Re: How to make xinit / startx to read my custom xorg.conf.d files?

[Thomas Lübking]

This has nothing to do w/ xinit or lightdm.
The driver match isn't met because of a race condition that has the GPU
not ready before systemd kicks off the graphical.target.
https://wiki.archlinux.org/title/NVIDIA#DRM_kernel_mode_setting
https://wiki.archlinux.org/title/Kernel_mode_setting#Early_KMS_start

Am Fri, Jul 09, 2021 at 01:55:28AM + schrieb Me:

I am running Arch Linux with the i3 window manager.  Today I switched
from lightdm to xorg-xinit.  I have read the available documentation,
and configured things so that startx will start and run the X server,
and i3wm runs fine after that.  However, I have one very important
problem that I cannot find an answer to.

I have a Nvidia GPU with proprietary drivers installed, and I have
created a file named 10-nvidia-drm-outputclass.conf in the
/usr/share/X11/xorg.conf.d/ directory.  The contents of that conf file
are as follows:


Section "OutputClass"
   Identifier "nvidia"
   MatchDriver "nvidia-drm"
   Driver "nvidia"
   Option "AllowEmptyInitialConfiguration"
   Option "Coolbits" "28"
   ModulePath "/usr/lib/nvidia/xorg"
   ModulePath "/usr/lib/xorg/modules"
EndSection



An important line in that file is the Option "Coolbits" "28" line, as
it allows me to set a custom fan speed for my GPU, which I need for
these warmer months.


When I use startx that conf file is not being read, it seems.  It is
completely ignored, and I am unable to set a custom fan speed.

Is there a conf file for xinit that I can use, create, or configure
that will access the 10-nvidia-drm-outputclass.conf file, so that the
GPU will run cooler via a custom fan speed?

When using lightdm, it works like a charm.  But as I say, xinit seems
to not see/read/access that file.

Any help would be greatly appreciated!!

C.M.Millington
--
Thank you!
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Slow Window Drag Causes Transparent Window And Locks GUI

[Thomas Lübking]

Am Fri, 9 Feb 2018 14:19:07 -0600
schrieb Nathan Royce :

> [stuff]
> The short of it is this...

KWin skips a button release, maybe it's indeed not generated by the
server.
Check your xorg log on whether you use libinput or evdev and try the
other to see whether it's a problem with one driver exclusively.
Just to be sure: this isn't a touchpad, is it?
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: [xrdb] documentation error in man page

[Thomas Lübking]

On Wed, Sep 13, 2017 at 02:35:21PM +, Adesh Kumar wrote:


It means that we can specify properties


The term "specified" appears on almost every parameterless option and
refers to the set of properties on the server.
Nothing else in the manpage suggest parameter support for "remove" or
any other affected switch.

Iff (and only if) this is bad wording, it affects far more than just this one
switch, but that's something a native english speaker should judge.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Cannot open /dev/tty0 (No such file or directory)

[Thomas Lübking]

On Mon, Aug 14, 2017 at 12:33:49PM -0500, Mike Drennan wrote:

(EE)
Fatal server error:
(EE) parse_vt_settings: Cannot open /dev/tty0 (No such file or directory)
(EE)


https://bbs.archlinux.org/viewtopic.php?pid=1584440#p1584440

But I don't know how this relies to "crouton", your daughter or
minecraft ...

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Dell P2715Q Monitor does not wake up after sleep (off, standby, or suspend)

[Thomas Lübking]

On Sat, Aug 12, 2017 at 02:28:03PM -0400, Greg Gorsuch wrote:

Actually, I was looking at the output from the xrander -q I posted in my
first e-mail. It does show the display as connected still


or "again" - the only explanation I see for the DPMS (?) induced change.


I have tried udevadm monitor and do not see any events.


Not even when you actually unplug the device???


a rule  UBSYSTEM=="drm", ACTION=="change",
RUN+="/usr/local/bin/fix_dell_monitor.sh",
but it doesn't seem like that ever gets executed.


Hard to say. Typical pitfall is that xrandr needs to be run for the
current XAuthority and Display (iow. as you user and with some
environment set) - not as root. Try to "echo foo > /tmp/bar" to check
whether it's called.
Also do not forget to reload the rules.


Where is the logic for deactivating monitors and waking up from sleep
actually handled?


*Shrug*
The hardware triggers an interrupt, the kernel/module sends a udev event
and "some™" userspace process acts accordingly (or not)

I'd try with a pure X11 server first and foremost to see whether some
supersophisticated DE daemon makes a bad choice. (Ie. launch only the
server and an xterm)

The nvidia blob actually has an "UseHotplugEvents" device option, but
that won't work w/ DP (because hotplug events are mandatory for this
connection)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Setting new mode for 2nd monitor fails

[Thomas Lübking]

intel says the limit is 3840x2160@60Hz
https://ark.intel.com/products/83505/Intel-Core-i7-4770HQ-Processor-6M-Cache-up-to-3_40-GHz

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: remap Mode_switch+mouse_drag to Alt+mouse_drag

[Thomas Lübking]

On Thu, Jun 08, 2017 at 09:57:49AM +0800, Amos Bird wrote:

Hi Thomas,

Thanks for your help. However I cannot remap the Mode_switch key itself. It has 
other bindings.
I would like only the Mode_switch+left_mouse_drag remapped to 
Alt+left_mouse_drag. Is it possible?


You're likely rather looking for a way to configure your WM to handle the this 
shortcut instead of the other?


What *might* work is a daemon interpreting the event to XTestFake*Event
Alt+LMB, but if eg. the WM tracks the mouse button condition, it won't
be cheated by two presses w/o a release or still consider the mode
switch (is it or is it the meta/win key?) pressed and not act etcetc.
ie. you need to know the exact behavior of the client you wish to cheat.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: confused by xrandr man page example

[Thomas Lübking]

Am Wed, 10 May 2017 04:38:33 -0400
schrieb Felix Miata :

> It seems obvious why x is 1600, but y=0 seems inconsistent with the
> specification of x. Are 0 and 768 equally correct for y=768?

Also in that manpage ;-)
> --panning
> [...]
> A  width  or  height  set to zero disables panning on the
> according axis."

>   --panning 1600x384

Doesn't work for me, 
> X Error of failed request:  BadMatch (invalid parameter attributes)
>  Major opcode of failed request:  140 (RANDR)
>  Minor opcode of failed request:  29 (RRSetPanning)
>  Serial number of failed request:  31
>  Current serial number in output stream:  31

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: sending mouse events to another Xorg server

[Thomas Lübking]

see extensions/XTest.h, notably XTestFakeButtonEvent and XTestFake*MotionEvent

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: X11 Color Map creation problem, XAllocColorCells call throw BadValue error

[Thomas Lübking]

looks like npixels is coffset is 0 on vnc, right?

On Mon, Apr 10, 2017 at 09:48:48AM -, Subrata Dasgupta wrote:

Hello All,
I have an 32 bit X11 application running on RHEL 7 and use 
libX11-1.6.3-2.el7.i686 rpm X11 library. This application is facing problem 
while creating its own color map. Please suggest what changes need to be made 
in the following code to make it workable.

This X11 application is getting BadValue error while calling XAllocColorCells 
function when I run the application from VNC. But it is working properly on 
Citrix. Application tries to create a new color map by calling XCreateColormap.

From the gdb debugging I can see two major difference in Xlibrary call behavior 
between VNC and Citrix.

1> XDisplayCells return values are different . on VNC XDisplayCells returns 64 
but on Citrix client it return 256.

2> On VNC the XCreateColormap call is returning a bigger color map ID (compared 
with Citrix client) which maybe out of range and so bad value error occur.
For VNC XCreateColormap return 37748737 but for Citrix it return 8388609.

I am not sure how to use Xlibrary to create color map for different X clients. 
Please help.

Below is the high level code of the color map creation…

void createColormap(Display *display)
{
int i, n, snr, coffset, ncolors;
Display *dnr;
XColor dcolors[256];
Window win;
Colormap colormap0, colormap1;
Visual *vis;

win = DefaultRootWindow(display);
ncolors = XDisplayCells(display, DefaultScreen(display));   // Return 64 on VNC 
, and 256 for Citrix.
dnr = display;
snr = XDefaultScreen(display);
colormap0 = DefaultColormap(dnr,snr);
colormap = colormap0;

if(ncolors > 256) {
return;
}
if(ncolors > 2) {

vis = XDefaultVisual(dnr, XDefaultScreen(dnr));
if(vis == NULL) {
return;
}

//- portion between this dotted lines 
do not execute on VNC as ncolors =64 and thus coffset =0 
 //
if(ncolors > 16) {
if(ncolors > 128) {
coffset = 32;
} else {
coffset = 0;
}
} else {
coffset = 0;
}
if(coffset > ncolors) {
coffset = 0;
}
n = ncolors - coffset;

for(i=0; i dcolors[i].pixel = i;
XQueryColor(dnr, colormap0, &(dcolors[i]));
dcolors[i].flags = DoRed | DoGreen | DoBlue;
}
//- portion between this dotted lines 
do not execute on VNC as ncolors =64 and thus coffset =0 
 //


colormap1 = XCreateColormap(dnr, win, vis, AllocNone); // This line execute on 
both. Returns 37748737 for VNC but 8388609 Citrix.
if(colormap1 == colormap) {
return;
}

// portion between this dotted lines do not execute on 
VNC as ncolors < 256 and coffset < 128 == //
if(ncolors >= 256 && coffset < 128) {
int r, g, b;
for(r=0; r < 4; r++) {
for(g=0; g < 4; g++) {
for(b=0; b < 4; b++) {
dcolors[coffset].pixel = coffset;
dcolors[coffset].flags = DoRed | DoGreen | DoBlue;
dcolors[coffset].red = (short) (r * 21845);
dcolors[coffset].green = (short) (g * 21845);
dcolors[coffset].blue = (short) (b * 21845);
coffset++;
if(coffset >= 256) break;
}
if(coffset >= 256) break;
}
if(coffset >= 256) break;
}
}
// portion between this dotted lines do not execute on 
VNC as ncolors < 256 and coffset < 128 == //

unsigned long plane_masks_return[1];
unsigned int nplanes= 0;
unsigned long pixels_return[256];
unsigned int npixels = coffset;
int status;
status = XAllocColorCells(dnr, colormap1, True, plane_masks_return, nplanes, 
pixels_return, npixels); //this call throw BadValue error.
if(status == 0) {
return;
} else {
XStoreColors(dnr, colormap1, &(dcolors[0]), coffset);
colormap = colormap1;
}
}
}

Thanks
Subrata



___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s


___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: seg fault in Fedora 25 with all desktop managers marco error 4 in libX11-xcb.so.1.0.0

[Thomas Lübking]

On Tue, Apr 04, 2017 at 10:21:39AM -0400, Robert Kudyba wrote:


Rebooted. Still seg fault. I created a Bugzilla entry at 
https://bugzilla.redhat.com/show_bug.cgi?id=1438059



Just a hunch but you're loading the (builtin) modesetting driver, vesa
and fbdev and they're apparently all in the process of probing the HW
and then you run into a pciaccess segfault.

