← Back to team overview

hybrid-graphics-linux team mailing list archive

Re: ACPI experts anyone? Re: NVIDIA Optimus (Dell L702X)

 

On Monday 06 June 2011 12:25 AM, Albert Vilella wrote:
Awesome! Is this working on L502X and L501X models?

For my L502X I use these commands, and it has been working perfectly since the last 2 months.

Turn nVidia card on:  \_SB.PCI0.PEG0.PEGP._ON
Turn nVidia card off: \_SB.PCI0.PEG0.PEGP._OFF

I'm using Bumblebee, with the enablecard/disablecard scripts enabled. Its working perfectly, the card is turned off by calling \_SB.PCI0.PEG0.PEGP._OFF at bootup, and when I run optirun, \_SB.PCI0.PEG0.PEGP._ON is passed, the program starts up, and after the program is terminated, the card is switched off again.

Thanks to all who were involved in this effort, because of which my Ubuntu is now able to use the nVidia card as in Windows, with proper power saving as well :)

Bilal Akhtar


On Sun, Jun 5, 2011 at 7:29 PM, James<jgeboski@xxxxxxxxx>  wrote:
For my L702X with an NVIDIA 550M, I used the ACPI methods from Jos
Hickson's dellL702X.sh script.

Methods:
    Turn on NVIDIA: \_SB.PCI0.PEG0.PEGP._PS0
    Turn off NVIDIA: \_SB.PCI0.PEG0.PEGP._PS3

The dellL702X.sh script:
https://lists.launchpad.net/hybrid-graphics-linux/msg00663.html

James

On Sun, Jun 5, 2011 at 11:44 AM, Albert Vilella<avilella@xxxxxxxxx>  wrote:
As far as I know, we haven't been able to decode the on/off switch for
the modern XPS models, and it's a pity because we have dozens of users
in the launchpad team, for the L501X, L502X and L702X models. The
DSDTs are here:

https://bugs.launchpad.net/lpbugreporter/+bug/752542

If there is any ACPI expert in the list, please have a look at those,
see if we can find the on/off graphics switch.

On Sun, Jun 5, 2011 at 7:51 AM, James<jgeboski@xxxxxxxxx>  wrote:
Hello all,

I'm recently got a Dell L702X and so far it's been a rather nice
machine aside from the Optimus hurdle.  So far, it works just fine
with the Intel chip, actually compiz capable but, I've got a 550M next
to it.  So far, I've managed to use the acpi_call module
(http://github.com/mkottman/acpi_call) to turn on and off the NVIDIA
chip and I can confirm this by the amount of power being drawn from
the battery.  My question is, with the NVIDIA chip enabled, is there
anyway to have the NVIDIA proprietary driver take over?  I've tried
this with and without the Intel VESA module but, with the same result,
a black screen.  The monitor does not turn off into a standby state,
it simply just display no content.  My guess would be that everything
must go through the Intel chip before it can hit the NVIDIA and that
the NVIDIA chip has to send everything back to the Intel chip due to
the monitor being connected to the Intel chip.  Ideally I'd like to
just have the NVIDIA chip always on with the NVIDIA driver handling
things but, I suppose if it was this easy, there would be no need for
the bumblebee project.  Is something like this even currently doable?

Thanks,
James

_______________________________________________
Mailing list: https://launchpad.net/~hybrid-graphics-linux
Post to     : hybrid-graphics-linux@xxxxxxxxxxxxxxxxxxx
Unsubscribe : https://launchpad.net/~hybrid-graphics-linux
More help   : https://help.launchpad.net/ListHelp




_______________________________________________
Mailing list: https://launchpad.net/~hybrid-graphics-linux
Post to     : hybrid-graphics-linux@xxxxxxxxxxxxxxxxxxx
Unsubscribe : https://launchpad.net/~hybrid-graphics-linux
More help   : https://help.launchpad.net/ListHelp


_______________________________________________
Mailing list: https://launchpad.net/~hybrid-graphics-linux
Post to     : hybrid-graphics-linux@xxxxxxxxxxxxxxxxxxx
Unsubscribe : https://launchpad.net/~hybrid-graphics-linux
More help   : https://help.launchpad.net/ListHelp



--
Bilal Akhtar - Ubuntu Developer <bilalakhtar@xxxxxxxxxx>
IRC Nick: cdbs


Follow ups

References