If an implicitly redirected window is unredirected by the reparent
operation, cw will be a stale pointer.
Signed-off-by: Peter Harris <pharris@opentext.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
If a window that is fully covered by an automatic-redirected descendant
(even implicitly, eg. via BackingStores) is reparented, the automatic
updates could be broken if the new parent is not marked as having
damaged descendants.
Fix this issue by propagating the damagedDescendants flag whenever a
window is reparented.
Reviewed-by: Adam Jackson <ajax@redhat.com>
Signed-off-by: Peter Harris <pharris@opentext.com>
As of ea483af9 we're calling this unconditionally from the GLX code so
the synthetic visual is in a lower select group. If Composite has been
disabled then GetCompScreen() will return NULL, and this would crash.
Rather than force the caller to check first, just always return FALSE if
Composite is disabled (which is correct, since none of the visuals will
be synthetic in that case).
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Aaron Plattner <aplattner@nvidia.com>
This comment is above compIsAlternateVisual, which used to be the only
thing determining whether implicit redirect was needed. It's not anymore
due to the redirection exception list. That job is now performed by
compImplicitRedirect, whose code is self-explanitory.
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
When going from border width zero to a non-zero border width, the
Composite extension is informed via the ConfigNotify callback. The
call-chain looks like this: compConfigNotify -> compReallocPixmap ->
compSetPixmap -> TraverseTree -> compSetPixmapVisitWindow. However, at
this time, pWindow->borderWidth was not yet updated. Thus, HasBorder()
is false and the window border will not be repainted.
To fix this, thread the new bw through to the window visitor, and
inspect that rather than HasBorder(). For the other callers of
compSetPixmap the border does not change size, so we can pass
pWin->borderWidth instead.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=98499
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Removes the last cpp conditional on ROOTLESS from dix code.
Reviewed-by: Jasper St. Pierre <jstpierre@mecheye.net>
Signed-off-by: Adam Jackson <ajax@redhat.com>
Check the current window pixmap to see if it matches the new pixmap
and avoid calling SetWindowPixmap in that case.
x11perf -ucreate highlights an instance where compCreateWindow reassigns
the same Pixmap to the Window. Currently this triggers an expensive
invalidation of the entire window hierachy, making sure that DRI2
clients are kept informed of the handle changes. However, as the backing
Pixmap for the Window is actually unchanged, there is no need to
do anything in this case.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Tested-by: Chris Wilson <chris@chris-wilson.co.uk>
Normally composite will decide to add implicit redirection when a
window with an alternate visual is a parent of a window with a regular
visual or vice versa. This uses extra pixmap memory and incurs an extra
copy. This exception mechanism provides a way for a DDX to override this
if the DDX knows that its acceleration architecture will render correctly.
The relevant case is that of an RGB window reparented into a ARGB parent
frame window. If the DDX knows that the acceleration architecture in use
will pad the alpha channel to opaque when rendering to the RGB window,
the implicit redirection can be avoided.
This patch adds a new composite function:
CompositeRegisterImplicitRedirectionException()
which lets a DDX register a pair of parent and child window visuals, that
will not be implicitly redirected even if the normal policy would have
made that choice.
Signed-off-by: Kristian Høgsberg <krh@bitplanet.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
There's nothing between the previous fetch and this one that could
cause the window pixmap to have changed.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
This lets us stop using the 'pointer' typedef in Xdefs.h as 'pointer'
is used throughout the X server for other things, and having duplicate
names generates compiler warnings.
Signed-off-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Tested-by: Daniel Stone <daniel@fooishbar.org>
This code existed in 3 different forms, perhaps it should be
consolidated.
Reviewed-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
This is strictly the application of the script 'x-indent-all.sh'
from util/modular. Compared to the patch that Daniel posted in
January, I've added a few indent flags:
-bap
-psl
-T PrivatePtr
-T pmWait
-T _XFUNCPROTOBEGIN
-T _XFUNCPROTOEND
-T _X_EXPORT
The typedefs were needed to make the output of sdksyms.sh match the
previous output, otherwise, the code is formatted badly enough that
sdksyms.sh generates incorrect output.
The generated code was compared with the previous version and found to
be essentially identical -- "assert" line numbers and BUILD_TIME were
the only differences found.
The comparison was done with this script:
dir1=$1
dir2=$2
for dir in $dir1 $dir2; do
(cd $dir && find . -name '*.o' | while read file; do
dir=`dirname $file`
base=`basename $file .o`
dump=$dir/$base.dump
objdump -d $file > $dump
done)
done
find $dir1 -name '*.dump' | while read dump; do
otherdump=`echo $dump | sed "s;$dir1;$dir2;"`
diff -u $dump $otherdump
done
Signed-off-by: Keith Packard <keithp@keithp.com>
Acked-by: Daniel Stone <daniel@fooishbar.org>
Acked-by: Alan Coopersmith <alan.coopersmith@oracle.com>
compAllowPixmap() is not called when changing between manual and
automatic redirection modes. That means pWin->redirectDraw is left
with an incorrect value, and miComputeClips() gets confused whether
the window is supposed to be treated as transparent or not. Fix
the issue by updating pWin->redirectDraw in compCheckRedirect()
even when not calling compAllocPixmap().
