesn't either).

Ben.




Reply to: