Screenlocker grabs input. Not KWin related.

Weren't all those blogs about how the screenlocker was now part of kwin since that was the only secure way to do it? Or did i read too much in between the lines?

It's in ksmserver - ide is that when you kill the screen grabber, you exit the session.

This are a little different for wayland (where I assume KWin will *have to* redirect the screensaver input as well) but even there the logics would happen inside the screenlocker process (which I doubt to be KWin then)

(In reply to Albert Astals Cid from comment #2)
> Weren't all those blogs about how the screenlocker was now part of kwin
> since that was the only secure way to do it? Or did i read too much in
> between the lines?

see http://blog.martin-graesslin.com/blog/2011/11/work-not-done/

On the actual bug: it's tricky, I don't think the greeter process or ksld knows that the screen got blanked. Without that information one cannot do much. Personally I would say this is the wrong layer, such tasks should be done in X or Wayland to suppress the first input event.

Wouldn't it be sufficient (atm, on X11) to simply omit forwarding the unlocking event?

screenlocker/lockwindow.cpp:341
// XSendEvent(QX11Info::display(), targetWindow, False, NoEventMask, &ev2);
// TODO: send FocusIn instead ;-)

> screenlocker/lockwindow.cpp:341
> // XSendEvent(QX11Info::display(), targetWindow, False, NoEventMask, &ev2);
> // TODO: send FocusIn instead ;-)

is that a kde4 path, because line 341 doesn't make sense on my checkout (it's 
a break).

Yupp, is now

0329                         if (responseType == XCB_KEY_PRESS || responseType == XCB_KEY_RELEASE) {
0330                             sendEvent<xcb_key_press_event_t>(event, window, targetX, targetY);
0331                         } else if (responseType == XCB_BUTTON_PRESS || responseType == XCB_BUTTON_RELEASE) {
0332                             sendEvent<xcb_button_press_event_t>(event, window, targetX, targetY);
0333                         } else if (responseType == XCB_MOTION_NOTIFY) {
0334                             sendEvent<xcb_motion_notify_event_t>(event, window, targetX, targetY);
0335                         }

Sorry.

yeah that would work, but how do we know that the screen is blanked?

http://xcb.freedesktop.org/manual/group__XCB__DPMS__API.html

More readable API for brief oversight:
http://www.x.org/releases/X11R7.7/doc/libXext/dpmslib.html


Of course that does not hold for some in-screen magic, but I assume that's not the target here.

Another issue *could* be that we recive the event too late - we'd have to poll the dpms level to compare the event and the latest suspend time.

I had a look at the extension yesterday. Unfortunately it does not provide any events to notify when DPMS switches states. This means the state needs to be polled which I wouldn't want to do after each key event as it introduces a roundtrip to the X server. As well it's probably rather pointless as the X server handles input events before sending to the clients. Whenever we get an input event DPMS will be off. The screen locker has no idea whether the input event triggered a disabling of DPMS.

This shows that there is exactly one place where this information is available and this place can filter out events: the X server. On further thought: this is not a screen locker specific problem. Also if you have DPMS enabled and the lock screen disabled, you probably want the first event to be discarded. That's something only the X Server could do -> RESOLVED UPSTREAM.

Git commit 57b11f84296b69944b907a89d403dc50d36a75dc by Martin Gräßlin.
Committed on 15/02/2016 at 14:53.
Pushed by graesslin into branch 'master'.

[backends/drm] Use an InputEventFilter to reenable outputs

So far the DrmOutput connected to all input events when going into
power saving. As we now have the input filters it's better to just
install a filter when an output goes into powersave and remove the
input filter again when all outputs are enabled again.

To make this work InputRedirection gains a new method to add a new
filter as the first filter. This is a potentially dangerous method
as it allows to have a filter before LockScreenFilter gets the
events. But in case of DPMS it's something we actually want.

A nice new feature possible with the input filter is that we can
filter out the event which re-enables the outputs. Thus when getting
on a system with output off and screen locked, the first key hit
doesn't go to the lock screen.

Reviewed-By: Bhushan Shah
Fixed-in: 5.6.0 (Wayland-only)

M  +76   -14   backends/drm/drm_backend.cpp
M  +25   -1    backends/drm/drm_backend.h
M  +5    -0    input.cpp
M  +9    -1    input.h

http://commits.kde.org/kwin/57b11f84296b69944b907a89d403dc50d36a75dc