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
Since extra expose events are no longer generated during window
unredirection, the window contents must be preserved by the server.
So copy the window contents back from the pixmap. The copy can only
be done after the clips have been recomputed, so delay the copy and
the pixmap destruction until ValidateTree is done. Window borders are
restored by HandleExposures and thus don't need to be copied back.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Rename compUpdateWindow to compPaintWindowToParent and split the child
walk to compPaintChildrenToWindow. Calling compPaintChildrenToWindow
allows an arbitrary subtree to be updated, instead of having to update
all the windows. This will be used to make sure all the descendants are
copied to the parent when the parent window contents need to be accessed
in IncludeInferios sub-window mode.
WindowRec has a new member 'damagedDescendants' that is used to keep
track of which subtrees need updating. When a window is damaged,
'damagedDescendants' will be set for all the ancestors, and when a
subtree is updated, the tree walk can be stopped early if no damaged
descendants are present.
CompScreenRec no longer needs the 'damaged' member since the root
window's 'damagedDescendants' provides the same information.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
TraverseTree visits the parent before the children. When performing
the automatic redirection updates, the children must be visited before
the parent.
If there are automatically redirected windows on multiple levels of the
tree, updating the parents before the children would cause the parent
updates to use stale data for areas covered by the children. Also
updating the damaged children would re-damage the parent, which would
cause additional walks over the tree.
In the worst case with an unbroken chain of automatically redirected
subwindows, all of which are damaged, only the leaf window would be
properly updated on the first round. Then it's parent would be properly
updated on the second round, and so on. And on every round all of the
ancestor windows would be updated as well, but with stale data.
So with N damaged windows you would end up with (N^2+N)/2 updates,
instead of the expected N.
This reverts commit 648c8871c9.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Reviewed-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Keith Packard <keithp@keithp.com>
Taken from:
50d2d8c896dbffd0d44a9b5b10216375f9b98af307fba8b1f7
With minor style fixes, ported to dixLookupResourceByType, and ported
away from client->noClientException and xalloc/xfree.
v2: Fix a memory leak in PanoramiXCompositeNameWindowPixmap, spotted by
James Jones.
v3: Fix a buglet in PanoramiXCompositeUnredirectSubwindows, spotted by
Dave Airlie.
v4: Fix a style issue with resource lookup noted by Jamey Sharp.
Reviewed-by: Dave Airlie <airlied@gmail.com>
Signed-off-by: Adam Jackson <ajax@redhat.com>
In all these cases, any rendering implied by this damage has already
occurred, and we want to get the damage out to the client. Some of
the DamageRegionAppend calls were explicitly telling damage to flush
the reportAfter damage out, but not all.
Bug #30260. Fixes the compiz wallpaper plugin with client damage
changed to reportAfter.
Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
v2: Use != RedirectDrawNone, since we aren't called for manual windows.
Signed-off-by: Adam Jackson <ajax@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
The fixes for the composite reallocation failure, were freeing the oldpixmap
straight after reallocating the new one, however this led to some wierd
effects in e16 compositing, and I think in a few other places. This patch
moves the freeing of the pixmap into the post wrapped stage. I'm not sure if
we are actually better off breaking ABI and adding another callback from the
ConfigureWindow to composite to make sure the old pixmap is always freed,
but this should be satisfactory as we should always hit one of the freeing
paths or else its a bug in the DIX.
bug: https://bugs.freedesktop.org/show_bug.cgi?id=28435
Reported-by: Andrew Randrianasulu <randrik@mail.ru>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
This is a combination of a huge mechanical patch and a few small
fixups required to finish the job. They were reviewed separately, but
because the server does not build without both pieces, I've merged
them together at this time.
The mechanical changes were performed by running the included
'fix-region' script over the whole tree:
$ git ls-files | grep -v '^fix-' | xargs ./fix-region
And then, the white space errors in the resulting patch were fixed
using the provided fix-patch-whitespace script.
$ sh ./fix-patch-whitespace
Thanks to Jamey Sharp for the mighty fine sed-generating sed script.
The hand-done changes involve removing functions from dix/region.c
that duplicate inline functions in include/regionstr.h, along with
their declarations in regionstr.h, mi.h and mispans.h.
Reviewed-by: Jamey Sharp <jamey@minilop.net>
Signed-off-by: Keith Packard <keithp@keithp.com>
Since reallocating the backing pixmap can fail, we need to try and do
it before any other side effects of reconfiguring the window happen.
This changes the ConfigNotify hook to return status, and moves the
composite window reconfiguration wrappers to ConfigNotify. They all
basically did the same thing, so we can drop the MoveWindow,
ResizeWindow, ChangeBorderWidth wrappers, and allow ConfigNotify to do
all the work. If reallocation fails we fail before we send any
confiureNotify events, or enter the area we can't recover from.
The only place we now enforce 32k limits are in EXA/UXA/fb, so drivers
that don't use this should probably deal with it in their pixmap
allocate if they don't already.
