From a915739887477b28d924ecc8417ee107d125bd6c Mon Sep 17 00:00:00 2001 From: marha Date: Sun, 6 Sep 2009 18:48:27 +0000 Subject: Switched to xorg-server-1.6.99.900.tar.gz --- xorg-server/ChangeLog | 17397 +++++++++++++++++++++++++++++++++++++++++------- 1 file changed, 14877 insertions(+), 2520 deletions(-) (limited to 'xorg-server/ChangeLog') diff --git a/xorg-server/ChangeLog b/xorg-server/ChangeLog index 73582f3ad..47a09432d 100644 --- a/xorg-server/ChangeLog +++ b/xorg-server/ChangeLog @@ -1,2786 +1,14691 @@ -commit 3044711412d0a08ba65a491bd2441c0c8980f5e2 -Author: Keith Packard -Date: Tue Aug 25 22:37:16 2009 -0700 +commit e6e83d81ee7366779d600cbb128034e248815339 +Author: Peter Hutterer +Date: Fri Sep 4 14:51:08 2009 +1000 - Bump version to 1.6.3.901 (1.6.4 RC1) + xserver 1.6.99.900 - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit d8e48deb212767c4dfce982e6cb9d1b1fce63b96 -Author: Keith Packard -Date: Tue Aug 25 22:29:25 2009 -0700 +commit bc0841b6e87cdd752b44b17ccb021c831a3e02a5 +Author: Peter Hutterer +Date: Fri Sep 4 15:58:36 2009 +1000 - Block signals while delivering XTest events (fixes bug 23100) + shave: hack around the configure ld check for make distcheck. - XTest uses the same event list as all input devices, and input devices - use it at SIGIO time. This causes mass confusion when a SIGIO event - occurs in the middle of XTest event processing. + configure calls $(CC) -print-prog-name=ld to get the binary for the linker. + The shave script prints " CC \nld" which can't be parsed by configure, + resulting in a fatal error. - The master branch has a much different fix, using a separate input - queue which reduces latency that may be caused by XTest event queuing; - it's a 'nicer' fix in some ways, but this simple fix should solve the - problem. + This patch special-cases -print-prog-name and passes it though directly + without any additional output. - Reviewed-by: Peter Hutterer - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 76a1839b0a7dcb82e46c43533d252288677c6dbb -Author: Keith Packard -Date: Tue Aug 25 18:07:00 2009 -0700 +commit 58d9a4c6356c98c32be2826410df027ece70d44d +Author: Peter Hutterer +Date: Fri Sep 4 15:06:45 2009 +1000 - Ensure that rotation updates happen frequently - - The smart scheduler is designed to minimize scheduler overhead by - increasing the interval between WaitForSomething calls when a single - client is running. However, the software rotation code depends on - its BlockHandler being invoked for screen updates; the long delays - caused by the smart scheduler optimizations means that screen updates - can be delayed a long time as well. + xfree86: fix make distcheck after removal of Domain.note and RAC.Notes. - The change is simple -- prevent the smart scheduler from increasing - the scheduling interval while any screen is using software rotation. - (cherry picked from commit e7dd1efef408effe52d0bd3d3aa0b5d4ee10ed90) - - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 31dc4fe0be57da4e9458ee490811fbd92598b7a7 -Author: Keith Packard -Date: Tue Aug 25 16:58:07 2009 -0700 +commit 8af2793a90e8e644a6a598adce3f752c15b3e37a +Author: Peter Hutterer +Date: Thu Sep 3 11:30:48 2009 +1000 - Perform rotation redisplay before calling driver block handler (which may flush rendering) + xkb: xkbGetKbdByName on the lastSlave needs to change the master (#21859) - The rotation block handler uses regular driver rendering functions to - repaint the screen, if those functions queue commands in the driver, - it's important that the driver block handler be invoked after the - rotated image is drawn. + If the layout is changed on a master's lastSlave, the master needs to change + layout immediately. Otherwise, the master stays on the same layout until the + lastSlave changes - which may not happen if only a single keyboard is + available. - Signed-off-by: Keith Packard - (cherry picked from commit 1740cda7a37abc7d0a169ab4555b446adaa62211) + X.Org Bug 21859 + + Signed-off-by: Peter Hutterer -commit cebc0a7aa7a368b2ade3a808195d2e738fda83e9 -Author: Keith Packard -Date: Tue Aug 25 16:54:16 2009 -0700 +commit 219ae5b628f9ffe2be0876d5911cc04e52bef5f8 +Author: Peter Hutterer +Date: Fri Aug 21 16:38:53 2009 +1000 - xf86_reload_cursors: fix cursor position to eliminate jumping after mode set + dix: hide the sequenceNumber key repeat hack behind two functions. - xf86_reload_cursors restores the cursor to the correct position, but - that must adjust for cursor hot spot and frame before calling down to - the hardware function, otherwise the cursor jumps to the wrong - position until it is repositioned by the user. + Just in case we figure out a better way to do it in the future. - Signed-off-by: Keith Packard - (cherry picked from commit 4aab05e3b3231f1ec9795a66a075d17a722634a7) + Signed-off-by: Peter Hutterer -commit 5848fca11f3c5b65bb6634816c0df6aeb6eda87d -Author: Michel Dänzer -Date: Tue Jun 23 16:45:40 2009 +0200 +commit bfb219f532f3c78ba905424365ee7c5f7b5f21a2 +Author: Daniel Stone +Date: Fri Aug 21 16:03:36 2009 +1000 - glx: Clean up more thoroughly if the drawable of a current context goes away. + input: allow for detectable autorepeat. - Fixes crash when restarting compiz, due to cl->currentContexts[x] being stale. - (cherry picked from commit 3020b1d43e34fca08cd51f7c7c8ed51497d49ef3) - -commit fad2de1cca057586a88fb20ebf9a9964f8e3e298 -Author: Michel Dänzer -Date: Thu May 14 11:46:41 2009 +0200 - - glx: If a destroyed window is bound to the current context, make it not current. + For core and XI1 events, store the key_repeat flag in the sequence number + until TryClientEvents. The sequenceNumber is unset until TryClientEvents. - Avoids subsequent crashes due to stale pointers to the DrawableRec, see - https://bugs.freedesktop.org/show_bug.cgi?id=21132#c15 and previous comments. + [Also thrown in, some random indentation changes. Thanks] - Signed-off-by: Michel Dänzer - (cherry picked from commit 2075d4bf9e53b8baef0b919da6c44771220cd4a5) + Signed-off-by: Peter Hutterer -commit a32d717ac64b21e15a5736ae4981cdd32f2da56b -Author: Richard Hughes -Date: Mon Aug 17 09:15:32 2009 +0100 +commit 3d988e01e41e98fc5160f825a250522ba274d09f +Author: Daniel Stone +Date: Fri Aug 21 15:15:41 2009 +1000 - Do not reset lastDeviceEventTime when we do dixSaveScreens + dix: improve code flow in TryClientEvents, better debugging messages. - When we turn off DPMS with DPMSModeOff and do dixSaveScreens, don't reset the - event time else session clients using IDLETIME will be reset. + Instead of a massive if (blah && blah), return early where possible. Signed-off-by: Peter Hutterer - (cherry picked from commit 6b5978dcf1f7ac3ecc2f22df06f7000f360e2066) -commit 32115e690f43fde619234722aa674ff1ae1b44ae +commit 6fb01c8286c16968fd07e4a5b78da89cd8768d79 Author: Peter Hutterer -Date: Fri Aug 14 11:48:54 2009 +1000 +Date: Fri Sep 4 10:57:33 2009 +1000 - dix: update the sprite trace for all masters && floating slaves (#23257) - - When the windows are restructured, CheckMotion needs to be called for all - masters and floating slaves to update the spriteTrace. - - X.Org Bug 23257 - (cherry picked from commit 6c292d17053eb2a7e7054e51210f423dbc0cb7e8) + require xf86vidmodeproto 2.2.99.1 or newer. + +commit 2122fedb2c058dc27f9df0e856d2193fabe725e2 +Author: Peter Hutterer +Date: Fri Sep 4 11:37:58 2009 +1000 + + os: silence bigreqsproto compiler warning - Tested-by: Thomas Jaeger Signed-off-by: Peter Hutterer - Signed-off-by: Keith Packard -commit af7e62a913c3c5b9639b4557b31537853f3f58bb -Author: Dan Nicholson -Date: Sun Aug 16 18:07:49 2009 -0700 +commit 59e731ef66095723e74a16b2e5548aed627089ff +Author: Peter Hutterer +Date: Fri Sep 4 11:06:37 2009 +1000 - xfree86: Link libselinux with Xorg system libraries + xfree86: silence some xf86dgaproto compiler warnings - Put the reference to -lselinux in XORG_SYS_LIBS so it isn't added as a - make dependency for libxorg.la. Otherwise, make goes looking for a file - -lselinux in the current directory, which it obviously won't find. + We already require xf86dgaproto > 2.0.99 since 6fffcd582 - Signed-off-by: Dan Nicholson - Tested-by: Justin P. Mattock - (cherry picked from commit c73cd3b265c301b8a54ffe484d6c696f2abefb46) + Signed-off-by: Peter Hutterer -commit 397f7c42cd775f1dbfced58bc1dfaead48e86440 +commit ead57a13531aa14033839112fa66e33e142f67da Author: Peter Hutterer -Date: Fri Aug 14 09:48:45 2009 +1000 +Date: Fri Sep 4 11:03:55 2009 +1000 - config: don't shutdown the libhal ctx if it failed to initialize (#23213) - - Regression introduced by b1c3dc6ae226db178420e3b5f297b94afc87c94c. - Shutting down the libhal_ctx if the init failed may cause an abort. - This can happen if hald is not yet running at server startup. - - X.Org Bug 23213 + xfree86: silence 'unused variable' compiler warning - Tested-by: Stefan Dirsch Signed-off-by: Peter Hutterer - (cherry picked from commit 49046088f10cceaea7da97401d742d3fb59371f5) -commit 55c26d8e4d110b689aea9d806e9d4fa7bbbdd32a +commit 21f04f008e6b1ab66a63de3de668b1f261a6928a Author: Peter Hutterer -Date: Thu Aug 13 16:37:00 2009 +1000 +Date: Fri Sep 4 11:01:09 2009 +1000 - Xext: fix up wrong conditions for negative sync transitions. - - If the counter had a value higher than the trigger value for a negative - transition, the trigger value did not get set. - - The correct sequence of checks is: - if (positive transition) - if (counter value < trigger value) - set up trigger - if (negative transition) - if (counter value > trigger value) - set up trigger - - Red Hat Bug 501601 + os: don't redefine GNU_SOURCE Signed-off-by: Peter Hutterer - (cherry picked from commit db568f9eabf3450d8a023597ff007df355b13ea8) -commit f274e595ed56722450f6ecebad70500c92e50c58 -Author: Keith Packard -Date: Fri Jul 31 23:37:53 2009 -0700 +commit 09f0ee1324b046f2830bdd260f6e2a67ee8f2973 +Author: Martin-Éric Racine +Date: Fri Sep 4 04:17:09 2009 +0300 + + Formatted recent Geode PCI ID changes to fit a 80-column coding style. + +commit a17b50b130fe01c058a74f2696a8b97c756acb49 +Author: Martin-Éric Racine +Date: Fri Sep 4 03:51:01 2009 +0300 + + Further explained what the Geode PCI ID is all about as comments. - Bump release number to 1.6.3 +commit afd800dae4f7491baee08509094a30160bb9c849 +Author: Martin-Éric Racine +Date: Fri Sep 4 03:41:08 2009 +0300 + + Removed unnecessary curly braces for Geode LX PCI ID. + +commit 842b7a47d583a7729099de3031ae20977dddf774 +Author: Martin-Éric Racine +Date: Fri Sep 4 03:25:32 2009 +0300 + + Updated the PCI Vendor and Device IDs for all Geode variants. - Signed-off-by: Keith Packard + In practice, some of the native drivers for older Geode products + have become deprecated due to lack of e.g. libpciaccess upgrade, + but that's OK, since most distributions don't ship them anymore. + In that case, we'll let X server fall back to good old VESA. -commit 55dcfdbbd6a2a4debb717b104c4b54f4a2c76fff -Author: Alan Coopersmith -Date: Tue Jul 28 15:02:37 2009 -0700 +commit 8e873185f4cbfb2a36e1f43fe7da47fd9fd5aeea +Author: Michel Dänzer +Date: Thu Sep 3 19:11:02 2009 +0200 - Remove hardcoded gcc -Wall option from configure.ac + EXA: Preserve pPixmap->devPrivate.ptr in exaPixmapIsOffscreen_mixed. - Signed-off-by: Alan Coopersmith - (cherry picked from commit 442967c90dd9d8483a56bdc9237c49e33d619126) + Otherwise we may incorrectly clobber it to NULL on repeated PrepareAccess + calls. -commit 4d349b6c333449517913a5e981b4fc3155da4855 -Author: Adam Jackson -Date: Fri Mar 27 15:56:15 2009 -0400 +commit 120286aef59dabdb7c9fa762e08457e5cc8ec3a6 +Author: Michel Dänzer +Date: Thu Sep 3 08:05:59 2009 +0200 - selinux: Only activate if policy says to be an object manager - (cherry picked from commit 283a081572d8db787c77d09e5ba6bcadebf4f7fe) + glx: Add screen DestroyWindow wrapper to destroy the GLX drawable. + + Fixes crashes exitting MacSlow's rgba-glx demo. -commit d0875154f4f4b751264a2f6781500e97409a499b -Author: Ben Skeggs -Date: Wed May 27 15:24:34 2009 +1000 +commit f04fe06ae244b851b38be824b1a80f2f8a030591 +Author: Peter Hutterer +Date: Thu Sep 3 14:17:14 2009 +1000 - quirk: use first detailed timing as preferred for PEA prod 9003 (rh#492359) + dix: don't return BadMatch from GetProperty (#23562) + + dixLookupWindow may return BadMatch if the window in question isn't actually + a window. In this case, GetProperty needs to return BadWindow - not + BadMatch. + + X.Org Bug 23562 Signed-off-by: Peter Hutterer - (cherry picked from commit 048697ccfa31cf7f7a29afa90a2f702d43efb7d4) -commit c941479ecc2dead9c3deaee2620c9b9518c3da9a -Author: Rémi Cardona -Date: Mon Jul 27 12:07:51 2009 +0200 +commit 84eb4c66a4a09c360cef260fb2f35dfb6d8a93c6 +Author: Peter Hutterer +Date: Thu Sep 3 10:00:42 2009 +1000 - config: add HAL error checks + test: add protocol testing for XIWarpPointer. - This patch simplifies error handling in the HAL code and fixes a - segfault if libhal_find_device_by_capability() failed. + TODO: some way to check src_x/y coordinates would be good. - Fixes http://bugs.gentoo.org/278760 + Signed-off-by: Peter Hutterer + +commit 1b7858e8469aea6d2031039ba41d7191a4b80f28 +Author: Peter Hutterer +Date: Thu Sep 3 10:58:00 2009 +1000 + + test: expose the default screen to tests, some cleanup work. - Based on a patch by Martin von Gagern + Provide common #define for invalid window IDs. + Init the sprite's hotPhys, provide a common #define for the initial sprite + position. - Signed-off-by: Rémi Cardona - Acked-by: Peter Hutterer - (cherry picked from commit b1c3dc6ae226db178420e3b5f297b94afc87c94c) + Signed-off-by: Peter Hutterer -commit 546f913ff5461dd93d4a0b29b24d2267557326c7 -Author: Alan Coopersmith -Date: Fri May 8 21:31:01 2009 -0700 +commit 61a6e1f074d9ff75d61446b946aab6c04019c287 +Author: Peter Hutterer +Date: Thu Sep 3 10:00:00 2009 +1000 - Don't printf NULL pointers on HAL connection error - - Fixes Solaris bug 6801386 Xorg core dumps on startup if hald not running - http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6801386 + Xi: return BadDevice for master kbds and attached slaves in XIWarpPointer - Signed-off-by: Alan Coopersmith - (cherry picked from commit 1e816065e5ec3b9394dc1fa5815457a664e15fd9) + Signed-off-by: Peter Hutterer -commit 155cb2f9a376d40b699a72ac3bdede71af1b518f -Author: Keith Packard -Date: Sun Jul 26 14:22:35 2009 -0700 +commit 8939ad2b2aa0385f072d3e1169eaf99289ed737a +Author: Peter Hutterer +Date: Thu Sep 3 09:44:11 2009 +1000 - Bump to version 1.6.2.901 (1.6.3 RC1) + Xi: return error values to client from XIWarpPointer. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 87900645c22d6066fe70190ab838a687808db5c3 -Author: Dave Airlie -Date: Thu Jun 11 14:21:53 2009 +1000 +commit d481720c57f558c47273bf89854fc69254319772 +Author: Peter Hutterer +Date: Wed Sep 2 15:28:28 2009 +1000 - xfree86: move didLock assignment down to where the function pointer is valid. + Xi: standardise XI2 headers. - crtc->funcs->lock is NULL, so it's no use calling it here. Move it down so - it's actually defined before we use it. + Adding missing dix-config.h include, adding one missing header guard. - Introduced with 6f59a8160042ea145514fdcb410f17f33fd437c2. + Signed-off-by: Peter Hutterer + +commit 377a5f655ca88cd836da24ab42361df9f3ff51a2 +Author: Peter Hutterer +Date: Wed Sep 2 11:16:15 2009 +1000 + + Xi: if XISetEventMask fails, return this to the client. - Tested-by: Peter Hutterer - (cherry picked from commit 0de58c88aba7ddd69b04f24ab5b2967c359aa69e) + The only failure point can be a BadAlloc. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 72c9bc64ff86999a4643663e5699a6547a9e2997 -Author: Benjamin Defnet -Date: Wed Jun 10 21:35:48 2009 -0700 +commit 993ca5acbdccef9e43cd6580e7cc22cfda1a2578 +Author: Peter Hutterer +Date: Wed Sep 2 10:53:40 2009 +1000 - randr: fix operation order so that rotation+transform works + dix: when unsetting a cursor, update the sprite immediately (#23608) - The matrix multiply to combine rotation and projective transforms was being - done in the wrong order. + Removing the device cursor while the cursor was within the window did not + update the visible sprite until the next enter/leave event. - Signed-off-by: Keith Packard - (cherry picked from commit b2bf67b61c564a4b92a429ca9ad455403161f33a) + X.Org Bug 23608 - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 7f95d18397757facafcba39e4613b2ed5047c5c4 -Author: Benjamin Defnet -Date: Mon Jun 8 21:45:42 2009 -0700 +commit b41db8fe460fc0956922b79261752d1bb104684f +Author: Peter Hutterer +Date: Wed Sep 2 08:57:40 2009 +1000 - hw/xf86/modes: Set crtc mode/rotation/transform before calling set_mode_major + Xi: fix broken swap code in XISelectEvents request processing. - This moves code out of each implementation of set_mode_major and back into - the X server. The real feature here is that the transform is now available - in the crtc for use by either xf86CrtcRotate or whatever the driver wants to - do. Without this change, the transform was lost for drivers providing the - set_mode_major interface. + The pointer advanced 12 bytes too short. Rather unfortunate if both the code + and the test have the same bug. - Note that users of this API will want to stop smashing the transformPresent - field, and could also stop setting mode/x/y/rotation for new enough X servers, - but there's no reason to make that change as it will break things when - running against older X servers. + Signed-off-by: Peter Hutterer + +commit 727de7c90de4198222e0dc58d7503b4a1672f642 +Author: Peter Hutterer +Date: Tue Sep 1 16:33:56 2009 +1000 + + Xi: Unify checking for invalid bits in grab masks. - Signed-off-by: Keith Packard - Acked-by: Daniel Stone - (cherry picked from commit 6f59a8160042ea145514fdcb410f17f33fd437c2) + Bits above XI2LASTEVENT are invalid and cause in BadValues. These checks + must be performed anywhere where a mask_len parameter is given. - Signed-off-by: Keith Packard + This patch also adds the missing checks to grab masks. + + Signed-off-by: Peter Hutterer -commit abc89e2c48a88c29f49fe44d03778deab4b1a3b9 -Author: Matthias Hopf -Date: Tue Mar 10 15:49:49 2009 +0100 +commit 59a6d7d478903a8bc9c5d4cc8b2e62e2ad102dba +Author: Peter Hutterer +Date: Tue Sep 1 16:30:52 2009 +1000 - randr: Nuke broken set_origin shortcut + Xi: don't overrun memory for grab masks. - Shortcut is impossible to implement this way, because we don't know for sure - whether the crtc of an output has changed or not. - (cherry picked from commit cadf65a6e190a8952ad3cc216dc9ea55241de91a) + A grab mask provided in the request may be larger than the one used in the + server. Cut down to size before memcopying. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit b1eed8970f7c5b07152dbc6b7dcc4c8504aeb67c -Author: Julien Cristau -Date: Fri May 22 09:54:38 2009 +0200 +commit 83db2b126e8623824e1303e74070375994984599 +Author: Peter Hutterer +Date: Tue Sep 1 15:58:02 2009 +1000 - randr: fix server crash in RRGetScreenInfo + Xi: don't try to set oversized or non-existing masks. - We don't return rates to randr < 1.1 clients, so don't allocate space - for them. This fixes a FatalError due to not all allocated space being - used. + Fixes crash if the first XISelectEvents has a zero sized event mask. + Fixes crash if the mask provided is larger than others->xi2mask[]. - X.Org bug#21861 + Signed-off-by: Peter Hutterer + +commit 6fe235d3635aaf995dfa198c60ded1a43fe9dd45 +Author: Peter Hutterer +Date: Tue Sep 1 15:46:34 2009 +1000 + + test: fix request length calculation, add length tests for XISelectEvents - Reported-by: Guillaume Quintin - Signed-off-by: Julien Cristau - (cherry picked from commit 12e725d08b4cf7dbb7f09b9ec09fa1b621156ea9) + Signed-off-by: Peter Hutterer + +commit 7ab1b50c545ae1a62384bedee60168e227bb2a94 +Author: Peter Hutterer +Date: Tue Sep 1 15:27:29 2009 +1000 + + test: fix build after changing libxf86config.la - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 1addf6fe235f8537024140aff56dffb4738701a7 -Author: Keith Packard -Date: Wed Aug 6 15:26:24 2008 -0700 +commit 58c298acc1045927e0d90be73b8dbc8837252589 +Author: Peter Hutterer +Date: Tue Sep 1 15:16:17 2009 +1000 - Replace dixLookupResource by dixLookupResourceBy{Type,Class} + Xi: extra length checking for requests providing masks. - dixLookupResource attempted to automatically detect whether the caller - wanted a lookup by-type or by-class, unfortunately, it guessed wrong for - RT_NONE. Instead of trying to make the guess better, this patch just reverts - the unification and creates separate functions for each operation. - (cherry picked from commit f8dd80d13bb5313a11b38b280f8ad3e22f0a6300) + masks can be of arbitrary length. If the client did not initialize mask_len, + some sort of boundary check is needed to avoid running over memory. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit c838a03e891a6fcd2f7b71f727c1b93f7c26e7a7 -Author: Robert Noland -Date: Wed Apr 15 12:06:19 2009 -0500 +commit 0e4dd3b2d28d3dbbfc152d6f5030901ec063a7ae +Author: Peter Hutterer +Date: Mon Aug 31 09:50:36 2009 +1000 - One = is more than adequate here. Make is sh safe. - (cherry picked from commit b3e3154cce47add97f5561088036ce3b9e7dc937) + Un-export ApplyPointerMapping. This is an in-server function. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 68cb38bed65c49716104eeb345c4b1b85a4b15bc -Author: Rémi Cardona -Date: Fri Jul 3 10:51:50 2009 +0200 +commit a470e8426ee290046a927a17fc969235c4b1e2e7 +Author: Eamon Walsh +Date: Wed Sep 2 17:56:52 2009 -0400 - configure: libXinerama isn't needed anymore - - since libXinerama commit 90d4d23bf2e94721149ddc0a80093b10a82e8845 and - xineramaproto commit 21477147613c28c968b5e1eb9d8aea7017dd399d, the - server no longer needs libXinerama. + Correct outdated e-mail address in "Author" statements. - Signed-off-by: Rémi Cardona - (cherry picked from commit 2c69deb92e11542f615df0f24fdc03e3b4415475) + Signed-off-by: Eamon Walsh + +commit 17545ed80ee3bb6a058a2748fc7bfb0fc5139fc1 +Author: Adam Jackson +Date: Wed Sep 2 14:40:07 2009 -0400 + + randr: Fix crtcs using set_mode_major() - Signed-off-by: Keith Packard + We'd never mark the crtc as active, meaning (among other things) gamma + upload wouldn't work. -commit ec60f4ca65dbc25f9a40636f5280b1a1ec68407f -Author: Michel Dänzer -Date: Fri Jul 10 20:06:02 2009 +0200 +commit 291408980f33b1e541c89d958535e6fad55fdac9 +Author: Dave Airlie +Date: Wed Sep 2 20:14:30 2009 +1000 - EXA: Only pass CT_YXBANDED to RECTS_TO_REGION() if that is really true. + vgaarb: protect fini as well just in case + +commit 8aeadbff0840efc0c9a073e60d6fec0dba313060 +Author: Dave Airlie +Date: Wed Sep 2 19:07:40 2009 +1000 + + vgaarb: if arb init fails, make sure locking doesn't occur. - Fixes http://bugs.freedesktop.org/show_bug.cgi?id=22642 . - (cherry picked from commit 35758544813f156eaac28844e693b2a28f6de316) + pointed out by Martin Jansa on irc. - Signed-off-by: Keith Packard + Signed-off-by: Dave Airlie -commit b10c4fe343d2339bae4736793939e44a6befaf88 -Author: Kim Woelders -Date: Sat Jul 11 18:55:17 2009 +0200 +commit 0e46066efea4e385b2874d4a62a786957e572fe5 +Author: Jerome Glisse +Date: Tue Sep 1 13:59:47 2009 +0200 - Fix key repeat problem. + Xext: fix a typo for bigreqsproto.h header file - Signed-off-by: Kim Woelders + Signed-off-by: Jerome Glisse -commit 606f6dba16d42e3546a82a386d5a01087467b511 -Author: Adam Jackson -Date: Tue Apr 14 10:54:25 2009 -0400 +commit c61d8ce1507318476b75b81bf871da19cb1bf747 +Author: Jeremy Huddleston +Date: Mon Aug 31 17:57:50 2009 -0700 - xdmcp: Don't crash on X -query with more than 255 IP addresses. (#20675) - - You could be more clever than this, but the wire protocol says this - really is an array of not more than 255 ARRAY8, so it's not just a - matter of changing the types. + Xext: Require newer versions of xcmiscproto, bigreqsproto, and xf86bigfontproto for new *proto.h header file names - (cherry picked from commit 0eb19f9437b7d8c19592e49eedb028771d300d80) + bigreqsproto >= 1.1.0 + xcmiscproto >= 1.2.0 + xf86bigfontproto >= 1.2.0 -commit 396d3a7762abd0dd84042833b75f2ebf9d100bb0 -Author: Alan Coopersmith -Date: Wed Jul 8 11:26:46 2009 -0700 +commit 1c37be80982151a5acd74fe4b960695f604f22ca +Author: Michael Witrant +Date: Tue Sep 1 10:46:50 2009 +1000 - Fix build of drivers with 1.6.2 when not using --install-libxf86config + XFree86: Linux: Fix 100% CPU usage with ShareVTs and kbd - c859b736d1d23c5dc2f53958b1e76660e6d45018 removed duplicate entries for - these from the Makefile, but removed a different set than was done in - master branch, causing xf86Parser.h to not be installed, which in turn - breaks the build of drivers like -ati, -intel & -nv that use xf86Modes.h, - which includes xf86Parser.h + Leave consoleFd open over the course of the server, even though any use + of it in this context is likely to be disastrous. - Signed-off-by: Alan Coopersmith - Signed-off-by: Keith Packard + Signed-off-by: Michael Witrant + Reviewed-by: Daniel Stone -commit 6f1aff5a2b45bc2985081abc240a8fed37170386 -Author: Keith Packard -Date: Mon Jul 6 14:08:31 2009 -0700 +commit 65183dc3153232943dff3cfbaacab141e7ed0e9c +Author: Daniel Stone +Date: Sun Jul 5 19:53:55 2009 +0300 - Bump to version 1.6.2 + Input: Mark Xi input events as critical - Signed-off-by: Keith Packard + Note that the Xi events are critical and should thus cause a flush to + the client when an input event is pending. + + Signed-off-by: Daniel Stone -commit 499f9f62e408510ed4efb6f2321a3f851f535567 -Author: Keith Packard -Date: Wed Jul 1 14:01:57 2009 -0700 +commit a4e614d301cfa2577de1ee6a15abf023590586d1 +Author: Daniel Stone +Date: Sat Apr 25 16:46:10 2009 +1000 - dri2: Preserve compatibility with 1.6 DRI2 API/ABI + Xi: Fix harmless ButtonPress/ButtonRelease confusion - The old DRI2 buffer allocation API wasn't great, but there's no reason to - make the server stop working with those drivers. This patch has the - X server adapting to the API provided by the driver, using the new API where - available and falling back to the old API as necessary. A warning will be - placed in the log file when the old API is in use. - (cherry picked from commit 2e2c5b216cc1c7a9bc26bd2c68226aaed5fc52ca) + A missing break meant that ButtonPress would fall through into + ButtonRelease, but luckily it appears to have been completely harmless. - Signed-off-by: Keith Packard + Signed-off-by: Daniel Stone -commit c859b736d1d23c5dc2f53958b1e76660e6d45018 -Author: Arkadiusz Miskiewicz -Date: Fri Jul 3 11:03:32 2009 +0200 +commit 5ef53a94ce4e48e11de26290cd677266308640c8 +Author: Matt Turner +Date: Mon Aug 31 20:23:26 2009 -0400 - xfree86: remove duplicate header entries in Makefile.am + alpha: kill xf86SlowBCopyToBus and xf86SlowBCopyFromBus - fixes https://bugs.freedesktop.org/show_bug.cgi?id=22547 + xf86SlowBCopyToBus and xf86SlowBCopyFromBus cause segfaults on my + system. - This bug was fixed in a big clean-up commit in master, which cannot be - backported (see commit b1dac41fb3853ca8182048ea57b88b6e84ecceb3) + Also remove associated slowbcopy_tobus/slowbcopy_frombus macros. - Signed-off-by: Rémi Cardona - Signed-off-by: Keith Packard + Signed-off-by: Matt Turner -commit 6f82ca0f8bfc074474407cb5ffe782c3c128f838 -Author: Alan Hourihane -Date: Mon Feb 16 11:39:34 2009 +0000 +commit a8e634b86d47353ef354717904fa66ebe35e1303 +Author: Jeremy Huddleston +Date: Mon Aug 31 15:53:58 2009 -0700 - dri2: support glXWaitGL & glXWaitX by copying fake front to front and - vice-versa. - (cherry picked from commit 2a8b8077d8f6001eb57deba60e1009fc99c28668) - - Signed-off-by: Keith Packard + XQuartz: GLX: Drawable does not contain resize anymore. + (cherry picked from commit 1e642f22fbf029bb3917091cb5fa2f78b4a92f62) -commit d478bc948838d11b3abcf6d2cdd3a00dccb344c4 -Author: Eric Anholt -Date: Tue Feb 17 09:59:17 2009 -0800 +commit 1b30545c04a51bfa3ff95a26d64962907a62ff15 +Author: Matt Turner +Date: Mon Aug 31 16:16:02 2009 -0400 - xinerama: Put the proto version in the code instead using proto headers. + Make sys.c use compiler.h unaligned access functions - Proto headers updating resulting in the server advertising new versions is - broken. This should be applied to every extension. - - This fixes the build against slightly-older xineramaproto. - (cherry picked from commit b349a764e98f0d8f221190157ffa0904b91beca5) + Signed-off-by: Matt Turner + +commit 9522ab9a98e98bc394703038e34b1c12aa39a179 +Author: Matt Turner +Date: Mon Aug 31 16:12:29 2009 -0400 + + Remove unused Delay.c - Signed-off-by: Keith Packard + Signed-off-by: Matt Turner -commit a7a93c12f91e0de72868f17a555215f1795d9e2f -Author: Keith Packard -Date: Mon Jun 29 23:18:55 2009 -0700 +commit 4fac13ff8848bc38654b5567170fbc0d7b2c96d6 +Author: Matt Turner +Date: Mon Aug 31 16:11:59 2009 -0400 - Build against DRI_TEX_BUFFER_VERSION 1 + Remove unused BUSmemcpy.c - setTexBuffer2 isn't present in this version of the structure, so don't try - to call it. + BUSmemcpy.c provides xf86BusToMem and xf86MemToBus, which are are memcpy + wrappers written to avoid glibc's memcpy on Alpha. glibc'c memcpy on + Alpha has improved much since this was written, so it's no longer + needed. Neither function is used inside the xserver, and no module on + my machine uses either as well. - Signed-off-by: Keith Packard + Signed-off-by: Matt Turner -commit 5b49aff4f8b5e4d15bc883f74509d93324f09f5f -Author: Keith Packard -Date: Mon Jun 29 14:40:46 2009 -0700 +commit b65146fc8418b53b7e43dd34cc20000369475671 +Author: Matt Turner +Date: Mon Aug 31 16:11:10 2009 -0400 - Bump version to 1.6.1.902 + Remove unused ia64_flush_cache function - Signed-off-by: Keith Packard + Signed-off-by: Matt Turner -commit e97cabce4abdaf6b83d1a442f7972d4c45979b33 -Author: Peter Hutterer -Date: Wed Mar 25 12:55:42 2009 +1000 +commit f201705e29bdc3309dd611b461ec67740cb706ef +Author: Matt Turner +Date: Mon Aug 31 16:09:33 2009 -0400 - xfree86: fix SWCursor check in xf86CursorSetCursor. + Simplify unaligned access code - Wrong check for inputInfo.pointer resulted in a SW cursor being rendered when - the pointer left the screen (in a Xinerama setup). - We must call the sprite rendering function if - - SW cursors are enabled, or - - The current device is not the VCP and not attached to the VCP. + All architectures should be able to use the same unaligned access code, + regardless of whether they need special unaligned access instructions. + Let's let gcc do the heavy lifting. - Backported from commit 66089e9129a821cfb1983d3d35f41b975a52de5e for - server-1.6-branch by Peter Hutterer + In the case that we're not using a gcc-compatible compiler, use memmove. - Reported-by: Gordon Yuan - Signed-off-by: Peter Hutterer - Signed-off-by: Keith Packard + The xserver already requires pixman, so include pixman.h for its uint*_t + types. + + Signed-off-by: Matt Turner -commit 17c5b2caf01deec1d1be9fc2ac68a3d636e64c44 -Author: Julien Cristau -Date: Mon May 18 19:01:46 2009 +0200 +commit 3718beb69cc5b4cdb79dd97242fce9231a26a39a +Author: Matt Turner +Date: Mon Aug 31 16:08:52 2009 -0400 - Change default for ExaOptimizeMigration to false + Clean up unaligned access functions on alpha - Quoting Michel Dänzer in <1234862541.4823.541.camel@thor>: + Checks for __GNUC__ are superfluous since the only other compiler for + the platform is Compaq C, and it doesn't support GCC style inline + assembly. - Unfortunately, there are still bugs left in EXA which prevent it from - working 100% correctly with the option enabled, see + Signed-off-by: Matt Turner + +commit e1fdd7671346ce374a2b77fe94b8c1a25c0520da +Author: Matt Turner +Date: Mon Aug 31 16:08:20 2009 -0400 + + Move unaligned access functions to common section - http://bugs.freedesktop.org/show_bug.cgi?id=16416 + Signed-off-by: Matt Turner + +commit 628b863bcc6a2a14fd3c0e6c2be4394833cc314c +Author: Matt Turner +Date: Mon Aug 31 15:53:41 2009 -0400 + + Remove unused arm_flush_cache function - http://bugs.freedesktop.org/show_bug.cgi?id=19940 + Signed-off-by: Matt Turner + +commit a544139196ac7f5bc89bc642c6c5c0ea1b0bb60e +Author: Matt Turner +Date: Mon Aug 31 15:53:19 2009 -0400 + + Remove unused ppc_flush_icache function - So disable it for now to avoid corruption. + Signed-off-by: Matt Turner + +commit 52aa0495218dc7821a23b045b2c74b2a66e34616 +Author: Matt Turner +Date: Mon Aug 31 15:52:39 2009 -0400 + + Replace 8 nops with proper sync instruction on mips - Signed-off-by: Keith Packard + Cc: Ralf Baechle + Acked-by: David Daney + Signed-off-by: Matt Turner -commit 8d65439d5c950ea01ec8e1e4dd989aff0fb4c3f0 -Author: Keith Packard -Date: Thu May 28 14:43:27 2009 -0700 +commit 563fa1c5d762173cd6fb78f9e33cb960ef3153bb +Author: Matt Turner +Date: Mon Aug 31 15:52:10 2009 -0400 - Make RANDR 'set' timestamps follow client specified time. Bug 21987. + Use sfence and mfence instructions on amd64 - The lastSetTime value which indicates when the configuration within the - server was last changed was not getting set in the appropriate RandR - requests. + Signed-off-by: Matt Turner + +commit 430c4af0d6e2cad03ca5f49ec6e87f3717d0d9e2 +Author: Matt Turner +Date: Mon Aug 31 15:51:41 2009 -0400 + + Add x86 barrier macros - Signed-off-by: Keith Packard - (cherry picked from commit 69a9545d1f8110841538410818df19fd960412c5) + Signed-off-by: Matt Turner + +commit 0d87f77a3ff4b22729220b7edcb98c6f06d0787c +Author: Matt Turner +Date: Mon Aug 31 15:51:05 2009 -0400 + + define barrier macros as nops if not otherwise defined - Signed-off-by: Keith Packard + Signed-off-by: Matt Turner -commit d98f2792d2f523d7397eac70cc021761e9561219 -Author: Federico Mena Quintero -Date: Tue Jun 9 13:28:28 2009 -0700 +commit 1b43ee13b30d9e3f828e94cdf686d103ce1cac07 +Author: Matt Turner +Date: Mon Aug 31 15:53:58 2009 -0400 - dix/randr: Add missing fields to SRR*NotifyEvent() + Move memory barrier macros into common section - Also, remove redundant field swaps and make others match the order in which - they are declared in the xRR*NotifyEvent structs. + alphabetize by architecture also. - Signed-off-by: Federico Mena Quintero - Signed-off-by: Keith Packard - (cherry picked from commit e244a5991e2cc55f5aa2f6e5255f1dabf56f0235) + Signed-off-by: Matt Turner + +commit 295e598d9a83ff7162a093810ca9f546d00c0968 +Author: Tiago Vignatti +Date: Mon Aug 31 20:04:51 2009 +0300 + + xfree86: remove _more_ RAC junk - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti -commit f697b6c582095aa1eec3244fa05835922249425b -Author: Peter Hutterer -Date: Mon May 11 12:53:56 2009 +1000 +commit 38627b2eaf64aa8fe694fc3e1f5e4fee0c9c11c3 +Author: Tiago Vignatti +Date: Mon Aug 31 19:36:02 2009 +0300 - kdrive: set Activate/Deactivate grab for input devices (#21591) + xfree86: shut up vgaarb warnings when server doesn't support it - X.Org Bug 21591 - (cherry picked from commit b1b5ec45c1cb650ccb8c659218f9481379c777d9) + Signed-off-by: Tiago Vignatti + +commit 79182538b43ecd35bf2b47eccfb2dd3bbcdc12eb +Author: Tiago Vignatti +Date: Mon Aug 31 19:25:13 2009 +0300 + + xfree86: removal of some dead code due VGA arbiter's inclusion - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti + +commit 6f3cc08ec01609c7efe424ac3e30fe4ff9391007 +Author: Jeremy Huddleston +Date: Thu Aug 27 16:21:11 2009 -0700 + + XQuartz: GL: Unset GL_EXT_gpu_program_parameters for Tiger/ppc + + See http://trac.macports.org/ticket/20638 + (cherry picked from commit 0f3a89d306838b3c75a73cd1e9e2928737222b70) -commit c91accd173b5f17af771bd9f4e60b5860af15718 +commit fdb29ebeed143fa05630966b847b05399a446ddc +Author: Jeremy Huddleston +Date: Thu Aug 27 11:04:00 2009 -0700 + + XQuartz: GL: Explicitly set GL_EXT symbols to 0 if they are not in OpenGL.framework to prevent X11's glext.h from setting them to 1. + (cherry picked from commit 7fe37137d826d1b698e87a5b35050dd02f0a4d4b) + +commit 3be80bd98357cf362aa9a004d1292e5167d83a17 Author: Peter Hutterer -Date: Mon May 11 12:51:40 2009 +1000 +Date: Fri Aug 28 16:20:59 2009 +1000 - dix: ensure Activate/DeactivateGrab has a valid value. + Xi: silence compiler warning "type may be used uninitialized" - Xephyr doesn't manually set Activate/DeactivateGrab for new devices, - resulting in a NULL-pointer dereference later when a grab is activated. - Avoid the segfault by ensuring that the pointer is always valid. + GrabKey and GrabButton are only called from XI/XI2 code. Set type to -1, + just in case. Signed-off-by: Peter Hutterer - (cherry picked from commit 