=> try uninstalling xf86-video-vesa and xf86-video-fbdevhw (so you'll
only use the modesetting driver since apparently modesetting itself
seems to work and you don't have xf86-video-ati installed) and see
whether that makes any difference.
Also see eg. https://bugs.freedesktop.org/show_bug.cgi?id=81678 and
google for pci_device_vgaarb_init - this seems a quite common issue...

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Have keyboard events "bypass" grabbed pointer

[Thomas Lübking]

Am Thu, 16 Feb 2017 09:27:32 +1000
schrieb Peter Hutterer :

> find documentation for xlib, that's usually quite ok for these bits
> (or the protocol directly). But yes, ReplayKeyboard ungrabs, so you
> have to immediately grab the keyboard again.

If the regrab fails, this is likely because some other client (the WM?)
grabbed the keyboard for the input. Check the status return code and
cycle regrab attempts. If they fail "too long" cancel the screenlocker
(while that's of course still a security risk, it's at least exposed)

If you've a specific setup, it's more reliable to pass down the
keypress by a side channel (dbus, or eg. alter an x property on the
root window which the WM tracks) than to open the keyboard.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Have keyboard events "bypass" grabbed pointer

[Thomas Lübking]

grab synchronously and conditionally xcb_allow_events (for acceptable keys)
Notice that in case the locker does not yet grab synchronously, you may
screw input processing and may have to kill/disconnect the locker (and
then fix the dead lock condition) during development.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: "Fatal server error: no screens found"

[Thomas Lübking]

On Mon, Feb 06, 2017 at 10:55:48PM -0600, Blaze Gottschalk wrote:

[47.114] (EE) Failed to load module "intel" (module does not exist, 0)
[47.115] (EE) Failed to load module "vesa" (module does not exist, 0)
[47.116] (EE) Failed to load module "fbdev" (module does not exist, 0)


Leaving aside the potential kernel mismatch, this looks as if no video
driver is installed. => ensure to instal install xserver-xorg-video-intel

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: X won't restart without reboot

[Thomas Lübking]

On Mon, Jan 23, 2017 at 12:03:30PM -0500, Jim4Prez wrote:

https://bbs.archlinux.org/viewtopic.php?id=31
...
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Multiplexing general windows similar to terminator

[Thomas Lübking]

http://kde-apps.org/content/show.php/?content=157471
You'll have to substitute QApplication and maybe Xlib with xcb.

And you're aware that you're cheating your homework and thus basically
failed the class?
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Xt / Motif C++ application hangs in the XtDispatchEvent call

[Thomas Lübking]

On Fri, Dec 09, 2016 at 09:14:22AM -, Subrata Dasgupta wrote:

after finishing _XtDefaultDispatcher() call.

There's no such "call" - _XtDefaultDispatcher is assigned to a local
variable twice (once conditionally), ie. there's either an ABI issue or
the code is not actually hanging there, but it's just the position you
randomly jumped into when gdb-attaching to the cycling process.

https://cgit.freedesktop.org/xorg/lib/libXt/tree/src/Event.c

It's more likely to hang in XtDisplayToApplicationContext() because the
application is processing events after closing the display or something.
Alternatively LOCK_APP dead locks.

Figure where the problem *really* is.

Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Question about X on the arm's.

[Thomas Lübking]

On Mon, Nov 28, 2016 at 11:59:26PM -0500, Gene Heskett wrote:


root   797  0.0  1.9 271380 33600 tty7 Ssl+ 20:50
0:02 /usr/lib/xorg/Xorg -core :0 -seat
seat0 -auth /var/run/lightdm/root/:0 -nolisten tcp vt7 -novtswitch

gene  4114  0.0  0.0   5564   908 pts/8S+   23:48   0:00
grep --color=auto listen



builders took advantage of that to get rid of 100k of object code.  So I
have challenged the odroid people to actually make it work.

Got any clue on this or just a hunch? There's an explicit nolisten
switch that didn't end up there magically but will oc. prevent tcp
access for sure.


That server starts automatically.

That's the "problem" - ask the odroid people about "automatically" (i've
no experience with that system, sorry)


The line that starts it, xinit/xsessionrc specifically says -listen tcp
as an argument for /usr/bin/X


I doubt the call is simply altered from listen to nolisten.
Pass it some unexpected bullshit argument ("-gnarf") and see whether
that has some impact - otherwise i'd say whatever starts X in that
environment doesn't care about xinit/xsessionrc at all.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Question about X on the arm's.

[Thomas Lübking]

On Sun, Nov 27, 2016 at 07:29:14PM -0500, Gene Heskett wrote:


Okay anyone, I've had 3 or 4 folks over the last 36 hours claim that X is
its own forwarding agent, why am I even using ssh? saying I'm not
needing ssh at all.


pass it "-listen tcp", it should open a connection on port 6000+n (where
n is the server, starting with 0) and you can "export
DISPLAY=remotedomain:0.0" to run X11 clients on that server.

Notice that this is no good idea unless on a local, friendly network.
Also, all caveats with "modern" toolkits apply here the same (they tend
to be dead slow over tcp because of the massive image putting, you'll
only get indirect GL etcetc.)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Question about X on the arm's.

[Thomas Lübking]

On Mon, Nov 28, 2016 at 11:15:52AM -0500, Gene Heskett wrote:

remove the '-nolisten tcp' from its launch of
exec /usr/bin/X "$@"
in xinit/xserverrc, and that would enable it, no one else has said to
just remove the "no".

What would be the nmap line to show that it is indeed listening?


Well, that didn't get me a listener according to an nmap scan of the
odroid server from this machine, after I was logged into the
odroid "server" and x was running.



Did you ensure the server is started invoking this file (by startx)?
   ps aux | grep listen

   nmap -p 6000 
should then be open if X runs on display :0

In doubt, start the server by hand "Xorg -listen tcp :0" to test the
behavior.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Installed Nvidia GT630 and S server fails to boot up.

[Thomas Lübking]

On Mon, Oct 24, 2016 at 02:16:35PM +0900, aki wrote:

By the way, the GeForce is not recognized by my BIOS/UEFI and by GRUB, 


It's more likely that the windows update screwed up things since this
happens on the VESA level, ie. whatever is there and responds to the
call will be used as visual output by a dead old, rock solid,
universally supported protocol. No response there means your UEFI or
hardware is broken.

For the other thing (ie. if the above claim is some false assumption of
yours about what UEFI or GRUB actually are), simply install the nvidia
or nouveau driver packages from Mint.

None of this is btw. even remotely X11 related (except for, you
apprently didn't install the relevant Xorg drivers - distro matter)


With nomodeset

Don't do that with the intel IGP, afaik no longer supported by the
kernel module. *ONLY* relevant for the nvidia kernel module.

In general, I suggest to attend the newbie corner of your distros forum
and ask for help. You're rather wrong on *this* list with "how to
use linux" questions.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: XQuartz fails to Launch or Reinstall

[Thomas Lübking]

On Mon, Oct 10, 2016 at 11:57:04PM -0700, Jeremy Huddleston Sequoia wrote:

Apple didn't do that.  The user did.  We ship with /tmp as a symlink to 
/private/tmp with correct permissions.


Ok, sorry.
Maybe you should consider chattr'ing some things against your users ;-)
(I mean, it's odd that ppl. manage to constrain permissions and otoh
fail to understand what they do. They'll have required root privs itfp
*shrug*)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: XQuartz fails to Launch or Reinstall

[Thomas Lübking]

On Tue, Oct 11, 2016 at 02:12:18AM +, Wendy M Calvin wrote:

Thanks for your response.

I get

touch: /tmp/.tX0-lock: Operation not permitted


I get the same error “Operation not permitted” if I try to mv the /tmp 
directory to something like tmp_old in order to recreate the link.


According to
https://discussions.apple.com/thread/482205?start=0=0 /tmp is now
755, root:wheel what means only root can write there.
You could add yourself to wheel and make it 775, but I'm not sure why
apple chose those (idiotic) permissions and it's likely not a persistent
change (reg. the /tmp permissions, group assignments should be fine)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Need help understanding X server freeze

[Thomas Lübking]

On Thu, Oct 06, 2016 at 12:31:59AM +0530, jeetu.gol...@gmail.com wrote:


Can you think of something else we can try so we can further pinpoint
where the problem lies or any thoughts at all on the above?


Only more wild guesses, maybe input/event handling?
GTK_IM_MODULE=xim GDK_CORE_DEVICE_EVENTS=1 gedit

You could also export GTK_DEBUG and GDK_DEBUG, see
https://developer.gnome.org/gtk3/3.8/gtk-running.html

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Need help understanding X server freeze

[Thomas Lübking]

On Wed, Oct 05, 2016 at 12:38:52AM +0530, jeetu.gol...@gmail.com wrote:

Hi Thomas,

Thank you for replying. I truly am at wits end here and appreciate any help.


