[ohm] Some thoughts on OHM

[Koen Kooi]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Richard Purdie schreef:
> On Wed, 2007-02-28 at 17:21 +0000, Richard Hughes wrote:
>>> 1. The screen can rotate so the device can be configured in a portrait
>>> mode (no keyboard) or a landscape mode (keyboard beneath the screen).
>>> There are switches that detect the position.
>> Surely abstract that out in HAL? That's got to be useful for a desktop
>> application. display.rotation or something like that might work.
> 
> It is partly abstracted in HAL already. The input SW_LID event is sent
> and I think HAL understands that. HAL doesn't know about SW_TABLET_MODE
> though.
> 
> Also, you need something, somewhere to tell the system that a keyboard
> isn't available in "tablet mode" but available in the landscape mode.
> The kernel can't really provide that, HAL doesn't, does OHM? Or does
> this have to be coded into then policy OHM uses? See my comments below
> about information.

What should happen in the following cases?

1) plug in usb keyboard, swivel to tablet mode
2) swivel to tablet mode, plugin in usb keyboard

I think tablet mode should mark a specific keyboard as being unavailable, but I', not sure
how to do that.

> At the back of my mind, I'm thinking of how a build system (say
> OpenEmbedded) might build HAL/OHM for multiple different style devices.
> I don't really want to have to teach OE a new configure line for HAL for
> every device added to it.
> 
> Information about different machines needs to go somewhere but where?

In the worst case we can use the MACHINE_FEATURES variable in OE to disable PCI and friends.

regards,

Koen
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFF5qVVMkyGM64RGpERAp8nAJ9C6/a/T8F/qB+j1gphoATUDeNvJQCgie/m
8Nk+G0QfPZxm0BBMccrFF0c=
=W9to
-----END PGP SIGNATURE-----