| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
want that. Likely.
|
|
|
|
| |
Fixes ArcticaProject/nx-libs#701
|
|
|
|
| |
Fixes ArcticaProject/nx-libs#694
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
nxagentHandleConfigureNotify() has an optimization that accumulates
ConfigureNotify events from the queue to only perform the changes of
the last ConfigureNotify event in the queue. But that code used to
ignore position changes and only adapt the new window position if the
last event happened to contain a position change.
This change ensures the latest position change - if any - found in the
queue will be applied after the accumulation.
Fixes: ArticaProject/nx-libs#688 (second part)
|
|
|
|
| |
Was 0,0 on every call...
|
| |
|
|
|
|
|
|
| |
This is the preferred way. It ensures, that
a) the data is zeroed
b) a correct size in case xlib is changing the size of the structure in future releases
|
| |
|
|
|
|
| |
nxagentSetWMNormalHints was slightly adapted therefore
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
nx-X11/ subdir (except nx-X11/lib/).
|
| |
|
| |
|
|
|
|
| |
UseTIRPC.
|
| |
|
|
|
|
| |
since libXcompext doesn't actually need it.
|
|
|
|
| |
instead of echo for variable value.
|
|
|
|
|
|
|
|
|
|
| |
for nxagent regeneration.
Updating timestamps on object files and libraries used to compose
nxagent is only meaningful for actual files within the buildroot.
External dependencies would be ignored anyway, and worse, cause older
GNU Make versions to fail if they contain special characters like a
percent sign, which is typically interpreted as a wild card character.
|
|
|
|
|
|
| |
if the output was disconnected before.
Fixes: ArcticaProject/nx-libs#677
|
|
|
|
| |
avoid referencing random data.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
commit cad5a1050b7184d828aef9c1dd151c3ab649d37e
Author: Nathan Kidd <nkidd@opentext.com>
Date: Fri Jan 9 09:57:23 2015 -0500
Unvalidated lengths
v2: Add overflow check and remove unnecessary check (Julien Cristau)
This addresses:
CVE-2017-12184 in XINERAMA
CVE-2017-12185 in MIT-SCREEN-SAVER
CVE-2017-12186 in X-Resource
CVE-2017-12187 in RENDER
Reviewed-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
Reviewed-by: Julien Cristau <jcristau@debian.org>
Signed-off-by: Nathan Kidd <nkidd@opentext.com>
Signed-off-by: Julien Cristau <jcristau@debian.org>
Backported-to-NX-by: Mike Gabriel <mike.gabriel@das-netzwerkteam.de>
|
|
|
|
| |
implemented in intersect_bb() function.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If the agent window is moved around on screen, it can happen
that it is moved into an invisible area of the real Xserver,
we calls this "beyond the bounding box".
.
If the agent window is partially beyond the bounding box, we
don't want Xinerama to re-adjust the RandR parameters inside the
agent. Near the bounding box, the session shall stay intact.
.
This means, desktop env wise, the desktop session control
elements can be moved (with the agent window) into the invisible
areas of the real Xserver and moved out again without RandR
events arriving inside the agent session.
Fixes ArcticaProject/nx-libs#662.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Keeping it enabled leads to nxagent consuming 100% of CPU resources
after some time.
Older code used three different timers for each DPMS mode (standby,
suspend, off), with each timer checking that the elapsed idle time is
strictly less than the DPMS mode timeout value.
Newer code started merging DPMS and ScreenSaver timers into a single
one, with checking DPMS modes and timeouts in a fall-through fashion.
The code expects that, if a timeout is reached, the mode is set
accordingly, so that the old timeout is disregarded next time.
Since we stub out DPMSSetMode() in nxagent, this doesn't happen.
In this case, the old DPMS timeout will be checked the next time around
and we will be calculating DPMS_MODE_TIMEOUT - IDLE_TIME. If IDLE_TIME
is bigger than DPMS_MODE_TIMEOUT, we run into problems with unsigned
integers. What happens next will be *another* overflow once DoTimers()
is executed and from that function SetTimers(), which will add the
current timestamp to the timeout value. Later functions subtract the
current timestamp again (which will be even higher since some time
elapsed) and underflow the value again.
It looks like this leads to consistent firing of the timer - something
we do not want to have.
For new, disable DPMS support in nxagent. We do not execute it on actual
hardware that could change DPMS modes.
At a later time, implementing a DPMS timeout/mode passthrough to the
connected X server/display might be beneficial.
Fixes: ArcticaProject/nx-libs#671
|
|
|
|
|
|
| |
nothing better can be detected.
Fixes ArcticaProject/nx-libs#668.
|
| |
|
|
|
|
| |
is used nowhere anyway
|
|
|
|
|
| |
make them appear in the nx section. Also separate the nx section from the
other options by an extra emtpy line.
|
| |
|
|
|
|
|
|
|
| |
In XORG-7_1 xkbconfig.c had been dropped. We must keep this file
because it is used by hw/nxagent/Keyboard.c to load the xkb
configuration file. Maybe we should move the relevant code to there or
drop xkb configuration file support from nxagent.
|
|
|
|
| |
only effect on session startups.
|
| |
|
|
|
|
| |
magicpixel takes a boolean argument for now.
|
|
|
|
|
|
| |
of file.
Fixes: ArcticaProject/nx-libs#664
|
|
|
|
| |
left-overs.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Backported from X.org:
commit 39ce5f1544029412f4060f3e89ce1d87222ef42b
Author: Adam Jackson <ajax@nwnk.net>
Date: Fri Jan 6 17:05:26 2006 +0000
Remove unused X11R4 DDX compatibility function miClipNotify.
Backported-to-NX-by: Mihai Moldovan <ionic@ionic.de>
|
| |
|
|
|
|
| |
I left out glx changes to miinitext.c for now.
|
| |
|
| |
|
|
|
|
| |
changes only
|
| |
|