summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2011-06-20EXA: fix typoRob Clark1-1/+1
The incorrect drawable deltas were applied if dst was a redirected window. Resulting in a bogus region passed to prepare_access_reg(). Signed-off-by: Rob Clark <rob@ti.com> Reviewed-by: Michel Dänzer <michel@daenzer.net> Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2011-06-15XQuartz: GLX: Fix a -Wformat-security compilation warningJeremy Huddleston1-1/+1
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-15XQuartz: GLX: Use _glapi_create_table_from_handle to create out dispatch tableJeremy Huddleston1-905/+17
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-15glx: Add _glapi_create_table_from_handleJeremy Huddleston3-0/+7587
_glapi_create_table_from_handle was recently added to mesa to ease creation of a _glapi_table. This commit brings this API over to xserver for use in XQuartz (next commit). The API was generated with scripts from current mesa followed by various script-foo to remove entries that are not in the server and add those which are still in the server but not in mesa (CullParameterdvEXT and CullParameterfvEXT) Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Adam Jackson <ajax@redhat.com>
2011-06-15XQuartz: Add missing include for <pthread.h>Jeremy Huddleston1-0/+2
BuildFix for Leopard and earlier. https://trac.macports.org/ticket/29717 Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-08dix: rename ProcessRawEvents to dix/events.c:DeliverRawEventPeter Hutterer3-30/+35
No functional changes, prep work for future changes. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Cyril Brulebois <kibi@debian.org> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Daniel Stone <daniel@fooishbar.org>
2011-06-08Xi: use temporary variable for filter.Peter Hutterer1-1/+4
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Daniel Stone <daniel@fooishbar.org>
2011-06-08dix: split DeliverEventToWindowMask up a bit more.Peter Hutterer1-19/+33
Move out the actual event delivery, it needs to be used from elsewhere. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Daniel Stone <daniel@fooishbar.org>
2011-06-08dix: rename DeliverEventsToClients to DeliverEventsToWindowMaskPeter Hutterer1-5/+5
To avoid confusion with a future patch and it better describes what this does anyway - delivering events to all clients that have the event mask on the window. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Daniel Stone <daniel@fooishbar.org>
2011-06-08dix: split client list retrieval out of DeliverEventToClientsPeter Hutterer1-14/+36
No functional change, but "other" was renamed to "clients". Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Daniel Stone <daniel@fooishbar.org>
2011-06-07Merge remote-tracking branch 'kibi/master'Keith Packard1-3/+4
2011-06-07Merge remote-tracking branch 'whot/for-keith'Keith Packard4-20/+36
2011-06-07dix: fix crashers with floating device.Peter Hutterer2-3/+3
dc57f89959e549403f8488eb9f23425bd7118b22 accidentally reversed the conditions. in dix/events.c we try to detach floating devices. This leads to a NULL-dereference on GetMaster()->id. in dix/getevents.c we try to get the master device for the floating slave and dereference it. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-06randr: void function cannot return valueNicolas Kaiser1-1/+1
Providing an argument to return in a function with void return type is not allowed by the C standard, and makes the Sun compilers unhappy. Signed-off-by: Nicolas Kaiser <nikai@nikai.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Alan Coopersmith <alan.coopersmith@oracle.com> Signed-off-by: Keith Packard <keithp@keithp.com>
2011-06-06xkb: Fix case checks for Latin 4.Cyril Brulebois1-1/+2
That one was missing _XkbKSLower: XK_kra: U+0138 LATIN SMALL LETTER KRA Reviewed-by: Daniel Stone <daniel@fooishbar.org> Signed-off-by: Cyril Brulebois <kibi@debian.org>
2011-06-06xkb: Fix case checks for Latin 2.Cyril Brulebois1-1/+1
Those ones were getting _XkbKSLower for no reasons: XK_ogonek: U+02DB OGONEK XK_doubleacute: U+02DD DOUBLE ACUTE ACCENT Reviewed-by: Daniel Stone <daniel@fooishbar.org> Signed-off-by: Cyril Brulebois <kibi@debian.org>
2011-06-06xkb: Fix case checks for Latin 1.Cyril Brulebois1-1/+1
That one was missing _XkbKSLower: XK_ssharp: U+00DF LATIN SMALL LETTER SHARP S That one was getting _XkbKSLower for no reasons: XK_division: U+00F7 DIVISION SIGN For reference, XK_multiply was already excluded from the _XkbKSUpper check, it's no big surprise XK_division has to be excluded from the _XkbKSLower check. Reviewed-by: Daniel Stone <daniel@fooishbar.org> Signed-off-by: Cyril Brulebois <kibi@debian.org>
2011-06-03XQuartz: AIGLX: Remove unnecessary includes in indirect.cJeremy Huddleston1-23/+9
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-03XQuartz: AIGLX: Setup dispatch table based on runtime capabilities rather ↵Jeremy Huddleston1-996/+880