Can you calso cause this with something >simple as "xterm" (ie, running
gnome terminal under xfce likely won't >help much)


If I X forward something simple like xterm over my Retroshare tunnel via
SSH, it works perfectly without any hiccups. Even if I try to run say a du
on / just to generate a lot of scrolling traffic.

It's only when I run something like gnome-calculator or gedit that the
whole desktop hags up at some point down the line usually when I have
pressed a button or tried to trigger a menu or dialog box.


Ie. gtk3 issue (so far), maybe GL?
LIBGL_ALWAYS_INDIRECT=1 LIBGL_ALWAYS_SOFTWARE=1 LIBGL_DRI3_DISABLE=1 gedit

If this works, I'd suspect LIBGL_ALWAYS_SOFTWARE to be the key.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: xrandr: cannot find output 0x95

[Thomas Lübking]

On Tue, Oct 04, 2016 at 08:40:19AM -0400, John Lewis wrote:


It was the R7 360 that was causing the issue. Dual screens work just
fine with Intel HD Graphics 4000.

Fits https://forums.gentoo.org/viewtopic-p-7746020.html
Seems a bug in the radeon driver then (and a quite odd one)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Need help understanding X server freeze

[Thomas Lübking]

On Tue, Oct 04, 2016 at 09:30:31PM +0530, jeetu.gol...@gmail.com wrote:


- Used Xfce4 instead of Gnome. No luck.


Can you calso cause this with something simple as "xterm" (ie, running
gnome terminal under xfce likely won't help much)


- Managed to get the modesetting driver to work with Xfce4 (didn't
work with gnome). No luck and X still freezes.


Tried staying with intel but resorting to uxa acceleration?

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Bug (?) in compose

[Thomas Lübking]

On Sat, Oct 01, 2016 at 08:02:11PM +0100, Ken Moffat wrote:


In ~/.xinitrc, try adding

export GTK_IMMODULE=xim


Better try
export GTK_IM_MODULE=xim

;-)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: xrandr: cannot find output 0x95

[Thomas Lübking]

On Fri, Sep 30, 2016 at 07:20:37PM -0400, John Lewis wrote:


john@thunderguard:~$ xrandr --output DVI-0 --mode 1920x1080 --pos 1936x0
--rotate normal
xrandr: cannot find mode 1920x1080


Ok, like before.


john@thunderguard:~$ xrandr --output HDMI-3 --mode 1920x1080 --pos 16x0
--rotate normal
xrandr: cannot find output 0x95


WTF?


I don't care which output covers 0x0 I just want to have two desktops
instead of a mirrored setup.

And does this provide it or do you get (other?) errors?
xrandr --output HDMI-3 --auto --rotate normal --output DVI-0 --auto --rotate 
normal --right-of HDMI-3

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: xrandr: cannot find output 0x95

[Thomas Lübking]

On Sat, Oct 01, 2016 at 02:21:29AM -0400, John Lewis wrote:


And does this provide it or do you get (other?) errors?
xrandr --output HDMI-3 --auto --rotate normal --output DVI-0 --auto
--rotate normal --right-of HDMI-3

Cheers,
Thomas


john@thunderguard:~/.screenlayout$ ./twoscreens2.sh
xrandr: cannot find mode 1920x1080
john@thunderguard:~/.screenlayout$ cat twoscreens2.sh
#!/bin/sh
xrandr --output DisplayPort-3 --off --output HDMI-3 --mode 1920x1080
--pos 1920x0 --rotate normal --output DVI-0 --mode 1920x1080 --pos 0x0
--rotate normal

Nope


So what happens on the auto + relative call? The twoscreens2.sh output
we did know.

Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Bug (?) in compose

[Thomas Lübking]

On Fri, Sep 30, 2016 at 04:25:20PM +0300, Clock Source wrote:


   : "ff"   Ufb00 # LATIN SMALL LIGATURE FF
- not work, xev output:
KeyPress event, serial 39, synthetic NO, window 0x501,
   root 0x262, subw 0x0, time 2593224, (321,71), root:(331,152),
   state 0x0, keycode 0 (keysym 0x100fb00, UFB00), same_screen YES,
   XLookupString gives 0 bytes:
   XmbLookupString gives 3 bytes: (ef ac 80) "ff"
   XFilterEvent returns: False


Everything is perfectly fine with the output. In case you see some hollow
rectangle, that's simply because your font doesn't support the glyph (and
either fontconfig is not used or cannot find a substitute that supports
the glyph either)

Ie. this is not an issue with glyph composition, but with your toolkit
or the installed fonts. Iow. you'll have to not in which applications it
does not work, which font you use there and which fonts are installed.
Afaics, the DejaVu fonts should support it, but not the monospace
variant (actually, I've no monospace font available supporting it, might
be a problem related to the ligature nature)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: xrandr: cannot find output 0x95

[Thomas Lübking]

On Sat, Oct 01, 2016 at 02:58:30AM -0400, John Lewis wrote:


I think that is what you asked for.

No. I asked for "leave explicit geometry alone and tell it to
auto-configure the screens and put one right to the other", ie. this:
xrandr --output HDMI-3 --auto --rotate normal --output DVI-0 --auto --rotate 
normal --right-of HDMI-3

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: xrandr: cannot find output 0x95

[Thomas Lübking]

On Tue, Sep 27, 2016 at 11:00:13PM -0400, John Lewis wrote:


xrandr --output DisplayPort-3 --off
...
DisplayPort-3 disconnected


No idea why arandr wants to configure a disconnected output, but this is
most likely the cause. Seems an arandr bug.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Huge mouse-click delay

[Thomas Lübking]

On Tue, Sep 20, 2016 at 02:08:49PM +0200, martin f krafft wrote:


All of them, including urxvt.


swap xf86-input-libinput for xf86-input-evdev and or try behavior under
a different environment (ie. start only an unmanaged xterm, launch
clients from there and test their behavior)



Does "export GDK_CORE_DEVICE_EVENTS=1" help anything on the matter


No, nothing :(


No surprise since urxvt isn't gdk ;-)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Huge mouse-click delay

[Thomas Lübking]

On Tue, Sep 20, 2016 at 10:34:18AM +0200, martin f krafft wrote:

Hello,

for a few days now, I'm experiencing (sometimes multi-second!)
delays between mouse click and reaction on my laptop. It's running:

xev reports the events ASAP, so the delay is at a later stage.

In the clients event handling.
=> Which clients in particular? All? Only gtk? Only Qt?
Is xterm affected?

Does "export GDK_CORE_DEVICE_EVENTS=1" help anything on the matter
(notice that exports affect only the shell, ie. you need to start the
client from the same terminal, not the desktop shell)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: x cut and paste

[Thomas Lübking]

On Fri, Sep 16, 2016 at 12:29:03AM -0400, Christopher Barry wrote:


Hi,

For some reason my X cut/paste stopped working.


Such does not exist. The data is transferred between clients on request,
using the PRIMARY or CLIPBOARD selection.


No idea why.


Check if every client (all applications) are affected.
If so, retry by explaining what you're trying to do.
If not, is simply a bug in the affected clients (or toolkit) - you
report a bug there.

(Ifff you're referring to the primary selection, ie. pasting w/ mmb, and
it fails everywhere, run "xev -event mouse | grep button" and ensure the
button is number "2".)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: display/GFX issue with Ubuntu 16.04 not seen in 14.04

[Thomas Lübking]

On Tue, Aug 23, 2016 at 06:53:13AM -0400, Jim Abernathy wrote:



On 08/19/2016 10:17 AM, Jim Abernathy wrote:
Not having much luck with the Distro forums on this issue so I 
thought I'd ask for advice here.


I've been running Ubuntu 14.04.5 for some time on Intel Core i5 GFX 
without any issues.  My display is HDMI from the PC to an Onkyo A/V 
receiver then to a Sony KDL-XBR6.


When I try a fresh install of 16.04.1, I get a display that blanks 
out for 3-5 seconds at random times depending on what is being 
displayed at the time.  I also have small abouts of video noise that 
is random horizontal lines about 1 pixel tall scattered through the 
page.


I noticed that 14.04.05 uses X server 1.17.2 and 16.04.1 uses 1.18.3.

When I do the command xrandr --verbose I see some small differences 
between 14.04 and 16.04, but it's things like 2 digits displayed 
instead of 1.  i.e. 60.0 vs. 60.00. I'm only concerned with 1080P 
HD.


I've tried this on different computers all with Intel GFX. same results.

This does not occur on other monitors or TVs.

I'd blame it on Sony and Onkyo if they didn't work so well with 
Windows, all versions, and ubuntu all versions except 16.04.


It's almost like the auto detection software in the newest version 
of X is off just enough to make my display flaky.  Kind of like the 
old analog CRT TVs that were sensitive to small tweaks to the 
horizontal and vertical adjustments.


Any help would be appreciated.

Jim A



Can someone let me know where I can ask questions regarding problems 
with X??  I thought this forum would be the place?


Though this is very most likely a bug in the kernel and not in X11,
check whether you're using the intel (likely) or the modesetting driver,
then use the other one (uninstalling xf86-video-intel could be
sufficient for this, but I never used ubuntu. Ask them.)

Since it's also likely a bug in the kernel, check "dmesg | grep intel"
before doing so. You might face the same issue as this guy:
https://bbs.archlinux.org/viewtopic.php?id=216088

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: High CPU usage - slow/lag/skip problem

[Thomas Lübking]

On Wed, Aug 10, 2016 at 07:16:57PM -0400, Zurd wrote:

Hi, me again, I'm bumping this. I realized I have a big post but I wanted
to put as much information as possible. All I want to know is how to
minimize the xrandr calls so that there's less EDID information in
Xorg.0.log.

My system is not slow yet but it should happen soon. My Xorg log is at 23M
in just 6h12m. I don't think this is normal.


Whenever "something" calls even the equivalent of "xrandr -q" the nvidia
driver will add such log entries. Given the frequency it's either some screen
config daemon or some broken (afair Qt5 used to do this) client code
determining the current screen geometry this way.

You can alter the way you start X11 (likely some DM) to pass
"-logverbose=0", no guarantees but it *should* silence this (and
everything else)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: PutImage vs BackgroundPixmap (was Re: Open Shared Memory and Render Pictures)

[Thomas Lübking]

On Sun, Aug 07, 2016 at 06:45:29PM +0900, Carsten Haitzler wrote:

things. there is still a race condition here as wm resizes window first, and
client some time later responds with a redraw.


And sometimes never :-)
But visually, this isn't true. Either the compositor can paint last or
(and that's probably more relevant) you configure the window w/o
actually resizing the frame and only resize the frame when the client
ACKs the change. This is basically what allowed us to remove gaps
between the decoration and (supporting) clients.
The pointer will easily go interim off-position, though.


actually be rather slow on some drivers

*cough* nvidia *cough*. It's not that bad anymore, though.

For KWin I had played with lazy configurations (covered by texture
scales in the compositor rather than in the client) - prone to cause
some sea-sickness, though ;-)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: compositing enabled with Radeon mobility X300

[Thomas Lübking]

On Fri, Aug 05, 2016 at 09:24:01AM -0400, Alex Deucher wrote:

How many displays are you using?  What resolution(s)?

This translates to "the output of 'xrandr -q'" ;-)
You might also try other compositors like xcompmgr or compton to rule
out a bug in the one used (xfwm?)

Finally, check the performance on low-resolution (XGA). The device is
short on RAM and compositing quite memory intense (notably in case of GL
compositing)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Radeon driver not works on PPC with X800

[Thomas Lübking]

On Thu, Aug 04, 2016 at 06:06:58PM -0500, Scott Thompson wrote:

https://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-nouveau/+bug/1609135
...
is that the machine locks up when the display is corrupted so I don't 
know how to get that information.


ssh/telnet

Otherwise the kernel would have halted and you'd be in deeper
trouble reg. any investigation on the issue (and likely wrong here ;-)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Corrupted XImage retrieved from a Window area

[Thomas Lübking]

On Sun, Jul 31, 2016 at 07:51:08AM +, Fabien Lelaquais wrote:

Thanks a lot for your answer.
Unfortunately I may not be able to rely of the Composite extension (app would 
be deployed in environments I don't control).

Regarding the XGetImage documentation, that I've read ten times:
My drawable (mainWindow) is indeed a viewable window.
It has no inferior, and an overlapping window on its center. The specified 
rectangle that I provide, which is the center part of the source window, is 
both fully visible on the screen and wholly contained in mainWindow.
I have no X error.


Ahh, sorry - my bad. I had interpreted it vv.
Your demo code works fine here - my guess would be


options:backing-store NO, save-unders NO

   ^

this one to be the troublemaker - enable the backing store (set "WHEN
MAPPED" here) and try again.


Option "BackingStore" "on"


in the screen section - or pass +bs to Xorg

Unfortunately I don't know whether the BackingStore is a hard condition
or things are supposed to work anyway (ie. this would be a bug in the
server)

Cheers and sorry for the fuzz,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Corrupted XImage retrieved from a Window area

[Thomas Lübking]

On Sat, Jul 30, 2016 at 08:34:20AM +, Fabien Lelaquais wrote:

Hi all,
I'm trying to create an XImage that represents a rectangular portion of a 
Window (because I need to be able to access the actual pixel values).
My experimentations show that if the source window has an hidden region 
(overlapping window) and if the origin of the rectangle I query is not (0, 0), 
then the data is corrupted, resulting in pixels set to 0 in the image data (and 
a black area in the XImage).

I use a raw XGetImage.


You want to use XCompositeRedirectWindow. This redirects the window into
a pixmap. (What compositors like xcompmgr do)


From man XGetImage:
---
If the drawable is a window, the window must be viewable, and it must be
the case that if there were no inferiors or overlapping windows, the specified
rectangle of the window would be fully visible on the screen and wholly
contained within the outside edges of the window, or a BadMatch error results.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Indirect OpenGL -- partially working?

[Thomas Lübking]

On Wed, Jun 15, 2016 at 05:05:28PM -0700, L. A. Walsh wrote:


Thomas Lübking wrote:

Indirect GL is confined to GL 1.4 (ie. fixed function path, notably no
glsl)
The driver and GPU support 4.5, but only on direct contexts.

You should check whether indirect GL works generally (there're somet
pitfalls), ie. whether glxgears works.
If so, you best contact the authors of the failing GL clients and ask
whether they provide a fixed function path (and how to select it)




Glx gears draws the gears but they don't move.

Despite the glxgears specific issues, the answer is "yes" ;-)


Why is it using a an SGI glx on the server (where cygwinX is running)?

Cygwin defaults to software rendering (but that's only relevant for
local clients)


glx lib or a default -- why would it be limited to 1.4?

The problem is not that the software rasterizer is limited to 1.4 but
that indirect GLX is. No matter what the local cygwin server (you may
enable experimental WGL) or the remote nvidia server can do. A remote GL
client can only make use of 1.4.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Touchpad scrolls to the top of the page upon touching

[Thomas Lübking]

On Wed, Jun 08, 2016 at 10:01:34AM +0200, Ingo Bürk wrote:

is quite annoying and reproducible in different applications. I've ran


All gtk3 based?
Does it work correctly when activating the window first?

https://bugzilla.gnome.org/show_bug.cgi?id=750994

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: How could an application know the desktop scale factor? ????

[Thomas Lübking]

On Fri, May 06, 2016 at 08:10:59PM +0300, Ilya Anfimov wrote:


It looks like there is no practice of regular scaling windows by
a compositing WM, so you can assume that scaling factor is 1.


Lacking general input redirection, this is actually an often requested
feature, but unsupported by all compositors I'm aware of.

nb. that Qt (and maybe gtk) also operate on overriding settings
(environment variables) to scale the particular client beyond the
physical resolution (not sure whether that's relevant for Alberto's
scenario)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Strange results from xdpyinfo

[Thomas Lübking]

On Sat, Apr 23, 2016 at 08:59:09PM +0300, Ilya Anfimov wrote:

Such a pity.


Oh, don't be such a cunt ;-P


   Is there any fix for this problem?


Fix the client. Reading output of a 3rd process into a static buffer is
plain broken. Not sure why it operates on xdpyinfo itfp (I guess it's
python? - Ok, now I'm bitching around) but it could likely just as much
operate on "xdpyinfo | grep resolution" or some such?

If you're downstream and cannot do anything about the broken client, but
know that it interprets the output of "xdpyinfo", alias xdpyinfo to some
filter chain in the context of the broken client.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Can I set different DPI for two displays in my system?

[Thomas Lübking]

On Dienstag, 8. März 2016 09:52:37 CEST, Łukasz Maśko wrote:

I was affraid of that... Windows, as I observed yesterday by accident, does 
some kind of window redraw, when I drag it between two screend 
with diferent 
DPIs. At certain point it - as I suppose - sends to a window a 
redraw request 
and provides it with a diffrent DPI setting, and you can easily bserve the 
window being redrawn with a ew DPI. I hoped we could obtain someting 
similar...


In theory™ such could be achieved on the client level, ie. if you move a 
window, it checks where it resides, calculates the natural dpi from device 
measures and resolution, alters (in case of Qt, not idea about gtk ;-) the 
device pixel ratio setting and cause a full update.
Of course you're still in some conflict whenever the window is between two 
screens.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Can I set different DPI for two displays in my system?

[Thomas Lübking]

Nope, sorry.

Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: What changes in xserver-1.18.x could affect KDE desktop effects?

[Thomas Lübking]

You're running into this:
https://bugs.kde.org/show_bug.cgi?id=356992

=> You need a more recent KWin (looks like you're on KDE4?) and until then simply use GLX 
instead of EGL (you're enforcing latter by some "KWIN_OPENGL_INTERFACE=egl" 
environment export)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: What changes in xserver-1.18.x could affect KDE desktop effects?

[Thomas Lübking]

Please provide /var/log/Xorg.0.log and the glxinfo output.

Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: setting primary in multihead xorg.conf causes disregard of DisplaySize

[Thomas Lübking]

On Mittwoch, 10. Februar 2016 08:05:03 CEST, Felix Miata wrote:

Sent too soon, and probably to the wrong place. I tried equivlent xorg.conf
files with Nvidia G84 and nouveau, and with ATI Cedar and radeon. Neither
have any such shortcoming, so I have to suspect this should have gone to
intel-gfx as a driver bug. Anyone here agree, or know a config solution?


tried
Option "IgnoreEDID" "true"
?

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Lost right edge of display

[Thomas Lübking]

On Dienstag, 9. Februar 2016 22:42:58 CEST, Ken Moffat wrote:

On Mon, Feb 08, 2016 at 02:23:02AM +, Ken Moffat wrote:

On Mon, Feb 08, 2016 at 01:25:39AM +0100, Thomas Lübking wrote: ...

A (good) follow up from the "this is weird" department :

I rebooted after that, and again had to power cycle the monitor to
get a full desktop to show.


According to the monitors manual, a short press of the power button (while the 
screen is on) should trigger the autosync - no need to cycle.

Better screens store the results of an autosync per resolution - no idea about 
yours, though ;-)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Lost right edge of display

[Thomas Lübking]

On Sonntag, 7. Februar 2016 21:19:48 CEST, Ken Moffat wrote:


But when I use X the right edge of the desktop is lost - perhaps
about 16 pixels, enough to lose the 'close' icon on a window at the
right edge of the desktop.

I have managed to lose the "everything full" brightness with xrandr,
but I am unsure how to try to address the display size.


check "xrandr -q" and notably the first line which should contain "current 1600 x 900" - 
if it's "1616 x 900", there's your problem.
In any case, post the output.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Lost right edge of display

[Thomas Lübking]

On Montag, 8. Februar 2016 00:16:07 CEST, Ken Moffat wrote:


Screen 0: minimum 320 x 200, current 1600 x 900, maximum 16384 x
16384


No, that's fine.


