[RFC] Screensaver/blanking inhibition

[Pekka Paalanen]

On Thu, 19 Nov 2015 12:06:43 -0800
Bryce Harrington <bryce at osg.samsung.com> wrote:

> On Thu, Nov 19, 2015 at 05:38:34AM -0500, Christopher Michael wrote:
> > On 11/19/2015 05:28 AM, Christopher Michael wrote:
> > >On 11/19/2015 05:15 AM, Jonas Ådahl wrote:

> > >>Not so sure about the scope though. If its not about surfaces on outputs
> > >>or input devices or focus or display protocol things it should just be a
> > >>D-Bus protocol.
> > >>
> > >
> > >Unsure if dbus would be sufficent. Not saying it isn't .. just saying
> > >"unsure". The reason I say unsure is (essentially) a simple request to
> > >"inhibit the screen" or "start the screensaver" would (in theory) I
> > >assume need to talk to the compositor, potentially for permission checks
> > >(ie: is this client Allowed to do this ?, is the compositor doing
> > >something else that second which is slightly more important and thus
> > >this request needs to be delayed ?, etc, etc).
> > >
> > >For various reasons (and perhaps unforseen ones), I would "think" that
> > >it would chat with the compositor (in some form). IF that is the case, I
> > >don't believe dbus would be sufficient as the compositor may need to
> > >talk to the client (give me your credentials, who is the user requesting
> > >this, etc, etc).
> > >
> > >But... these are just my random, rambling thoughts on the subject ;)
> > >
> > >Cheers,
> > >Chris
> > >
> > 
> > After a little more thought, it could potentially work ok with dbus
> > ...assuming the requests end up being some form of dbus
> > method/signal combo where a compositor could listen and client could
> > talk (or vice-verse). Having said that tho, I (personally) would
> > rather see dbus avoided here and just have it be "wl/weston
> > protocol" (just seems more pure to me)...
> 
> Jonas can correct me if I'm interpreting his comment wrong, but I
> understood him to mean, "For inhibition of things that aren't directly
> relevant to displays or surfaces or input devices, leave those out of
> this proposal; they can be done using dbus or whatnot."  I didn't take
> his suggestion to mean that the screensaver inhibition itself would be
> done via DBUS.
> 
> Of course, there's no reason there couldn't be a higher level wrapper
> available through DBUS that handles the Wayland calls, if someone
> decided it'd be convenient or whatever.

A fundamental problem with using DBUS for things related to Wayland
protocol objects like wl_surfaces is that we do not have a way to
identify the wl_surface in a DBUS message. Every Wayland connection has
a private object id space, which means that you would have to identify
the connection if you used protocol object ids. Essentially you'd have
to invent a way to get global names for things.


Thanks,
pq
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 811 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/wayland-devel/attachments/20151120/814cb930/attachment.sig>