than buildtime This will allow XQuartz built on older systems to pickup capabilities on newer systems and prevent runtime failures when building on newer systems and running on older ones. Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-03XQuartz: Remove explicit link against libGLJeremy Huddleston1-1/+0
This was indented to force a link against OpenGL.framework's libGL, but it actually resulted in linking against mesa's libGL due to the ordering of -L Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-03Xephyr: Mention Xserver's options are also accepted.Cyril Brulebois1-0/+2
Originally reported in the Debian BTS: http://bugs.debian.org/582650 Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> Signed-off-by: Jari Aalto <jari.aalto@cante.net> Signed-off-by: Cyril Brulebois <kibi@debian.org>
2011-06-03xkb: Fix case checks for Latin 8.Cyril Brulebois1-15/+26
Spotted by -Wlogical-op: | CC xkbfmisc.lo | xkbfmisc.c: In function '_XkbKSCheckCase': | xkbfmisc.c:104:3: warning: logical 'and' of mutually exclusive tests is always false [-Wlogical-op] | xkbfmisc.c:118:3: warning: logical 'and' of mutually exclusive tests is always false [-Wlogical-op] A quick look at the keysymdef.h file (from xproto) suggests the implementor chose to use interval checks to determine the case, but since lines weren't sorted by codepoints, checks were quite wrong. Implement _XkbKSUpper/_XkbKSLower checks based on a grep for CAPITAL/SMALL (respectively) on the Latin 8 part of the said file. Reviewed-by: Daniel Stone <daniel@fooishbar.org> Signed-off-by: Cyril Brulebois <kibi@debian.org>
2011-06-02test: fix memset size for WindowRec (#37801)Peter Hutterer1-1/+1
X.Org Bug 37801 <http://bugs.freedesktop.org/show_bug.cgi?id=37801> Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Daniel Stone <daniel@fooishbar.org> Reviewed-by: Cyril Brulebois <kibi@debian.org> Signed-off-by: Keith Packard <keithp@keithp.com>
2011-06-03Xi: use __func__ instead of function name.Peter Hutterer1-2/+2
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Cyril Brulebois <kibi@debian.org>
2011-06-03dix: use xi2_get_type instead of manual typecastPeter Hutterer1-1/+1
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-03dix: drop x/y back into the right valuators after transformation.Peter Hutterer1-7/+8
If the matrix is used for rotation, the coordinates affected may change. e.g. a valuator mask of (x, nil) becomes [x, lasty] and is rotated to [lasty, x]. Since the second value was unset, we would not drop x back into the mask, resulting in a loss of movement. Thus, drop any value that changed after applying the matrix into the valuators. Thus, the example above becomes (x, nil) → [x, lasty] → [lasty, x] → (lasty, x) Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Simon Thum <simon.thum@gmx.de>
2011-06-03dix: don't pass x/y to transformAbsolutePeter Hutterer1-13/+25
We passed in the mask, but didn't do anything with it. Move the logic to take the axes out of the valuator masks into transformAbsolute. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Simon Thum <simon.thum@gmx.de>
2011-06-03test: fix memset size for WindowRec (#37801)Peter Hutterer1-1/+1
X.Org Bug 37801 <http://bugs.freedesktop.org/show_bug.cgi?id=37801> Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Daniel Stone <daniel@fooishbar.org> Reviewed-by: Cyril Brulebois <kibi@debian.org>
2011-06-03dix: fix an error message.Peter Hutterer1-1/+1
event type, not device type. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2011-06-03test: don't test for double alignment on i386. (#36986)Peter Hutterer1-0/+3
i386 is one of the few architectures that doesn't need double alignment. X.Org Bug 36986 <http://bugs.freedesktop.org/show_bug.cgi?id=36986> Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2011-06-01XQuartz: xpbproxy: Correct NSUInteger format stringsJeremy Huddleston1-0/+27
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-01XQuartz: ASL: Use xpbproxy subsystem for logging xpbproxy messagesJeremy Huddleston2-7/+19
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-01XQuartz: ASL: Use GLXAqua subsystem for logging GLXAqua messagesJeremy Huddleston1-6/+2
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-01XQuartz: ASL: Use xpr subsystem for logging xpr messagesJeremy Huddleston4-3/+5
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-01XQuartz: ASL: Update logging to support differnet levels and subsystemsJeremy Huddleston2-4/+8
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-01XQuartz: Add a semicolon after DEBUG_LOGJeremy Huddleston1-1/+1
Don't rely on the macro to provide it Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-01XQuartz: Move -lXplugin to LDFLAGS from LDADDJeremy Huddleston1-1/+2
Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-06-01XQuartz: GLX: Create a new dispatch table rather than modifying the existing oneJeremy Huddleston1-1/+11