VGA-0 connected 1600x900+0+0 (normal left inverted right x axis y
axis) 434mm x 236mm


A VGA (only?) panel without a sync scan trigger???
Tried the power button (which model is it exactly)?

Otherwise you may be able to force a re-scan by altering the mode:
xrandr --output VGA-0 --mode 1280x720
xrandr --output VGA-0 --auto

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: https://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Diagnosing first vs subsequent performance

[Thomas Lübking]

Check dmesg, notably for NVRM, right after the failed gl call.
I assume any initial gl call will do, ie. running glxinfo will lead to a first 
instance glxgears on the GPU?

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Diagnosing first vs subsequent performance

[Thomas Lübking]

On Mittwoch, 20. Januar 2016 21:03:35 CEST, Lloyd Brown wrote:


[lbrown@m8g-1-8 ~]$ DISPLAY=:0.0 glxgears
Running synchronized to the vertical refresh.  The framerate should be
approximately the same as the monitor refresh rate.



try
DISPLAY=:0.0 __GL_GSYNC_ALLOWED=0 __GL_SYNC_TO_VBLANK=0 glxgears

no idea what impact a bogus swap interval would have on smi listing, though.

Cheers,
Thomas

___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: New Monitor Weirdnesses or how do I get Xorg to pay attention to my xorg.conf file?

[Thomas Lübking]

Am Donnerstag, 3. Dezember 2015 schrieb Robert Heller :

>
>

> What is nouveau?

http://nouveau.freedesktop.org/wiki/

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: New Monitor Weirdnesses or how do I get Xorg to pay attention to my xorg.conf file?

[Thomas Lübking]

VESA defined modes are
- 320×200
- 640×400
- 640×480
- 800×600
- 1024×768
- 1280×1024

No idea whether the vesa driver supports non-vesa modes, but did you try using 
nouveau instead of vesa?

/var/log/Xorg.0.log will tell you the servers opinion on what you added to your 
config.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: New Monitor Weirdnesses or how do I get Xorg to pay attention to my xorg.conf file?

[Thomas Lübking]

On Donnerstag, 3. Dezember 2015 20:58:33 CEST, Robert Heller wrote:


This is what it shows:


Can you please attach the entire log?
It sees your added modelines, but discards them (and I assume that is because 
of the VESA driver)

The onlyaccepted modes are

(**) VESA(0):  Built-in mode "1024x768"
(**) VESA(0):  Built-in mode "800x600"
(**) VESA(0):  Built-in mode "640x480"


So much is for sure.


My best suggestion would btw. to try the nouveau driver, but


compiled for 7.1.1, module version = 1.0.0
ABI class: X.Org Video Driver, version 1.0


Are you seriously using Xorg from 2006??
Any particular reason for doing so?

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: How to refresh the list of outputs detected by xrandr?

[Thomas Lübking]

On Freitag, 20. November 2015 03:21:56 CEST, Hacksign wrote:
 So anybody tell me how can I configure my Xorg in order to 
get effect of kali linux (2560x1440 with 96x97 dpi, but cursor & 
fonts are bigger)?


For fonts you can override the xft.dpi or fonconfig setting, some DE's provide 
a GUI for that. The cursor will likely just load a theme with increased size.
Again, the DE will likely allow you to configure that, you basically alter 
~/.icons/default/index.theme to include the desired one.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: fresh install Kubuntu 14.04 (15.10 same problem) XServer never starts

[Thomas Lübking]

On Samstag, 7. November 2015 22:11:48 CEST, Gregor Leskovšek wrote:


virtual box). Do You think I should try disabling double video mode

Yes, it's the apparent cause (from the backtrace, there's also a radeon bug 
about this, but i've neither experience nor an idea what's the problem here)


But, before all do I need to reinstall Kubuntu in between each such BIOS
change


Good god, no.


or is it enough that I just restart the server?

You need to reboot the system (to change the BIOS), i'd say ;-P


And do You think it would make a change if I replace RADEON with another
GPU?


Nvidia would require nouveau for PRIME support and I wouldn't expect any better 
there. The binary nvidia driver operates differently, you can try your luck 
with bumblebee, but if that's not a notebook, just disable the IGP.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Potential approach to EnterNotify problem in window manager

[Thomas Lübking]

On Donnerstag, 22. Oktober 2015 22:17:28 CEST, Ingo Bürk wrote:


This causes the problem that if the user puts their pointer "away" in
some corner, then shows the bar for a second and then hides it again,
X11 generates LeaveNotify for i3bar and EnterNotify for whichever window
is underneath the pointer. Since the user might have "focus follows
mouse", this causes said window to be focused, even though it's not at
all the intention of the user (they're not really looking to use their
mouse at all in this scenario).


KWin FFM has a general pointer position comparism. Only actual movements cause 
a focus change, not random crossing event caused by window un/maps (as eg. 
caused by popups)


Now, if i3bar was part of the window manager, we could obviously just
discard the EnterNotify in the window manager to avoid this.


Humm?
The EnterNotify is generated for something else anyway, but window detection is 
not related to the process at all. You could add a hinting property to the 
i3bar or just compare class and type properties to know that this is the i3bar.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Making xset stick on a secondary X session

[Thomas Lübking]

On Mittwoch, 14. Oktober 2015 17:00:27 CEST, Ilya Anfimov wrote:

On Wed, Oct 14, 2015 at 03:24:15PM +0800, Tony Mobily wrote:

Hi,
I tried again. I am starting to doubt that xset works properly!
I did:
$ export DISPLAY=:1
$ sudo X :1 -noreset & ...


 Confirming, reproduced, debian/jessie, X -version
X.Org X Server 1.16.4
Release Date: 2014-12-20

 Looks  like  xset  r  rate  works only when the server is active


Check the result from the VT where you set it with "xset -q" - looks good?
Now switch the VT back to the desired X server and re-check "xset -q" - looks 
bad now?

No idea whether it works, but i'd try
Option "AutoRepeat" "200 60" in some MatchIsKeyboard "on" config snippet in 
/etc/X11/xorg.conf.d/

I'm not sure whether the resetting behavior counts for a bug or is intended 
(since the same will happen if you re-plug the keyboard)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Finding the cursor

[Thomas Lübking]

On Mittwoch, 14. Oktober 2015 11:57:22 CEST, Ilya Anfimov wrote:


 However,  exactly  enlarging cursor shape or changing it's color
is a tough task -- as there is no easy way to get current  cursor
in X11.


Like eg. XFixesGetCursorImage(dpy), resp. xcb_xfixes_get_cursor_image_*

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: unclutter again

[Thomas Lübking]

PS, not as root, but as session user.
Don't ever run stuff as root unless you *really* have to.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: unclutter again

[Thomas Lübking]

You want to start unclutter with your X session scripts (what depends on how 
you start X11 and with what session, ie. KDE, Gnome or whatever)

On Mittwoch, 14. Oktober 2015 15:55:14 CEST, Gene Heskett wrote:
Because X is very poor at remembering what I have running and what screen 
its running on


Because that is totally not X11's job, there's a session manager protocol, but 
I doubt unclutter opts into it.

kmail on this workspace, and an incoming mailwatcher script that tells 
kmail to go get the mail after a new mail has been written 


KMail? KDE? "kcmshell[4|5] autostart" (the actual kcmshell command name depends 
on your distro and KDE version, sorry about that) - it's a system settings module.
Just add some "pkill unclutter; unclutter -foo -bar" there.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: unset Xft.dpi how?

[Thomas Lübking]

On Montag, 12. Oktober 2015 03:05:06 CEST, Felix Miata wrote:

Thomas Lübking composed on 2015-10-11 20:32 (UTC+0200):


Felix Miata wrote:



Remove (you need to reload the entire database) xrdb -query | grep -v
Xft.dpi | xrdb -load



This reads out the current database, strips every line containing
"Xft.dpi" and loads the result as new database.


This looks like should be what I'm after, but putting it above '. 
/etc/X11/Xsession' in Mint 17.2's /etc/X11/xinit/xinitrc effect on

cinnamon-session. Using the following ~/.xinitrc d



Did you check whether those files are invoked at all by cinnamon?


I have no idea how to find out.


xinitrc is related to xinit which is used by startx, but desktops tend to
operate on their own stuff (eg. startkde doesn't read ~/.xinitrc at all,
you would have to explicitly add it to its autostart stuff.



I know nothing about cinnamon nor Mint, sorry, but if cinnamon-session is
(typically) a script, you might find its inclusions there (otherwise best
ask cinnamon developers) You could also start a bare X server, xterm and
"strace cinnamon-session 2>&1 | grep open" to see what files it opens (iff
cinnamon-session is an ELF binary!)


I tried this ~/.xinitrc first at a test:

Then I tried:

#!/usr/bin/env bash
#xrdb -query | grep -v Xft.dpi | xrdb -load &
export LANG="en_US.UTF-8"
export LC_ALL="en_US.UTF-8"
export LANGUAGE="en_US.UTF-8"
export LC_CTYPE="en_US.UTF-8"
echo "Xft.dpi: 120" | xrdb -override
exec /usr/bin/xterm &
exec openbox
#exec cinnamon-session

That got an xterm open, from which I entered xrdb -query | grep Xft to find
120, then strace cinnamon-session 2>&1 | grep open. Screen went 
black, then I

was returned to shell prompt from which I ran startx. So next I tried this
~/.xinitrc:

#!/usr/bin/env bash
#xrdb -query | grep -v Xft.dpi | xrdb -load &
export LANG="en_US.UTF-8"
export LC_ALL="en_US.UTF-8"
export LANGUAGE="en_US.UTF-8"
export LC_CTYPE="en_US.UTF-8"
echo "Xft.dpi: 120" | xrdb -override
exec /usr/bin/xterm &
#exec openbox
exec strace cinnamon-session 2>&1 | grep open

That got both xterm and cinnamon session going, but left no 
output


exec strace cinnamon-session 2>&1 | grep open > /tmp/cinnamon.touches

it's however gonna generate a rather huge file.