This also breaks ABI, so we need an alternate fix for older servers,
working on the X server makes me realise why I'm a kernel hacker.
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
Most references to the savedScreenInfo array already had the
corresponding screen pointer handy, which meant they usually looked like
"savedScreenInfo[pScreen->myNum]". Adding a field to ScreenRec instead
of keeping this information in a parallel array simplifies those
expressions, and eliminates a MAXSCREENS-sized array.
Since dix uses this data, a screen private entry isn't appropriate.
Signed-off-by: Jamey Sharp <jamey@minilop.net>
Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com>
Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux)
The problem fixed by this patch can be reproduced on Linux with the
following steps.
- Access NULL pointer intentionally in ProcessOtherEvent on key press.
- Instead of saving core dump to a file, write it into a pipe.
echo "|/usr/sbin/my-core-dumper" > /proc/sys/kernel/core_pattern
- Dump the core by pressing a key.
While the core is being dumped into the pipe, the smart schedule timer
will cause a pending SIGALRM. Linux kernel stops writing data to the
pipe when there are pending signals. This causes the core dump to be
truncated. On my system I'm expecting a 6 MB dump but the size will be
60 kB instead. The problem is solved if we block the SIGALRM caused by
expired smart schedule timer.
I haven't been able to reproduce this problem in the following cases.
- Save core dump to a file instead of a pipe.
- kill -SEGV `pidof Xorg`
- Press a key to dump core while gdb is attached to Xorg.
- Give option -dumbSched to Xorg.
Also note that the fix works only when NoTrapSignals has the default
value FALSE. The problem can still be reproduced if error signals
aren't trapped. In addition to pending SIGALRM, there is a similar
problem with pending SIGIO from the keyboard driver during core dump.
Signed-off-by: Rami Ylimaki <ext-rami.ylimaki@nokia.com>
Reviewed-by: Keith Packard <keithp@keithp.com>
Signed-off-by: Keith Packard <keithp@keithp.com>
On 64-bit systems, int and pointers don't have the same size, so GCC gives
warnings about casts between int and pointer types. However, in the cases
covered by this patch, it's always a value that fits in int being stored
temporarily as a pointer and then converted back later, which is safe.
Casting through the pointer-sized integer type intptr_t convinces the
compiler that this is OK.
Signed-off-by: Jamey Sharp <jamey@minilop.net>
Signed-off-by: Daniel Stone <daniel@fooishbar.org>
- Redo damage naming for more consistency.
- Call post submission functions only where appropriate.
- EXA can now live without it's odd damage workarounds.
The -wm (when mapped) option for the BackingStore support has been
causing the server to dereference a NULL pointer.
This has probably been the case since backing store has been
implemented on top of Composite.
It looks like (some of?) Composite didn’t expect its WIndowPtr
argument to be the root window.
In Composite’s compCheckRedirect() function we now avoid calling
compAllocPixmap() and compFreePixmap() when the pWin pointer’s
parent member is NULL, as is it the case with a server’s root window.
This addresses:
https://bugs.freedesktop.org/show_bug.cgi?id=15878
- This allows some compositing managers to work, even after randr12 has changed the root window size.
- Thanks to ajax for figuring out the best place to put this.
- Example:
- xf86RandR12SetMode() calls EnableDisableFBAccess().
- That calls xf86SetRootClip() which in turn calls ResizeChildrenWinSize().
- The final step is the call to PositionWindow().
This was an attempt to avoid scratch gc creation and validation for paintwin
because that was expensive. This is not the case in current servers, and the
danger of failure to implement it correctly (as seen in all previous
implementations) is high enough to justify removing it. No performance
difference detected with x11perf -create -move -resize -circulate on Xvfb.
Leave the screen hooks for PaintWindow* in for now to avoid ABI change.
compNewPixmap copies bits from the parent window to the redirected child
pixmap to populate the pixmap with reasonable data. It cannot always use
CopyArea as that only works across matching depths. Use Composite when
the depths do not match.
This patch changes the semantics of manual redirect windows so that they no
longer affect the clip list of their parent. Doing this means the parent can
draw to the area covered by the child without using IncludeInferiors. More
importantly, this also means that the parent receives expose events when
that region is damaged by other actions.
Add XSERV_t, TRANS_SERVER, TRANS_REOPEN to quash warnings.
Add #include <dix-config.h> or <xorg-config.h>, as appropriate, to all
source files in the xserver/xorg tree, predicated on defines of
HAVE_{DIX,XORG}_CONFIG_H. Change all Xfont includes to
<X11/fonts/foo.h>.
to me.
Clean up transition between cheap and expensive GC wrappers by using the
prologue and epilogue macros. Before, the GC would be left unvalidated
sometimes which would cause all kinds of entertaining bugs against a
DDX which cares (XAA).
Eliminate needless (and, it turns out, dangerous) call to ChangeGC on
DestroyGCPrivate.
in cwSetWindowPixmap, check if the pixmap is the screen pixmap and disable
the wrapper by setting the private to NULL.