Fixes regression introduced by b0c665ac0fe6840dda581e4d0d0b76c703d62a7b 0 X11.bin 0x0000000100118293 __glXAquaScreenCreateContext + 684 1 X11.bin 0x00000001001315b0 DoCreateContext + 163 2 X11.bin 0x000000010013509f __glXDispatch + 211 3 X11.bin 0x00000001000c7dad Dispatch + 785 4 X11.bin 0x00000001000b97e5 dix_main + 1022 5 X11.bin 0x00000001000122bc server_thread + 50 6 libSystem.B.dylib 0x00007fff836554f6 _pthread_start + 331 7 libSystem.B.dylib 0x00007fff836553a9 thread_start + 13 http://lists.apple.com/archives/X11-users/2011/May/msg00045.html Signed-off-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Adam Jackson <ajax@redhat.com>
2011-06-01DIX: Set backgroundState correctly for root windowMarko Macek1-0/+2
When we change the root window's background to None, and we've run with -wr or -br for a forced solid background, make sure we also change the background state to BackgroundPixel, so we don't try to lookup either pScreen->whitePixel or pScreen->blackPixel as a pixmap. Signed-off-by: Marko Macek <Marko.Macek@gmx.net> Reviewed-by: Walter Harms <wharms@bfs.de> Reviewed-by: Daniel Stone <daniel@fooishbar.org> Signed-off-by: Keith Packard <keithp@keithp.com>
2011-06-01Version bumped to 1.10.99.901 (1.11 RC1)xorg-server-1.10.99.901Keith Packard1-2/+2
Signed-off-by: Keith Packard <keithp@keithp.com>
2011-05-31Merge remote-tracking branch 'ajax/xserver-next'Keith Packard8-13/+777
2011-05-31Merge remote-tracking branch 'sandmann/for-keithp'Keith Packard1-2/+0
2011-05-31fixes: Add support for pointer barriersAdam Jackson8-13/+777
Implements pointer barriers as specified by version 5 of the XFIXES protocol. Barriers are axis-aligned, zero-width lines that block pointer movement for relative input devices. Barriers may block motion in either the positive or negative direction, or both. v3: - Fix off-by-one in version_requests array - Port to non-glib test harness - Fix review notes from Søren Sandmann Pedersen, add tests to match Co-authored-by: Peter Hutterer <peter.hutterer@who-t.net> Tested-by: Peter Hutterer <peter.hutterer@who-t.net> Signed-off-by: Adam Jackson <ajax@redhat.com> Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
2011-05-27"privates.h", line 198: warning: void function cannot return valueAlan Coopersmith1-1/+1
Providing an argument to return in a function with void return type is not allowed by the C standard, and makes the Sun compilers unhappy. (They actually flag it as an error, unless using a new enough version to be able to downgrade it to a warning with "-features=extensions".) Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Reviewed-by: Daniel Stone <daniel@fooishbar.org> Reviewed-by: Cyril Brulebois <kibi@debian.org>
2011-05-27Use XORG_STRICT_OPTION from util-macros 1.14 to set -Werror flagsAlan Coopersmith1-4/+2
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-05-27Use XORG_COMPILER_BRAND from util-macros 1.14 to check for SUNCCAlan Coopersmith1-5/+3
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com>
2011-05-26Add a property for device/product ID.Peter Hutterer1-0/+5
In some cases, knowing about the device model number and the device's vendor is important to activate product-specific settings. Since this is nonetheless driver-specific, only provide the property but don't do anything with it. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Daniel Stone <daniel@fooishbar.org>
2011-05-26Xi: add device node property to known properties.Peter Hutterer1-0/+3
Since the server has little choice (or even knowledge) of the actual device node used by the driver, this property is merely provided for standardisation. It is up to the driver to set it to the appropriate value, usually a device node in the form of /dev/input/event0 or similar. Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net> Acked-by: Dan Nicholson <dbn.lists@gmail.com>
2011-05-25Don't call pixman_disable_out_of_bounds_workaround() anymoreSøren Sandmann Pedersen1-2/+0
Pixman used to have a workaround for a bug in old X servers, and this function was used to disable that workaround in servers known to be fixed. Since 0.22, which the X server depends on, the workaround doesn't exist anymore, so there is no point disabling it. Reviewed-by: Cyril Brulebois <kibi at debian.org> Signed-off-by: Soren Sandmann <sandmann@cs.au.dk>
2011-05-25randr: check rotated virtual size limits correctlyAaron Plattner1-6/+6
Commit d1107918d4626268803b54033a07405122278e7f introduced checks to the RandR path that cause RRSetScreenConfig requests to fail if the size is too large. Unfortunately, when RandR 1.1 rotation is enabled it compares the rotated screen dimensions to the unrotated limits, which causes 90- and 270-degree rotation to fail unless your screen happens to be square: X Error of failed request: BadValue (integer parameter out of range for operation) Major opcode of failed request: 153 (RANDR) Minor opcode of failed request: 2 (RRSetScreenConfig) Value in failed request: 0x780 Serial number of failed request: 14 Current serial number in output stream: 14 Fix this by moving the check above the code that swaps the dimensions based on the rotation. Signed-off-by: Aaron Plattner <aplattner@nvidia.com> Reviewed-by: Jeremy Huddleston <jeremyhu@apple.com> Tested-by: Robert Hooker <robert.hooker@canonical.com> Tested-by: Kent Baxley <kent.baxley@canonical.com> Signed-off-by: Keith Packard <keithp@keithp.com>