6f9e22049862ee9ac7f604411d005d8bb1b2dd1c) - - Signed-off-by: Keith Packard -commit bd94c7c986aed4496cd4e8141bb30c3e4412f1f4 -Author: Alan Coopersmith -Date: Tue May 12 16:11:01 2009 -0700 +commit 24716b9254fa7d609792596723a192bb044a7d3f +Author: Eamon Walsh +Date: Fri Aug 28 22:52:48 2009 -0400 - Resync COPYING file with notices in code base as of xorg-server-1.6.1 + xace: fix up access modes in dixLookupDrawable calls from dri2. - Remove notices for code no longer in tree (Xprint, Xgl, kdrive, cfb, etc.) - Add/update notices for new/changed code in tree + Referencing a screen through a drawable only requires GetAttr access. + Treat dri2 drawables as child windows (Add/Remove access). + Treat getting buffers as intent to read/write the drawable. - Signed-off-by: Alan Coopersmith - (cherry picked from commit b746a00cffca5c553b607a8e9c1074294a23b443) + Signed-off-by: Eamon Walsh + +commit 0f3a64c3a5bce6cf3f00fe42cc7d1d3e69822945 +Author: Eamon Walsh +Date: Thu Aug 27 15:11:12 2009 -0400 + + Add DRI2 requests to protocol.txt - Signed-off-by: Keith Packard + Signed-off-by: Eamon Walsh -commit 597747c6551cd67487069415297e5eb441038321 -Author: Jeremy Huddleston -Date: Sat May 9 11:42:17 2009 -0700 +commit 3d17c4f6e1538986fe7b1f17614d76e28f2d1e11 +Author: Adam Jackson +Date: Fri Aug 28 12:12:52 2009 -0400 - Fix a couple off-by-one array boundary checks. + EDID: Vendor detailed blocks aren't worth X_WARNING about + +commit 2c1a845521a31f08938b0f8d6e38835ce127c94a +Author: Adam Jackson +Date: Fri Aug 28 12:09:57 2009 -0400 + + EDID: Print 1152x864 in established timings, not x870 - Error: Write outside array bounds at Xext/geext.c:406 - in function 'GEWindowSetMask' [Symbolic analysis] - In array dereference of cli->nextSib[extension] with index 'extension' - Array size is 128 elements (of 4 bytes each), index <= 128 + The spec says x870, but we actually use x864 because that's a real DMT + mode and x870 isn't. This might or might not be wrong, but we should at + least tell the truth. + +commit 2a806d7fa372e28d039761c9b4087cf812e8e46b +Author: Tiago Vignatti +Date: Fri Aug 28 16:05:04 2009 +0300 + + os: remove unused -cursor option - Error: Buffer overflow at dix/events.c:592 - in function 'SetMaskForEvent' [Symbolic analysis] - In array dereference of filters[deviceid] with index 'deviceid' - Array size is 20 elements (of 512 bytes each), index >= 0 and index <= 20 + Signed-off-by: Tiago Vignatti + +commit 9040dab76182d1a019ca7fef7b29733d2c199e61 +Author: Tiago Vignatti +Date: Fri Aug 28 15:50:50 2009 +0300 + + xfixes: minor clean ups on createInvisibleCursor() - Error: Read buffer overflow at hw/xfree86/loader/loader.c:226 - in function 'LoaderOpen' [Symbolic analysis] - In array dereference of refCount[new_handle] with index 'new_handle' - Array size is 256 elements (of 4 bytes each), index >= 1 and index <= 256 + - unused return value + - no reason to declare static variable given the function is just called + once + - no reason to declare different type and cast it after. - These bugs were found using the Parfait source code analysis tool. - For more information see http://research.sun.com/projects/parfait + Signed-off-by: Tiago Vignatti + +commit 80ed8096f9b8e974b556a6c9f3b600fb71b994c3 +Author: Tiago Vignatti +Date: Fri Aug 28 15:33:19 2009 +0300 + + render: delete unused headers declaration - Signed-off-by: Alan Coopersmith - Signed-off-by: Adam Jackson - Acked-by: Peter Hutterer - (cherry picked from commit b680bda34da130ce408783f04214771471e41e8d) - (cherry picked from commit 04c9e80f083659e63cffec8969fb3a0cfc551a97) + Signed-off-by: Tiago Vignatti + +commit f959b1e5485b93700c7da449a98182f5ce067ca2 +Author: Tiago Vignatti +Date: Fri Aug 28 15:30:21 2009 +0300 + + render: AnimCurInit and AnimCursorCreate shouldn't be _X_EXPORT - Signed-off-by: Keith Packard + Pointed by Peter Hutterer on xorg-devel ml. + + Signed-off-by: Tiago Vignatti -commit 10c06ddeefccc195e70adfed664e9488eeb53804 -Author: Peter Hutterer -Date: Thu Apr 9 15:40:24 2009 +1000 +commit e454f106dc65ecfacc154a1fa0810935022a8fee +Author: Dave Airlie +Date: Fri Aug 28 11:39:42 2009 +1000 - xfree86: restore default off for DontZap + xf86 ddx: add vga arbiter support. - Zapping is triggered by xkb these days, so note in the man page that it's the - Terminate_Server action. Since it's XKB, personal preferences towards or - against zapping should be achieved through xkb rulesets. - If Terminate_Server is not in the xkb actions, then we can't zap anyway and we - don't need a default of DontZap "on". + This adds support for using the libpciaccess interface for + vga arbitration support on top of a kernel which supports it. - This patch restores the old meaning of DontZap - disallow zapping altogether, - regardless of XKB's current keymap. - Ideally, this patch should be accompanied by b0f64bdab00db652e in - xkeyboard-config. + Currently patches are queued for kernel 2.6.32 in jbarnes + pci tree, and shipping in Fedora kernel. - Signed-off-by: Peter Hutterer - (cherry picked from commit 737b49199a05299486064e6e762cf2a2f6f95be6) + Co-authors: + Tiago Vignatti - Signed-off-by: Keith Packard + Signed-off-by: Dave Airlie -commit 966349f8e068fa120470549b0db9a8fcf88aca5b -Author: Michel Dänzer -Date: Fri May 15 15:48:37 2009 +0200 +commit 6fffcd5825454a7fe58ffbcfb219f007cf38e731 +Author: Peter Hutterer +Date: Thu Aug 27 13:43:15 2009 +1000 - EXA: Always damage glyph cache pixmap manually after uploading a glyph. + xfree86: require xf86dgaproto 2.0.99.1. - Signed-off-by: Michel Dänzer - (cherry picked from commit 7c8327f0a75087a85864256a9cea80dd4b86def5) + Signed-off-by: Peter Hutterer + +commit 2fba2eac0b4c8d07bdf7bea20ef75ff579621728 +Author: Peter Hutterer +Date: Mon Aug 24 11:23:23 2009 +1000 + + test: add a few tests for xtest device initialization. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit cc5e4b632aa99623ff577776039287c5950d921d -Author: Michel Dänzer -Date: Fri May 15 15:48:37 2009 +0200 +commit 903c3db1d1685bd855dceed9e7b92890743663e1 +Author: Peter Hutterer +Date: Mon Aug 24 09:48:00 2009 +1000 - EXA: Take GC client clip type into account for migration. + Xext: rename Xtst* to XTest* - Fixes http://bugs.freedesktop.org/show_bug.cgi?id=18950 . + This patch corrects a misnaming of XTest-related functions. - Signed-off-by: Michel Dänzer - (cherry picked from commit 850675d4de4373e5df95507dbf2cd9affaaf54bc) + The extension itself announces itself as XTEST. Xtst is the library name + itself, but all library functions are prefixed by XTest. Same with the + naming in the server. - Signed-off-by: Keith Packard + - Rename all *Xtst* functions to *XTest* for consistency with the library + and in-server API. + - Rename the "Xtst device" property to "XTEST device" for consistency with + the extension naming. + - Rename the device naming to " XTEST device". The + default xtest devices become "Virtual core XTEST pointer" and "Virtual + core XTEST keyboard". + + Signed-off-by: Peter Hutterer -commit bbe0be1e459b0d5810a0e7b916cc890c440a4e21 -Author: Tomas Janousek -Date: Wed May 20 15:03:01 2009 +0200 +commit 8bfd23e144e51401e3756de9260a4811fcc59e91 +Author: Peter Hutterer +Date: Mon Aug 24 09:38:15 2009 +1000 - Bug #6428, #16458, #21464: Fix crash due to uninitialized VModMap fields. - - In ProcXkbGetKbdByName, mrep.firstVModMapKey, .nVModMapKeys and - .totalVModMapKeys were not initialized, contained random values and caused - accesses to unallocated and later modified memory, causing - XkbSizeVirtualModMap and XkbWriteVirtualModMap to see different number of - nonzero values, resulting in writes past the end of an array in XkbSendMap. + input: move XTest device initialization into Xext/xtest.c - This patch initializes those values sensibly and reverts commits 5c0a2088 and - 6dd4fc46, which have been plain non-sense. + XTest devices are non-optional but nonetheless specific to the XTEST + extension. - Signed-off-by: Tomas Janousek Signed-off-by: Peter Hutterer - (cherry picked from commit 525aa17f804d37d1cfcbbf6b8e6cddb45e999b20) - - Signed-off-by: Keith Packard -commit 6056d429e7dcc37fe706dd7bc8d3429fdb635e14 -Author: Kristian Høgsberg -Date: Wed Apr 1 17:42:33 2009 -0400 +commit a95f80fa914678d360f6dfd2b58926193df6ea4a +Author: Peter Hutterer +Date: Mon Aug 24 09:26:56 2009 +1000 - Support setTexBuffer2 in AIGLX. + dix: use IsXtstDevice instead of the direct key lookup. - Fixes broken GLX_tfp, specifically, lets compositors ignore un-defined - alpha channel for pixmaps. - (cherry picked from commit 91b697efdefba125348dbcaf584ee51a7f8c9bf6) - - Signed-off-by: Keith Packard - -commit 7cae08d21952b02ef664641af6588ef6bd727bf8 -Author: Adam Jackson -Date: Mon Jun 29 11:16:16 2009 -0400 - - EDID: Fix timing class names to match the spec - (cherry picked from commit 5c1afac5eea1d8327c74342d12d082b75f0cebde) - -commit dbac41b624e4aa86a6a184b7ebb52bfdd367bbf0 -Author: Adam Jackson -Date: Fri Jun 19 12:42:07 2009 -0400 - - pci: Dump vendor/devices ids in the printed device list - (cherry picked from commit eb35402d0a5290e8a73d7d1e92f173294c364cc2) + Signed-off-by: Peter Hutterer -commit 6be19e8f43086fb4b7fb30a47b89b5f3eed798ef -Author: Ian Romanick -Date: Wed Apr 8 14:54:30 2009 -0700 +commit 16b7ebd7d876034edfe8f74562bd06e747879d79 +Author: Peter Hutterer +Date: Mon Aug 24 14:34:23 2009 +1000 - Use a #define instead of a magic number + xfree86: Remove xf86GetMotionEvents from public API. - The number of buffers is likely to change in the future, so having - this as a define is the right way to go. + This function was used as the default motion event queue API until + including XINPUT_ABI 2 (server 1.5). - Signed-off-by: Ian Romanick - (cherry picked from commit 03aebed519986c4dd03e02b3b3d4af1f64595ca7) + This API was broken with 1883485 in May 2008 (wrong casting of parameters) + and isn't in use by input drivers past ABI 3. + + Signed-off-by: Peter Hutterer + Signed-off-by: Tiago Vignatti -commit 540d5b87a4e24d85ec46620dfedd7bd7979180ea -Author: Jerome Glisse -Date: Mon May 11 22:52:46 2009 +0200 +commit af76bddd7a47092e2f0f85b2547c1cc74eb5cebc +Author: Jeremy Huddleston +Date: Sun Aug 9 16:13:32 2009 -0700 - DRI2: update DRI2 private drawable width & height according to X drawable - (cherry picked from commit f250eea2e90fc50bec5214c2f41132b95edc2c46) + XQuartz: Check NSINTEGER_DEFINED to make sure the NSInteger and NSUInteger types are defined. + (cherry picked from commit 36e02a8649c0a545084cce5fb2c5717277b9273c) -commit ec9f1ae32474bc0507a3c66e63bdf2835d467a34 -Author: Ian Romanick -Date: Mon Apr 27 15:11:10 2009 -0700 +commit 07c935b72fb14cb12f1cb5afbf8cc81f97086bac +Author: Jeremy Huddleston +Date: Tue Aug 25 20:11:38 2009 -0700 - DRI2: Force allocation of real-front buffer for non-windows as well - - For redirected rendering we end up with pixmaps (which the app thinks are - windows) that are double buffered. + XQuartz: Use applewmproto 1.4 updated headers. + (cherry picked from commit 203df06c0eb2dcd5adfa788a1ba9569650c15f9c) + +commit 0f613007cd43545eda40825bd74ada3e6e5f36f6 +Author: Jeremy Huddleston +Date: Wed Aug 5 17:43:29 2009 -0700 + + CheckWindowOptionalNeed: Ensure w->optional is set to avoid SEGFAULT - Signed-off-by: Ian Romanick - Tested-by: Pierre Willenbrock - (cherry picked from commit 0d9d3f3e361f769822caedccf4c2a58cc9930ecc) + If CheckWindowOptionalNeed is called consecutively (and the first removes optional), then + we will SEGFAULT. This can happen in XQuartz because ReparentWindow will call + pScreen->ReparentWindow which can DeleteProperty which will CheckWindowOptionalNeed... then + ReparentWindow will call it again later indiscriminantly. + (cherry picked from commit b608c864ccc59628bd33c033393121b61580460e) -commit 4fad615d689c61c6c3a000295a1fa755359737cb -Author: Ian Romanick -Date: Fri Apr 24 12:49:19 2009 -0700 +commit e7dd1efef408effe52d0bd3d3aa0b5d4ee10ed90 +Author: Keith Packard +Date: Tue Aug 25 18:07:00 2009 -0700 - DRI2: Implement protocol for DRI2GetBuffersWithFormat + Ensure that rotation updates happen frequently - This change implements the protocol for DRI2GetBuffersWithFormat, but - the bulk of the differences are the changes to the extension / driver - interface to make this function work. The old CreateBuffers and - DeleteBuffers routines are replaced with CreateBuffer and DeleteBuffer - (both singular). + The smart scheduler is designed to minimize scheduler overhead by + increasing the interval between WaitForSomething calls when a single + client is running. However, the software rotation code depends on + its BlockHandler being invoked for screen updates; the long delays + caused by the smart scheduler optimizations means that screen updates + can be delayed a long time as well. - This allows drivers to allocate buffers for a drawable one at a time. - As a result, 3D drivers can now allocate the (fake) front-buffer for a - window only when it is needed. Since 3D drivers only ask for the - front-buffer on demand, the real front-buffer is always created. This - allows CopyRegion impelemenations of SwapBuffers to continue working. - As with previous version of this code, if the client asks for the - front-buffer for a window, we instead give it the fake front-buffer. + The change is simple -- prevent the smart scheduler from increasing + the scheduling interval while any screen is using software rotation. - Signed-off-by: Ian Romanick - Reviewed-by: Kristian Høgsberg + Signed-off-by: Keith Packard -commit 98c3c21735197fbd2c8166c9bdabf055e14c9009 -Author: Ian Romanick -Date: Fri Apr 24 12:09:21 2009 -0700 +commit 1740cda7a37abc7d0a169ab4555b446adaa62211 +Author: Keith Packard +Date: Tue Aug 25 16:58:07 2009 -0700 - DRI2: Add interface for drivers to query DRI2 extension version + Perform rotation redisplay before calling driver block handler (which may flush rendering) - Signed-off-by: Ian Romanick - (cherry picked from commit 28ddfc88d8d547941c7f4713db527a3c2f9ec35a) + The rotation block handler uses regular driver rendering functions to + repaint the screen, if those functions queue commands in the driver, + it's important that the driver block handler be invoked after the + rotated image is drawn. + + Signed-off-by: Keith Packard -commit 4cb4c210c365fd40ad314e0707eb38811f240a12 -Author: Ian Romanick -Date: Thu Apr 16 12:10:34 2009 -0700 +commit 4aab05e3b3231f1ec9795a66a075d17a722634a7 +Author: Keith Packard +Date: Tue Aug 25 16:54:16 2009 -0700 - DRI2: Add missing front-buffer flush callback. + xf86_reload_cursors: fix cursor position to eliminate jumping after mode set - Signed-off-by: Ian Romanick - (cherry picked from commit d1e916d29be8b470cbc8cadcf6e83991fdbc5a9f) + xf86_reload_cursors restores the cursor to the correct position, but + that must adjust for cursor hot spot and frame before calling down to + the hardware function, otherwise the cursor jumps to the wrong + position until it is repositioned by the user. + + Signed-off-by: Keith Packard -commit aa13faef2b1464f808e04de9826c6b8b8b91ae89 -Author: Ian Romanick -Date: Wed Apr 15 11:13:48 2009 -0700 +commit 0f9ffc887ca1471e98df746253d9300e03e46a15 +Author: Peter Hutterer +Date: Mon Aug 24 15:10:52 2009 +1000 - DRI2: Don't leave empty entries in private->buffers + Xi: fix XIWarpPointer up for FP3232 as input coordinates. - This should fix bug #21130. + requires inputproto 1.9.99.902 - Signed-off-by: Ian Romanick - (cherry picked from commit de1e43181bd670877b994db221ad8a04b5d63324) + Signed-off-by: Peter Hutterer -commit d7277296ed7aea7bd41b3489d4ceef750d400206 -Author: Ian Romanick -Date: Thu Apr 9 14:38:24 2009 -0700 +commit 5e96945cf54136afdb80cc17f67611251d59205d +Author: Peter Hutterer +Date: Fri Aug 21 12:15:54 2009 +1000 - DRI2: Synchronize the contents of the real and fake front-buffers + Xi: fix swapping for XIWarpPointer and XIChangeCursor requests. - Signed-off-by: Ian Romanick - (cherry picked from commit 567cf67959b30432ae30f4851ec17b3a375ab838) + Signed-off-by: Peter Hutterer -commit 73b786f7e7f46d40bf3b039538540c2e25f45947 -Author: Ian Romanick -Date: Thu Apr 9 14:31:01 2009 -0700 +commit 990d204ef8074665763ad129d0b015b7ab3ae9ef +Author: Peter Hutterer +Date: Mon Aug 24 09:22:16 2009 +1000 - DRI2: Do not send the real front buffer of a window to the client + Xext: remove un-used extern of DeviceMotionNotify. - Signed-off-by: Ian Romanick - (cherry picked from commit f1a995d1496d73741731e32f475097c44a8da972) + Signed-off-by: Peter Hutterer -commit 32d250a881341ece8a1f1d78359adc1b265b5174 -Author: Ian Romanick -Date: Wed Apr 8 15:44:34 2009 -0700 +commit 80f18a73267bd3090e459de609a7006a21e6c649 +Author: Peter Hutterer +Date: Mon Aug 24 09:21:03 2009 +1000 - DRI2: Add fake front-buffer to request list for windows + input: move CorePointer/KeyboardProc declarations into header. - If a front-buffer is requested for a window, add the fake front-buffer - to the list of requested buffers. + The extern declaration in xichangehierarchy.c was broken anyway. + This fixes a crash on creating a new master device. - Signed-off-by: Ian Romanick - (cherry picked from commit aa2928325fe51d94a636dde9c090e8f54a311a12) + Reported-by: Maxim Levitsky + Signed-off-by: Peter Hutterer -commit db61eff891675aeaf2466c3529424de4621005dc -Author: Adam Jackson -Date: Fri May 22 12:01:55 2009 -0400 +commit b96823713129f641ee99306e3cadba8dfe21afa2 +Author: Peter Hutterer +Date: Mon Aug 24 09:06:29 2009 +1000 - EDID: Add modes from Established Timings III descriptor to mode pool + mi: include shmint.h if needed, silence compiler warning. - EDID 1.4, section 3.10.3.9 - (cherry picked from commit 99e22b86c5f1a3653f3caaf01368a777d2b208d0) + Signed-off-by: Peter Hutterer -commit ebca49e0fedcc1c536f1dee7c58cae9903e74fa2 -Author: Adam Jackson -Date: Thu May 21 10:20:48 2009 -0400 +commit 1b3859a49a3861517c9b34f6e8750f7ae0ff2b5d +Author: Peter Hutterer +Date: Fri Aug 21 12:29:39 2009 +1000 - EDID: Be more cautious about finding vendor blocks. - - Many old monitors zero-fill the detailed descriptors, so check for that - to avoid a useless warning like: + Xi: add swapping hook for XIGetFocus reply. - (WW) RADEON(0): Unknown vendor-specific block 0 - (cherry picked from commit