No idea whether it's still up to date, but
https://forum.manjaro.org/index.php?topic=767.0

suggests there's "gnome-session-properties" - it may also fetch this out of 
some dconf, but yes: asking cinnamon expoerts sounds like a good idea.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: unset Xft.dpi how?

[Thomas Lübking]

On Sonntag, 11. Oktober 2015 07:49:19 CEST, Felix Miata wrote:

Can anyone tell me how to globally unset Xft.dpi, particularly on an
installation that does not set Xft.dpi via /etc/X11/Xresources, e.g. on
Linuxmint Cinnamon?


Override:
echo "Xft.dpi: 123" | xrdb -override

Remove (you need to reload the entire database)
xrdb -query | grep -v Xft.dpi | xrdb -load

This reads out the current database, strips every line containing "Xft.dpi" and 
loads the result as new database.
xrdb -remove wipes *everything*.

Whether and where Xft.dpi is set (KDE font config kcm enforcing a resolution?) 
by your distro, i don't know either - sorry.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: unset Xft.dpi how?

[Thomas Lübking]

On Sonntag, 11. Oktober 2015 11:40:50 CEST, Felix Miata wrote:

Thomas Lübking composed on 2015-10-11 09:38 (UTC+0200):



This would/should substitute 123 DPI for 96 DPI, correct?


Yes (though you likely already figured)


Remove (you need to reload the entire database)
xrdb -query | grep -v Xft.dpi | xrdb -load


This reads out the current database, strips every line 
containing "Xft.dpi" and loads the result as new database.


This looks like should be what I'm after, but putting it above '.
/etc/X11/Xsession' in Mint 17.2's /etc/X11/xinit/xinitrc
effect on cinnamon-session. Using the following ~/.xinitrc d


Did you check whether those files are invoked at all by cinnamon?
xinitrc is related to xinit which is used by startx, but desktops tend to 
operate on their own stuff (eg. startkde doesn't read ~/.xinitrc at all, you 
would have to explicitly add it to its autostart stuff.

I know nothing about cinnamon nor Mint, sorry, but if cinnamon-session is 
(typically) a script, you might find its inclusions there (otherwise best ask 
cinnamon developers)
You could also start a bare X server, xterm and "strace cinnamon-session 2>&1 | grep 
open" to see what files it opens (iff cinnamon-session is an ELF binary!)



xrdb -query | grep -v Xft.dpi | xrdb -load &
export LANG="en_US.UTF-8"
export LC_ALL="en_US.UTF-8"
export LANGUAGE="en_US.UTF-8"
export LC_CTYPE="en_US.UTF-8"
exec cinnamon-session


I bet that cinnamon-session ultimately re-sets the xrdb (as does KDE on 
starting up)


Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Booting without xorg running mint 17.1 cinnamon

[Thomas Lübking]

On Mittwoch, 7. Oktober 2015 20:35:23 CEST, Eric Gunther wrote:

On Wed, 2015-10-07 at 09:16 -0600, David wrote:
the full name of the driver is NVIDIA-Linux-x86_64-352.41.run  
i did try the nvidia support and they were not very helpful, they said

i had to contact the actual card manufacturer (evga), who said i had
to contact nvidia i looked through some of the nvidia forums and
they didn't have anything that was helpful.
it is mate display manager, although it is running in cinnamon. i
double checked because there were several others depending on distro
(lightdm, kdm) but mine does in fact have mdm by default.


OK


ill look at virtual terminals for mate, would that have x stopped
though?


I don't think it would have stopped X but uses another virtual terminal.


all of my f keys (2-7) result in a blank screen, f8 will switch back
and forth from the blank screen (the others will not switch back).  i
don't recall off the top of my head if there was anything with the
keys above f8 but nothing that seemed any different or i'm sure i
would remember.


OK


graphics is one of the things that doesn't make sense to me
sometimes, so when it comes to things like framebuffer i'm lost.  if i
know where to look to find out what it needs to be, and where it needs
to be changed to make that happen i can usually handle that.  anything
more and i'm at a loss though.


It is complicated. although as I mentioned before...


http://us.download.nvidia.com/XFree86/Linux-x86/256.35/README/index.html

this might be a place to look.


i also read that shift, esc, or del can gain access to boot options
while on the bios screen but the only thing i was able to get (once,
and not since then) was a grub> command line that i don't know how to
get around in.



OK, Try what Thomas has said because he is far more knowledgeable than
I:


Oh, flowers. How kind and nice =)

If the only point of this exercise is btw. to install the nvidia driver:
I kinda doubt it's really required to enter runlevel 3 or below for this.

Granted, the installer will fail to load the new kernel module while X is up, but unless 
it dares to restore the installation in turn, a reboot will "fix" that anyway.

inb4 you're currently using nouveau and end up with an unusable system, ensure 
to blacklist the nouveau kernel module:

echo "blacklist nouveau" | sudo tee /etc/modprobe.d/bl_nouvea.conf

or similar.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Booting without xorg running mint 17.1 cinnamon

[Thomas Lübking]

Do not mess around with files in /boot!

What you want to do is on the grub screen (where you can pick the OS to start)
a) stop the timer ;-)
b) select the Linux Mint entry
c) "[e]dit" it (this depends a bit on the grub setup, but usually there's either a selectable 
"button" or you just press "e"
d) append " 1" (nothing else)
e) (b)oot that command.

This should really get you into runlevel 1, ie. a root (text)shell

Runlevel 1 is a boot directly into a rootshell (no GUI, no network - but you 
can start either by hand), 3 is the CLI login (which will likely not work 
because this very interesting distro doesn't seem to install a/getty) and 5 is 
the GUI login.

I'd also say that you should rather seek advise in a Mint forum, since the 
problem seems to be specific to that distro.

Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Mouse blanker function?

[Thomas Lübking]

On Dienstag, 29. September 2015 06:05:43 CEST, Gene Heskett wrote:

It shuts off the pointer when you start typing, and brings it back when 
the mouse is moved.


"unclutter" [1], but i'm not sure whether the "-keystroke" switch works this 
way (it's usually used to hide the mouse in general if it's not moved for some time)

Cheers,
Thomas

[1] http://ftp.x.org/contrib/utilities/unclutter-8.README
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: How to get each monitor's geometry with xlib when there is more than one screen ?

[Thomas Lübking]

On Dienstag, 29. September 2015 10:56:33 CEST, Hacksign wrote:

My monitor configuration is listed at the end of this email post.
My problem:
 can not get each monitor's width & height widh xlib.
here is my code:



"monitor" = "output" != "screen" in X11 - nevertheless you're looking for
XRRScreenSize *XRRSizes(Display *dpy,
   int screen, int *nsizes);

Pass it the values you have and a reference to "int nsizes" and get back a list 
of
typedef struct {
   intwidth, height;
   intmwidth, mheight;
} XRRScreenSize;

don't forget to free that.

See http://linux.die.net/man/3/xrrscreenconfig

Notice that this only works as expected w/ the xrandr extension available. 
Otherwise you'll (afair) always have one screen with the combined size of the 
root window.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Color Management

[Thomas Lübking]

On Montag, 21. September 2015 19:26:38 CEST, Martin Kaffanke wrote:
Hi there, 

Is there a way to make colors on the display looking the same 
as on the printer, by printing a Page and compare the screen on 
sight? 


Try --gamma and --brightness on xrandr.
You can also try the more versatile (but harder to use ;-) 
http://xcalib.sourceforge.net/

"On sight" will however get you nowhere close to reliable results. There're 
special calibration devices for this purpose (like $100,- and up to really insulting 
offers ;-)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

driver installing requirement, [was Re: xorg Digest, Vol 121, Issue 4]

[Thomas Lübking]

On Samstag, 19. September 2015 16:22:14 CEST, David wrote:

as before...) and can't get xorg to stop.  Last time sudo service mdm stop
worked, but now when I enter that command I just get a black screen with no
console.


Sounds as if Mint isn't running any getty, nor autospawning them m(
(Maybe it's worse, and they don't install any getty providers by default...)

Try whether you can run "sudo agetty --noclear tty2 linux &"
That should get you a TTY on ctrl+alt+f2.

Notice that there are many getty daemons and I have no idea which are 
provided/installed my Mint.

Do not quit the display manager from a running GUI session, since that will 
simply kill the session (unclean exit)
Instead logout (so you see the login screen), then move to another VT (with a 
login shell), login there and quit the DM from there.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: maximized state and normal frame size lost after hide/show

[Thomas Lübking]

On Freitag, 14. August 2015 13:23:29 CEST, Cosmin Apreutesei wrote:

I noticed that if I unmap a window, when I map it back, it's maximized
state along with the restore-to size and position are all lost


Simple un/mapping a window should not loose its properties. Sure it's not 
destroyed?
(Otherwise some client, in doubt the WindowManager) would erase them)


(Ubuntu 10). Is there a way to get the normal size/pos so I can put it back
before mapping?


The restorage geometry is handled by the window manager internally.
You'd have to store the geometry before the window gets maximized and on 
unmapping (actually re-creation?) first set that geometry for  the 
unmaximized window and then set the maximization state (so the WM once saw the proper old 
geometry)

But again: for actual un/mapping this should no way be required.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: maximized state and normal frame size lost after hide/show

[Thomas Lübking]

On Freitag, 14. August 2015 18:00:52 CEST, Cosmin Apreutesei wrote:


Well, I'm using XWithdrawWindow() if it makes any difference, because
XUnmapWindow() doesn't properly hide minimized windows (it doesn't
send UnmapNotify).


Aha.

XWithdrawWindow() basically tells the windowmanager: please forget about this 
window. I'll now re-use it for entirely different purposes


The fact that you're skeptical of this makes me think this is a bug in
Unity.


