ratpoison-devel
[Top][All Lists]
Advanced

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

Re: [RP] ratpoison-1.4.9-xrandr4 snapshots


From: Fabian \"cbdev\" Stumpf
Subject: Re: [RP] ratpoison-1.4.9-xrandr4 snapshots
Date: Fri, 9 Dec 2016 23:36:24 +0100

Hi Mathieu,

> Could you try to rebuild it with the configure below :
> 
> --8<---------------cut here---------------start------------->8---
> ./configure --with-xrandr --enable-debug
> --8<---------------cut here---------------end--------------->8---

OK, when configuring the build that way, it complains about
missing Xrandr.h - which is easily fixed by installing the appropriate
package. I think this was the primary problem.

And it does work now! I actually get two screens though the windows 
overlap in a weird way (the boundary between the two displays is overlapping).
This happens when reconfiguring the display mode while in ratpoison and
reconfiguring the external display first.

Upon first starting, the internal display is configured to 1024x768 
(why I don't really know, but fixable). I then run 
        xrandr --output VGA1 --mode 1280x1024 --right-of LVDS1
which configures the external screen correctly (and also updates 
ratpoison to give me another screen). I then run
        xrandr --output LVDS1 --mode 1440x900
which gives me the overlapping windows.

I can fix it by reconfiguring again, first the internal and then the external
display. The problem also does not manifest itself when doing things in this 
order
initially.

An sdump in the overlapping situation looks like:
| VGA1 0 1152 0 1280 1024 0,LVDS1 1 0 0 1440 900 1

While the output of xrandr looks like:
|Screen 0: minimum 320 x 200, current 2432 x 1024, maximum 8192 x 8192
|LVDS1 connected 1440x900+0+0 (normal left inverted right x axis y axis) 304mm 
x 190mm
|   1440x900      60.06*+  59.89    50.18  
|[etc]
|VGA1 connected 1280x1024+1152+0 (normal left inverted right x axis y axis) 
338mm x 270mm
|   1280x1024     60.02*+  75.02  
|[etc]

This problem persists across `restart` calls.

When doing things in the correct order (or fixing it up using the method
described above), the sdump looks like
| VGA1 0 1440 0 1280 1024 0,LVDS1 1 0 0 1440 900 1

While the output of xrandr looks like
|Screen 0: minimum 320 x 200, current 2720 x 1024, maximum 8192 x 8192
|LVDS1 connected 1440x900+0+0 (normal left inverted right x axis y axis) 304mm 
x 190mm
|   1440x900      60.06*+  59.89    50.18  
|[etc]
|VGA1 connected 1280x1024+1440+0 (normal left inverted right x axis y axis) 
338mm x 270mm
|   1280x1024     60.02*+  75.02  
|[etc]

So this might be a bug in XRandr.


> --8<---------------cut here---------------start------------->8---
> ratpoison > rp_debug
> grep "Xrandr version" rp_debug
> --8<---------------cut here---------------end--------------->8---

That comes up empty, though some relevant info might be

|ratpoison:xrandr.c:244: debug: --- Handling RRNotify ---
|ratpoison:xrandr.c:254: debug: ---          XRRCrtcChangeNotifyEvent ---
|ratpoison:xrandr.c:223: debug: xrandr_crtc_change: crtc found, rotation
|RR_Rotate_0 ev->x 1152, ev->y 0, ev->width 1280, ev->height 1024
|ratpoison:xrandr.c:244: debug: --- Handling RRNotify ---
|ratpoison:xrandr.c:254: debug: ---          XRRCrtcChangeNotifyEvent
|--- ratpoison:xrandr.c:223: debug: xrandr_crtc_change: crtc found,
|rotation RR_Rotate_0 ev->x 0, ev->y 0, ev->width 1440, ev->height 900
|ratpoison:xrandr.c:244: debug: --- Handling RRNotify ---
|ratpoison:xrandr.c:249: debug: ---          XRROutputChangeNotifyEvent
|--- ratpoison:xrandr.c:244: debug: --- Handling RRNotify ---
|ratpoison:xrandr.c:249: debug: ---          XRROutputChangeNotifyEvent
|--- ratpoison:bar.c:629: debug: msg = ratpoison 1.4.9-xrandr4
|ratpoison:xrandr.c:244: debug: --- Handling RRNotify ---
|ratpoison:xrandr.c:254: debug: ---          XRRCrtcChangeNotifyEvent
|--- ratpoison:xrandr.c:244: debug: --- Handling RRNotify ---
|ratpoison:xrandr.c:249: debug: ---          XRROutputChangeNotifyEvent
|--- ratpoison:xrandr.c:180: debug: xrandr_output_change: Removing
|screen VGA1


Another problem just presented itself when returning from a TMPWM (jwm in my 
case) 
to ratpoison:
The display is overlaid by a black window and ratpoison is not responding to any
configured keys but the meta key. I can still send commands to ratpoison by 
switching to another VT and sending commands via ratpoison -c.
ratpoison seems to enter command mode when I press the meta key (C-s in my 
setup)
and performs `other` (bound to C-s in my setup) when I press it again, but does 
not
react to any other configured keys. Starting the tmpwm again by running
        ratpoison -c "tmpwm jwm"
in another VT shows me an empty window with an empty title in front of all 
others.

xprop on the window shows

|_NET_WM_STATE(ATOM) = 
|_NET_WM_ALLOWED_ACTIONS(ATOM) = _NET_WM_ACTION_SHADE, _NET_WM_ACTION_MINIMIZE, 
|_NET_WM_ACTION_MAXIMIZE_HORZ, _NET_WM_ACTION_MAXIMIZE_VERT, 
_NET_WM_ACTION_CLOSE, 
|_NET_WM_ACTION_RESIZE, _NET_WM_ACTION_MOVE, _NET_WM_ACTION_CHANGE_DESKTOP, 
|_NET_WM_ACTION_STICK, _NET_WM_ACTION_BELOW, _NET_WM_ACTION_ABOVE
|_NET_FRAME_EXTENTS(CARDINAL) = 4, 4, 20, 4
|_NET_WM_DESKTOP(CARDINAL) = 0
|WM_STATE(WM_STATE):
|               window state: Normal
|               icon window: 0x0

This was not the case in earlier versions.

Thanks for all the work!

Kind regards
Fabian




reply via email to

[Prev in Thread] Current Thread [Next in Thread]