Yesno. (Some) compositing windowmanagers (conditionally) do no unmap iconified (ie. 
minimized) windows in order to keep the redirected windows updated (when 
showing previews in exposé clones or the taskbar tooltips etc.)



The correct solution depends on what you're trying to do, but if you want to know when a 
window was minimized (in a managed environment and that *is* supposed to work 
with compositing WMs as well) you'd rather track the WM_STATE¹ and/or _NET_WM_STATE² (for 
_NET_WM_STATE_HIDDEN) properties.

If you want to minimize the window, the proper approach is to ask³ the 
windowmanager to do so.

FYI, X11 is not an API but a protocol. APIs like Xlib or xcb implement it and 
user code would rather use toolkits like Qt, gtk, fltk, ... which more or less 
nicely abtract the usual tasks.


function restore()
   if minimized() then
  unminimize()
   elseif maximized() then
  unmaximize()
   end
end

If that's not provided by the API already, how can I possibly
implement the above with an async API?


I'm not actually sure where you see an async problem in the above - minimized would query formentioned WM_STATE 
property (as minimized does *not* equal unmapped - notably windows can be unmapped w/o being tagged 
iconified, eg. when being shaded; for unmapped use XGetWindowAttributes[4]) and of course X11 
knows something like a blocking (not actually, though) process via XSync()[5] which will ensure at least the calls of your client 
have been processed (how other clients react to this is a different question - if you want to wait for a reaction of the window 
manager process, you obviously need to wait for that)

You're confusing things here in that X11 is, as mentioned, NOT an API and Xlib is a very 
low-level API. It's not meant to be simple.
Also Wayland is like X, but unlike Mir, a protocol - (rather) not an API.

Doing what you suggested in an abstraction layer like eg. Qt is indeed /that/ 
simple as you wrote it down.


Cheers,
Thomas

[1] https://tronche.com/gui/x/icccm/sec-4.html
[2] 
http://standards.freedesktop.org/wm-spec/wm-spec-latest.html#idm140200472615568
[3] 
https://tronche.com/gui/x/xlib/ICC/client-to-window-manager/XIconifyWindow.html
[4] https://tronche.com/gui/x/xlib/window-information/XGetWindowAttributes.html
[5] https://tronche.com/gui/x/xlib/event-handling/XSync.html
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Evga geforce730 with nvidia 352.30 driver in mint cinnamon 17.1

[Thomas Lübking]

Are you sure 1000x800 is a native resolution of that screen??
If not, and it's LCD, image will look horrible (the driver simply scales 
1000x800 to some native resolution of that screen, eventually even 640x480)

640x480 is VGA - an industrie standard and the ultimate failsafe (since 
something that doesn't support VGA is simply unsopported)

What's the output of xrandr -q and -to be sure- the model description of the 
HP monitor?

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Evga geforce730 with nvidia 352.30 driver in mint cinnamon 17.1

[Thomas Lübking]

On Sonntag, 9. August 2015 15:51:37 CEST, David wrote:

It's HP Compaq LA1956x which is supposed to get '1280 x 1024 (60 Hz) analog
input' but I found that to be a little small.
I'm not sure why but it refers to it as a CRT, I guess just because it's
VGA...


xrandr -q
Screen 0: minimum 8 x 8, current 2366 x 800, maximum 16384 x 16384
VGA-0 connected primary 1000x800+1366+0 (normal left inverted right x axis y 
axis) 0mm x 0mm
   640x480 59.9*+
   320x240 60.1
DVI-D-0 disconnected (normal left inverted right x axis y axis)
HDMI-0 connected 1366x768+0+0 (normal left inverted right x axis y axis) 160mm 
x 90mm
   1366x768   59.8*+
   1920x1080  60.0 59.930.0 24.0 60.1 60.0
   1280x720   60.0 59.9
   1024x768   75.0 70.1 60.0
   800x60075.0 72.260.3
   720x48059.9 60.1
   640x48075.0 72.8 59.9



I'd say the monitor doesn't provide any EDID data (or the transfer or data is 
broken) - check /var/log/Xorg.0.log on this!)

a) Don't try to drive it on 1000x800 - the outcome will always be blurry. Set 
it to 1280x1024
b) if you can, use the DVI connection. It will not only provide you a better 
image, but likely also bypass this problem.

Other than that you need to either
a) obtain the EDID data from the monitor (there're tools to dump it, but you'll 
most likely get no data or broken data as well)
b) write/calculate a modeline and add that (for this monitor) to 
/etc/X11/xorg.conf[.d/some_snippet.conf]

For the latter eg. fill this form with the data from the monitors tech sheet:
http://xtiming.sourceforge.net/cgi-bin/xtiming.pl

Cheers,
Thomas

PS: really try using DVI ;-)
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Evga geforce730 with nvidia 352.30 driver in mint cinnamon 17.1

[Thomas Lübking]

On Sonntag, 9. August 2015 16:26:59 CEST, David wrote:

Yeah I was actually thinking about trying dvi, I was gonna try HDMI but the
monitor doesn't have it.  I was thinking the display port was HDMI.  Sigh.
It looks just as bad but smaller at 1280x1024.  Thanks for the help!



That's (likely) because the monitor is actually still at 640x480 (the line with 
the asterisk)
What happens then is that the GPU scales down 1280x1024 to 640x480 and send 
that to the monitor which blows it up to 1280x1024 again.
The result is -expectably- suboptimal.

Changing the logic resolution won't resolve any of your issues.
You need the list to show up 1280x1024 and that requires usable EDID, an explicit 
modeline (though you can also make use of xrandr --output VGA-0 --newmode [...], but 
that's not persistent) or - hopefully - just using the digital connection ;-)

Cheers,
Thoams
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: intel atom D525MW w/i915 onboard gfx vs 16x9 monitor

[Thomas Lübking]

On Samstag, 18. Juli 2015 07:51:41 CEST, Gene Heskett wrote:

The default seems to be 1280x768.  There is a 13nnx800 mode but the 
images are bigger than the screen and circles are still smunched 
vertically. 17 AOC monitor on the machine doing the squashing.


What are complete output of xrandr -q, contents of /var/log/Xorg.0.log and 
the exact monitor model?

Usually the system asks monitor and IGP for supported modes and presents you 
the intersection - 1280x768 is 16:9.6, but if the monitors physical pixels 
aren't square, this can end up as 16:9

Cheers,
Thomas

___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: How to change mouse cursor to standard hand cursor

[Thomas Lübking]

First of all *you* do not deploy apps with toolkits, but dynamically link those 
libraries so it will use the version provided by the distro.
Linux doesn't do the windows dll hell.

And yes: using a toolkit is probably a good idea.


Secondly:
-
xcb_cursor_context_t *ctx;
if (xcb_cursor_context_new(connection, defaultScreen, ctx) = 0) {
  xcb_cursor_t cursor = xcb_cursor_load_cursor(ctx, pointing_hand);
  if (cursor == XCB_CURSOR_NONE) // they come with various names ...
 cursor = xcb_cursor_load_cursor(ctx, hand2);
  if (cursor == XCB_CURSOR_NONE)
 cursor = xcb_cursor_load_cursor(ctx, hand);
  if (cursor == XCB_CURSOR_NONE)
 cursor = xcb_cursor_load_cursor(ctx, hand1);
  if (cursor == XCB_CURSOR_NONE)
 cursor = xcb_cursor_load_cursor(ctx, pointer);
  if (cursor == XCB_CURSOR_NONE)
 cursor = xcb_cursor_load_cursor(ctx, e29285e634086352946a0e7090d73106);
  if (cursor == XCB_CURSOR_NONE)
 cursor = xcb_cursor_load_cursor(ctx, 9d800788f1b08800ae810202380a0822);
  if (cursor != XCB_CURSOR_NONE) {
 xcb_change_window_attributes(connection, window, XCB_CW_CURSOR, cursor);
  }
  xcb_cursor_context_free(ctx);
}


Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: How to change mouse cursor to standard hand cursor

[Thomas Lübking]

On Freitag, 17. Juli 2015 17:04:16 CEST, Cosmin Apreutesei wrote:

Um, the Qt/X11 deployment page specifically recommends statically
linking of Qt and even libstdc++[1].


This is if you intend to ship binaries of your stuff, what's pretty uncommon in 
the linux ecosystem.
Distribution is usually done by, well, distributions.


Where did you get the names (especially the weird hashes)?


Once looked them up in various cursor themes distributed on eg. kde-look.org to write a 
script¹ that sanitizes themes (ie. have all variants through symlinks)

Unfortunately, toolkits - let alone applications written on XLib/xcb - tend to 
differ in their opinion on what funky names to use...
There was an attempt for a freedesktop standard back in 2001 or so, but it 
miserably failed (to be adopted)

The hashes are hashes of original (unthemed) X11 cursor images which some 
applications look/ed up.

Cheers,
Thomas

[1] http://www.kde-look.org/content/show.php?content=161037
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: XIScrollClassInfo, generation of smooth scroll events from replayed buttons

[Thomas Lübking]

On Dienstag, 16. Juni 2015 06:53:31 CEST, Peter Hutterer wrote:

On Fri, Jun 12, 2015 at 09:21:34PM +0200, Thomas Lübking wrote:

See https://bugzilla.gnome.org/show_bug.cgi?id=750870

KWin passively grabs and casually replays buttons (including 4,5,6  7) on
inactive windows, but apparently that does not lead to the generation of
smooth scroll events from xinput2.

I suspect this could be related to the replay also generating framing ...


well, the problem here is that button and motion events are separate 
events 


Sorry, I forgot to cancel that.

Turned out that gtk raises a flag on crossing (as caused by the replay) and 
ignores the next motion event completely (replayed or not)

As by replaying every motion event is the first motion event and gtk also 
completely ignores wheel buttons no event processing takes place in the client (despite 
it receives all of them)

There was also https://bugs.freedesktop.org/show_bug.cgi?id=71762 about the 
inabilty to separate between XINotifyPassiveUnGrab and XINotifyUnGrab

Cheers and thanks for some more details,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

XIScrollClassInfo, generation of smooth scroll events from replayed buttons

[Thomas Lübking]

See https://bugzilla.gnome.org/show_bug.cgi?id=750870

KWin passively grabs and casually replays buttons (including 4,5,6  7) on inactive 
windows, but apparently that does not lead to the generation of smooth scroll 
events from xinput2.

I suspect this could be related to the replay also generating framing 
enter/leave events (at least I hope that's not somehow us causing them ;-)

= Is this somehow still supposed to work?

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: xserver-xorg.synaptics question

[Thomas Lübking]

On Mittwoch, 29. April 2015 02:51:04 CEST, Gene Heskett wrote:

Now, where is the last script that X executes as it completes it 
initialization? 


This depends on you DM (xdm, kdm, ldm, gdm, sddm, ...) resp. the desktop 
environment (eg. some source ~/.xprofile, but there's really no guarantee)

For LXDE, ~/.config/lxsession/LXDE/autostart should work, see
https://wiki.archlinux.org/index.php/LXDE#Autostart

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: xserver-xorg.synaptics question

[Thomas Lübking]

On Dienstag, 28. April 2015 12:47:21 CEST, Gene Heskett wrote:

Can you kind folks recommend a distribution with a lightweight gui that 
actually works?, or do I have to open this 12 year old puny powered 
lappy up and physically disconnect the touchpad?


I missed most of the thread and dunno why Mint nukes config paths, but you can 
probably use synclient to adjust the touchpad at runtime (in doubt from some 
login script)

synclient FingerLow=35
synclient FingerHigh=40

play with the values until it's nice for you. Keep Low below High.

There's also
synclient PalmDetect=1

for the clumsy among us =)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: xserver-xorg.synaptics question

[Thomas Lübking]

On Dienstag, 28. April 2015 19:16:17 CEST, Gene Heskett wrote:

You missed the point Thomas.  I have a mouse. I want that touchpad 
totally and absolutely nuked


Try
synclient TouchpadOff=1

also see
https://wiki.archlinux.org/index.php/Touchpad_Synaptics#Disable_touchpad_on_mouse_detection

on how to do that when plugging a rodent (distro doesn't matter here, arch and 
gentoo just have best docs ;-)

as if it is just an odd colored piece of 
plastic in the panel that does absolutely nothing.

Isn't there even a HW button (maybe behind some Fn) to toggle it?

If it's capacitive (likely) there're also other hardware solutions to turn it 
off -) but synclient is more elegant =)


This particular version of the pad control, called tpconfig I think, is

... a stupid generic - if you know it's a synaptics device (briefly check 
Xorg.0.conf) use synclient.

You can (likely) also disable it via xinput.


But its moot. I am now trying to get lubuntu 14.04-2 LTS to start the


I don't think things are gonna change because you switch to 
yet-another-ubuntu-derivate ;-)

Cheers,
Thomas

___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Mint14 Nadia - with interface graphic MATE - After login , video is black

[Thomas Lübking]

No log, no help :-P

- Attach /var/log/Xorg.0.log

On Sonntag, 8. März 2015 15:44:35 CEST, Antonio Vescovi wrote:

Hi, my name is Antonio and I write in this list, because I have a problem
with server X. I use Mint 14 Nadia with desktop enviroment,MATE. After the
reboot system, I do a login at my notebook, but after this, I don't  see
nothing. The video remain black. The log file I see an error code and tells
me that it is disabled . How do I enable the server x again ?? Thanks

Best regardes

Antonio




___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: DPMS switches to suspend too early (solved)

[Thomas Lübking]

On Mittwoch, 4. März 2015 10:05:53 CEST, Peter Münster wrote:


Thanks. Is this related to the DPMS problem?


In a way - ir-keytable is a bit more straight forward than lirc (RC becomes a 
keyboard and sends non-synthetic events...)


After replacing irxevent with xdotool the problem is solved, thanks!

...which, as you figured, is the core problem.

You can oc. also continue on lirc+xdotool

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: DPMS switches to suspend too early

[Thomas Lübking]

On Dienstag, 3. März 2015 21:35:01 CEST, Peter Münster wrote:


Perhaps the DPMS system does not like the synthetic YES


I'd say so - not even xdotool creates synthetic events.
Using irxevent?

You may want to look into
ir-keytable or: How I Learned to Stop Worrying about the LIRC Kernel
http://forum.kodi.tv/showthread.php?tid=101151

Alternatively you could map the input to something that creates non-synthetic 
events (ie. xdotool, dead key presses) in addition.

Cheers,
Thomas

___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Set mouse pointer resolution a.k.a sensitivity interactively?

[Thomas Lübking]

On Montag, 5. Januar 2015 16:47:47 CEST, Hi-Angel wrote:

I am one of those crazy peoples who don't want the pointer to be
accelerated: so, if I moved the mouse m space, the pointer would be
always moved n space — no matter how fast/slow I did it.


Resolution != acceleration

Many mice provide a button to control the resolution, for logitech devices 
there's http://www.lomoco.org/ and the synclient should be able to do this for 
synaptics touchpads.

To set mouse acceleration and threshold:
   xset m [acc_mult[/acc_div] [thr]]m default

ie. eg. xset m 3 8 or xset m 1 0

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Need help with M$ touchmouse

[Thomas Lübking]

On Samstag, 6. Dezember 2014 03:21:10 CEST, Gene Heskett wrote:

Is it on git, or do I have to dl a tarball?


Either. [1][2]
The problem rather is that you'll *have* to compile several packages and very 
much likely also need to update some build tools as well.



Other than the fact that even the LinuxCNC simulator would probably die

Sorry, I don't follow here. What do you mean by die?
You mean http://linuxcnc.org/ ?
I'd rather try to build that from sources (seems to be mostly python  tcl/tk 
anyway) on a parallel installation of an up-to-date distro of your choice.

Cheers,
Thomas


[1] http://ftp.x.org/pub/individual/, for mirrors: 
http://www.x.org/wiki/Releases/Download/
[2] http://cgit.freedesktop.org/, see Xorg repos.
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Need help with M$ touchmouse

[Thomas Lübking]

On Freitag, 5. Dezember 2014 03:43:44 CEST, Gene Heskett wrote:

(EE) FBDEV(0): FBIOPUTCMAP: Invalid argument


You might want to check whether there's a usable driver for your graphics chip 
provided by even that version of Ubuntu.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Need help with M$ touchmouse

[Thomas Lübking]

On Freitag, 5. Dezember 2014 18:12:09 CEST, Gene Heskett wrote:


using the nouveau driver. From a 3.16.0 kernel tree I assume.

Doesn't seem so.

Posted log segment says, that you're trying to load nv, then unload it and 
ultimately seem to be running fbdev (what causes the mny error messages)

This is not about the kernel module, but the X11 driver.
You'll need xserver-xorg-video-nouveau and libdrm-nouveau1, but rather not from 
2010.

Cheers,
Thomas

PS:
http://www.freedesktop.org/wiki/Software/libevdev/ is the very first goggle 
result here.
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Need help with M$ touchmouse

[Thomas Lübking]

On Freitag, 5. Dezember 2014 19:31:28 CEST, Gene Heskett wrote:


(II) LoadModule: nouveau
...
(EE) [drm] failed to open device


See
http://nouveau.freedesktop.org/wiki/TroubleShooting/#index1h3

From what I've read so far, I'd just bet that the updated kernel drm mismatches 
the XOrg driver.

I reconfigured, built and installed a 3.16.0 kernel with the generic fbdev 

added to the mix.
Again: this is NOT the kernel module, but the xorg driver.
You need to update XOrg (also for a new nouveau driver) or (actually) the 
entire distro.

Building XOrg is far less time consuming than back then, when X11 came as a 
blob, but it requires you to know what you need and also (likely) to update 
some toolchain elements.

I deleted most of the conversation, but since this is apparently not a 
production system: is there anything holding you back from just upgrading the 
distro version (resp. install a brand new one in parallel) to get a 
state-of-the-art SW stack?

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: letter x typed automatically and how to disable hotkeys?

[Thomas Lübking]

On Montag, 29. September 2014 18:49:07 CEST, Jeremy C. Reed wrote:

I can trigger this anytime I want by simply moving my laptop.


The notebooks accelerometer?
It could be exposed by the kernel (though usually they end up bein joysticks) and caught 
as a keyboard, typing random junk. Check xinput for strange stuff.

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: CVE-1999-0526 Vulnerbility on W2k8 R2

[Thomas Lübking]

On Mittwoch, 24. September 2014 01:45:57 CEST, Duane Fish wrote:

I just personally don't know much about the server, as it's not 
my primary responsibility. 


Then why are you trying to deal with something what you apparently do not 
understand the least instead of just telling the people in charge that they 
perhaps have a security issue they should care about?

Google will tell them how to fix this, since it's a configuration issue - not a server 
bug (ok: most likely)

Cheers,
Thomas
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s

Re: Are we need a gothic English keyboard layout?

[Thomas Lübking]

On Mittwoch, 27. August 2014 05:45:54 CEST, Hi-Angel wrote:


 So, the question is: are you need this layout?

Hardly. Those are mathemtical symbols and *not* for text at all.
Also it's no way Fraktur[1] and the only two fraktur additions in unicode are u+017f and u+2e17 
(special german fraktur letters/symbols that are not in use in latin, u+017f is a soft 
s.


Personally I think that it looks very beautiful

It will look like shit (or rather an empty box) if the client font has no 
representation for those characters.
If you want a beautiful look, install a font that suits you, eg. check 
www.dafont.com (or a thousand other font collections)



and since all over the internet using the Unicode


 *L.O.L.*

Cheers,
Thomas

[1] see eg. http://www.dafont.com/search.php?q=fraktur.de for examples of 
fraktur fonts
___
xorg@lists.x.org: X.Org support
Archives: http://lists.freedesktop.org/archives/xorg
Info: http://lists.x.org/mailman/listinfo/xorg
Your subscription address: %(user_address)s