a2c5ee36b21c2ee5c0468f1b251e74c1412dbecb) + Signed-off-by: Peter Hutterer -commit c643d24cdeaade8d7839691a3113c4d2d17be61e -Author: Adam Jackson -Date: Fri May 29 18:07:48 2009 -0400 +commit fe430d7d3b0a85fbea9e93171423688f76dc9e1e +Author: Peter Hutterer +Date: Fri Aug 21 12:09:32 2009 +1000 - vfb: Fix depth setup. + test: add XIQueryPointer protocol tests. - Initialize the depth corresponding to the root window before the - pixmap-only depths. Otherwise you end up with the root window depth in - the depth list twice, which is mildly confusing for clients and - catastrophically confusing for PanoramiXConsolidate(). - (cherry picked from commit 45530d16097459a756696e255ab4e72d6e51fbc4) + Signed-off-by: Peter Hutterer -commit 5cd5a01259ba349f1868ca4af04207cf120d69e4 -Author: Eamon Walsh -Date: Tue May 19 19:30:33 2009 -0400 +commit 285c88018ba4b3819500027c9b6a05b8c27ff2ee +Author: Peter Hutterer +Date: Fri Aug 21 11:38:42 2009 +1000 - xace: Fix a bad device access hook call. - - Add a proper access mode, and reverse the logic of the return value. - Zero ("Success") is returned on success from the hook calls. + Xi: correct length field for XIQueryPointer reply. - Signed-off-by: Eamon Walsh - (cherry picked from commit 3cea176d5abcb0f14eefbdcbe17fed0847524dd4) + Signed-off-by: Peter Hutterer -commit a9f85dcefbadfe44d8c58ad08430aaadb8c59d34 -Author: Eamon Walsh -Date: Mon May 11 15:27:46 2009 -0400 +commit e4ea91a02d6fac9dad844150c5d2fbe7dc9629b1 +Author: Peter Hutterer +Date: Fri Aug 21 12:12:05 2009 +1000 - xselinux: Relax ownership restriction on SetSelectionUseContext. - - Instead, clients should keep track of the selection instances they use. - (cherry picked from commit 0952d12717031e9dda9e48123bb922d0f4e81834) + Xi: allow XIQueryPointer requests for master pointers and floating slaves. - Signed-off-by: Eamon Walsh + Signed-off-by: Peter Hutterer -commit 0fbf6f30f771e60714fb3003a9bb1a6b79e50c6e -Author: Alan Coopersmith -Date: Thu Mar 26 22:22:32 2009 -0700 +commit 8e396f2b806496cdcac233cc731b0322735020c3 +Author: Peter Hutterer +Date: Fri Aug 21 12:11:19 2009 +1000 - Don't leak default font path when appending built-ins + Xi: fix swapping for XIQueryPointer request/reply handling. - Signed-off-by: Alan Coopersmith - (cherry picked from commit 66539cc05d0b017b9feb4a038499907810140623) + buttons_size is necessary as WriteToClient swaps the buttons_len field, + resulting in the wrong number of bytes being written later. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 773e4890b385a229b0919b4a249714d0bb7171a2 -Author: Alan Coopersmith -Date: Thu Mar 26 22:25:08 2009 -0700 +commit f1c856ef9073d9d02bc84ce1472883c8441a9758 +Author: Peter Hutterer +Date: Fri Aug 21 10:42:15 2009 +1000 - Don't leak canonical module name and patterns if module is built-in - - Signed-off-by: Alan Coopersmith - Acked-by: Peter Hutterer - (cherry picked from commit 1c101d75d4855b2698e3fc8d2dd662f20585812f) + test: fake initialization of a sprite trace and window locations - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 21f58a2b0d0d96492d2d7c7de1bd654ed0114279 +commit 58fd28f6b6bebab7f31c54e696f96d64ebb50f5a Author: Peter Hutterer -Date: Wed May 6 16:55:32 2009 +1000 +Date: Fri Aug 21 09:44:37 2009 +1000 - Xi: don't double-swap the XListDeviceProperties reply. + Xi: return deviceid as error value in XIQueryPointer. Signed-off-by: Peter Hutterer - (cherry picked from commit 50cc8adafca4ba3838d468278d6eb8a4692d2488) - - Signed-off-by: Keith Packard -commit b575af88d0e547a03b164f6f3b2d4b268b9ffc34 +commit b41dd4328bfb8a0981d299bb60f986a2bc55fd24 Author: Peter Hutterer -Date: Wed May 6 22:25:28 2009 +1000 +Date: Fri Aug 21 09:43:46 2009 +1000 - Xi: fix copy/paste error causing sizeof against wrong struct. - - This wrong check may cause BadLength to be returned to the client even if the - length is correct. + record: include recordproto.h, require recordproto 1.13.99.1 Signed-off-by: Peter Hutterer - (cherry picked from commit 8b583ca2b21155359c6255f406c96599b277c762) - - Signed-off-by: Keith Packard -commit 9c643b1e3ddc4ec9f1a924bd25f24df8536c7ea6 +commit 65354e5a698a5b527db09afc431110afba0e14b2 Author: Peter Hutterer -Date: Thu Apr 16 16:17:07 2009 +1000 +Date: Thu Aug 20 10:52:52 2009 +1000 - os: don't malloc memory in LogVMessageVerb. - - LogVWrite is limited to a buffer size of 1024, so we don't loose anything here - by truncating. This way we can use LogVMessageVerb (and xf86Msg and friends) - during signal handlers with the normal message types. + Xi: return BadDevice for changing device cursors on non-master pointers. Signed-off-by: Peter Hutterer - Acked-by: Alan Coopersmith - (cherry picked from commit 0e0642ee9466d3268476d0084a83a9d93a4aa555) + +commit ef75301854d41c6b11dd2742fa9d17a5de8a4cfa +Author: Peter Hutterer +Date: Thu Aug 20 10:52:24 2009 +1000 + + dix: FixUpEventFromWindow mustn't scramble non-device events. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 7cd596e0dd6b9c195d7e395c26bfb00ef544c525 -Author: Dave Airlie -Date: Wed Apr 15 18:26:06 2009 +1000 +commit c73cd3b265c301b8a54ffe484d6c696f2abefb46 +Author: Dan Nicholson +Date: Sun Aug 16 18:07:49 2009 -0700 - randr12: looking up these bits if randr isn't initialised is bad. + xfree86: Link libselinux with Xorg system libraries - When xinerama is enabled we don't get randr protocol, but the - driver might still want randr internals - (cherry picked from commit faf7dfa099f5b42a703313fbd1bf8afdad07a179) + Put the reference to -lselinux in XORG_SYS_LIBS so it isn't added as a + make dependency for libxorg.la. Otherwise, make goes looking for a file + -lselinux in the current directory, which it obviously won't find. - Signed-off-by: Keith Packard + Signed-off-by: Dan Nicholson + Tested-by: Justin P. Mattock -commit bc95f29441b3daa7b9d91bd3c2b3e502a3a8cc81 -Author: Alan Coopersmith -Date: Thu Apr 23 08:42:38 2009 -0700 +commit 9d8fc33cae2fee7777edd20658dbf6a3ffb3d557 +Author: Adam Jackson +Date: Thu Aug 20 16:56:04 2009 -0400 - Fix byte swapping of XF86VidMode{Get,Set}GammaRamp - - Fixes OpenSolaris Bug 8315: - Xorg segfaults when screensaver fades in cross-endian xdmcp session - + composite: Move screen init before extension registration. - Signed-off-by: Alan Coopersmith - (cherry picked from commit 7d0f7518c2235a9dc783029971259ddaada2db20) + Otherwise, you'd still advertise the extension even if no screens + actually supported it, and the first Composite protocol request would + probably crash. + +commit ae68708ebe24c4cdf8c5265e206f6913d9298f37 +Author: Adam Jackson +Date: Thu Aug 20 15:45:57 2009 -0400 + + xfree86: dump /proc/cmdline in the log on Linux + +commit 0b131a5cd91cea54240777c66a9cd385029e8cb2 +Author: Adam Jackson +Date: Thu Aug 20 15:28:57 2009 -0400 + + linux: Yet more malloc() avoidance for backtrace() - Signed-off-by: Keith Packard + Turns out, there's an initializer at the top of backtrace() that (on + some arches) calls dlopen(). dlopen(), unsurprisingly, calls malloc(). + So, call backtrace() early in signal handler setup so we can later + safely call it from the signal handler itself. -commit 3f0c73694b9ddc85b82a7a82a5d98fa8b18a01a8 -Author: Jon TURNEY -Date: Thu Feb 12 16:03:16 2009 +0000 +commit 792dee3854cbede64e17fdc4736831edad295706 +Author: Tiago Vignatti +Date: Wed Aug 19 18:20:54 2009 +0300 - Remove references to rgb.txt from files section of Xserver and Xorg man pages + xfree86: remove bus state change notification callback - The references to this file in the server code were removed in commit dda10c9066a660b647384179f82e1da8e063264f + No one is using bus notifications now. We hope that the kernel take care of + this properly. - Signed-off-by: Jon TURNEY - (cherry picked from commit d0dd649035fc3698c5b436f9d9d248116aa106a3) + For other not-so-urgent-notifications (ACPI wakeups, etc) we can just register + a handler on server's scheduler (using xf86AddGeneralHandler). And for + external applications, the "trend" is to use HAL to kick notifications. So + we're already provided of enough notification schemes. - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti + Signed-off-by: Aaron Plattner -commit 78eb366c65a953bbc059a989a92dfe28fbe1ae29 -Author: Jon TURNEY -Date: Thu Feb 12 16:02:17 2009 +0000 +commit eb031d4013d36f6aef4aba45840762ae8635cc13 +Author: Adam Jackson +Date: Wed Aug 19 15:42:34 2009 -0400 - Remove long-gone '-co' option from Xserver man page + linux: hand-roll a backtrace printer instead of using backtrace_symbols - The '-co' option was removed in commit dda10c9066a660b647384179f82e1da8e063264f + Why? Because backtrace_symbols calls malloc, which you can't do from a + signal handler. Face? Palm. + +commit 75b9383d8a4c113ab3c6cfc1d5efcb5d9982a1bf +Author: Jesse Adkins +Date: Tue Aug 18 21:13:40 2009 -0700 + + xserver doesn't stop all connections to localhost - Signed-off-by: Jon TURNEY - (cherry picked from commit b0ad9e1ced9619f37acf77764c395c57b86cf463) + X.Org Bugzilla #23329: http://bugs.freedesktop.org/show_bug.cgi?id=23329 + Patch #28648: http://bugs.freedesktop.org/attachment.cgi?id=28648 - Signed-off-by: Keith Packard + I noticed in xserver at os/access.c that xorg tries to stop connections + to localhost by checking against the address 127.0.0.1. However, RFC + 3330 defines the localhost network as 127.0.0.0/8. This means that any + IPv4 address that starts with 127 is just another name for localhost. -commit 31536307703fcffa14a027ec777310c9962f816c +commit 2d34eace13c2016048c627c4e96c3b2399901078 Author: Alan Coopersmith -Date: Sat May 2 00:26:18 2009 -0700 +Date: Tue Aug 18 20:11:23 2009 -0700 - XkbSetNamedIndicator should ignore SD's without LED's + Sun bug 6872917: Xorg not querying /dev/fb when no xorg.conf exists - When ProcXkbSetNamedIndicator is called on a core device, and we - walk the slaves to set the LED's on each of them, ignore any slaves - that do not have either a KbdFeedbackCtrl or LedCtrl structure. + commit 48ee5558333bd324463b6994735cabb23de262ec (OpenSolaris VT support) + broke the autoconfiguration code in xf86AutoConfig.c that uses the + Solaris-specific VIS_GETIDENTIFIER ioctl on a frame buffer device like + /dev/fb by changing xf86Info.consoleFd from /dev/fb to a /dev/vt/* + device. - (This is much more critical in xserver-1.5-branch, where we walk - *all* devices, not just the slaves of the specified master, and - thus return failure when setting an LED on the Core Keyboard and - hit a xf86-input-mouse device with no LED's to set.) + This fixes it by reworking the code to split the console device + (/dev/vt/*, the vtXX CLI option) from the frame buffer device + (/dev/fb, -dev option) to allow both VT and autoconfig to work. - Signed-off-by: Alan Coopersmith - Acked-by: Peter Hutterer - (cherry picked from commit 557dbadf3be273255e8fdb12d9321f4e88bf2b65) + It also fixes the console device to use /dev/fb when VT's are not + supported instead of throwing a Fatal Error because it can't open + /dev/vt/0. - Signed-off-by: Keith Packard + Signed-off-by: Alan Coopersmith -commit 01f0210b2ba34f724aa221bbe6b567a5751781a6 -Author: Keith Packard -Date: Fri May 8 22:29:15 2009 -0700 +commit d225230148b3cdab3b32fce3abee26b818d4bab0 +Author: Jesse Barnes +Date: Tue Aug 18 12:15:32 2009 -0700 - xserver 1.6.1.901 + xfree86/linux: don't change VT perms unless we're running as root - Signed-off-by: Keith Packard + In non-setuid root installations, we shouldn't try to adjust VT/tty + ownership. It will fail, and shouldn't be necessary anyway (since + startup scripts or PAM should be handling perms for us in that case). + + Signed-off-by: Jesse Barnes -commit e21c9949ac456ba9b162b6244a17e68c5b8112a9 -Author: Pierre Ossman -Date: Tue Apr 28 09:34:55 2009 +0200 +commit 41884b8f4dfe4d931c64a6046adfefcc2b7646c2 +Author: Tiago Vignatti +Date: Tue Aug 18 15:47:56 2009 +0300 - dix: fix calculation of number of fake KeyRelease events + xfree86: remove unused code (xf86Configure.c) - Signed-off-by: Pierre Ossman - Signed-off-by: Keith Packard + The xorg.conf generator was not assigning correctly the primary device + ("bootable") as screen zero. So just skip this kind of routines for now. + + Signed-off-by: Tiago Vignatti -commit 3b1ff33587458bad45777b8eb74b81c3e55987ad -Author: Pierre Ossman -Date: Tue Apr 28 12:29:07 2009 +0200 +commit 097376266f256e3ae37a16b343ed1e3441a8f003 +Author: Tiago Vignatti +Date: Tue Aug 18 13:13:57 2009 +0300 - Xi: don't send XKB mapping notifications when XKB is disabled + xfree86: remove unaffected code (xf86Configure.c) - Signed-off-by: Pierre Ossman - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti -commit 4b702526dec35f5cfe5c6baec6a77c02c95c2867 -Author: Peter Hutterer -Date: Wed Mar 25 13:08:27 2009 +1000 +commit dd8960a9f37fbef65cb99e058c3772edfda7b6db +Author: Tiago Vignatti +Date: Tue Aug 18 12:53:47 2009 +0300 - dix: ignore non-pointer events in XineramaCheckMotion (#20557) + xfree86: remove stupid macro (xf86Configure.c) - Note that deviceKeyButtonPointer and keyButtonPointer have the same wire - layout, so we only need to check for event types. + It was confusing more instead helping. - X.Org Bug 20557 + Signed-off-by: Tiago Vignatti + +commit 4ec7667bb17839bd4ec11091c75ecfff317fc96c +Author: Tiago Vignatti +Date: Tue Aug 18 12:33:35 2009 +0300 + + xfree86: remove wrong commentary - Signed-off-by: Peter Hutterer - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti -commit fdbb6fd3d3c0ce7078f8faaf089af51cc36cbcb3 -Author: Jesse Barnes -Date: Mon May 4 15:38:22 2009 -0700 +commit 2e8c2b8ae6a092871838b7cceaaf4a7590bce1d3 +Author: Tiago Vignatti +Date: Tue Aug 18 12:28:15 2009 +0300 - Don't prepare outputs & crtcs if set_mode_major is present + xfree86: header clean up (xf86Configure.c) - A driver with this hook will take care of preparing the outputs & crtcs, - so calling the prepare functions will just cause unnecessary flicker. + Signed-off-by: Tiago Vignatti + +commit 2df414a33321ae772af1e7e52c36aec222bb3c8f +Author: Tiago Vignatti +Date: Tue Aug 18 12:14:27 2009 +0300 + + xfree86: reorganize xf86Configure.c - Fixes bug #21077 - (cherry picked from commit 94648bb797d94b025746c60679c584e5be2fae28) + No semantical changes. Only code moved around. - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti -commit eac2b3658aacc8a64dc3c0fd7cc8fd77abcc59af -Author: Julien Cristau -Date: Wed Apr 22 20:01:49 2009 +0200 +commit 113e66a1ffe2a43c4d1c92a7d78ca43b4db58822 +Author: Tiago Vignatti +Date: Sun Aug 16 21:02:11 2009 +0300 - Bug#21324: Add quirk for Iiyama Vision Master 450 - - Reported-by: Jeremy Henty - Signed-off-by: Julien Cristau - (cherry picked from commit 0dfb97f15f591f85e079f5829c77d0c328d00464) + xfree86: remove unused PCI header - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti -commit b3c42e6450b0af97ead4f410983c803a959e1d04 -Author: Tormod Volden -Date: Wed Apr 1 19:32:21 2009 +0200 +commit 6b5978dcf1f7ac3ecc2f22df06f7000f360e2066 +Author: Richard Hughes +Date: Mon Aug 17 09:15:32 2009 +0100 - xfree86: edid quirk for Philips LCD LP154W01-TLAJ + Do not reset lastDeviceEventTime when we do dixSaveScreens - This panel reports its vertical size in cm. + When we turn off DPMS with DPMSModeOff and do dixSaveScreens, don't reset the + event time else session clients using IDLETIME will be reset. - X.Org bug#21000 + Signed-off-by: Peter Hutterer + +commit 4098ad72d00e65d142fb9fe8a1194be35338508b +Author: Eamon Walsh +Date: Sun Aug 16 00:06:10 2009 -0400 + + xace: fix access mode in dixLookupWindow within several RandR calls. - Signed-off-by: Tormod Volden - Signed-off-by: Julien Cristau - (cherry picked from commit b1dab580bdfb4acfe3feddeda6e760098ec4922a) + Referencing a screen using a window only requires GetAttr access. - Signed-off-by: Keith Packard + Signed-off-by: Eamon Walsh -commit 96eede556ac9f95204fa54287f74794acdbf3a2b -Author: Julien Cristau -Date: Wed Dec 24 14:17:27 2008 +0100 +commit d4b8f7602b5e266a0ebd3b1ba23724362cc7de3a +Author: Eamon Walsh +Date: Mon Aug 10 18:39:04 2009 -0400 - Add RandR 1.3 requests to protocol.txt - (cherry picked from commit 4f86ee61a4abf7a29e565d095aa08abd0ca9dc66) + xace: fix access mode in dixLookupWindow call within ProcUngrabKey. - Signed-off-by: Keith Packard + Referencing a window (as grab-window) only requires GetAttr access. + + Signed-off-by: Eamon Walsh -commit b7b520e3a93fde5579c48eabb9b8c694a7dfbcb1 -Author: Julien Cristau -Date: Wed Dec 24 14:23:37 2008 +0100 +commit 6d2f4e487869f10de4a62365b4d6de036c752ab8 +Author: Eamon Walsh +Date: Mon Aug 10 18:36:53 2009 -0400 - Add XI 1.5 event and requests to protocol.txt - (cherry picked from commit 063833f3a6d9f8f657e3de309c8d6d5c3d606513) + Add XI2 requests to protocol.txt - Signed-off-by: Keith Packard + Signed-off-by: Eamon Walsh -commit 0df12a3f01c953c0908b4b53a93fd3a5c704f69b -Author: Ander Conselvan de Oliveira -Date: Mon Apr 6 16:01:20 2009 -0300 +commit 73975ef3a39ce522c6206ca800ed175fbf851dcf +Author: Eamon Walsh +Date: Mon Aug 10 18:34:07 2009 -0400 - xfree86: Remove device from inputInfo.devices if ActivateDevice failed. + xselinux: Allow per-client device create contexts. - After the call to xf86ActivateDevice, the new device will be added to - inputInfo.devices. However, if the subsequent call to ActivateDevice - fails, the correponding InputInfoRec for the device is deleted but an - entry still remains in inputInfo.devices. This might lead to a server - crash later on (on InitAndStartDevices for instance) when the device - control proc would be called for an invalid device. - (cherry picked from commit efa31092d6703397121a0ada4f7205a8ecad3d3d) + The previous behavior was to set the serverClient's value which was used globally. + This is in support of XI2, where clients can create device pairs directly. - Signed-off-by: Peter Hutterer - Signed-off-by: Keith Packard + Signed-off-by: Eamon Walsh -commit 24dc7694a2ba72d45cda7720aeaec8350930edc3 -Author: Michel Dänzer -Date: Fri Feb 6 11:37:53 2009 +0100 +commit 66e32d252cffcd4fe7d505f1c211253f23c5002c +Author: Eamon Walsh +Date: Mon Aug 10 18:33:05 2009 -0400 - EXA: Guard empty pending region warning by DEBUG_MIGRATE. + xselinux: Add more new device permissions for XI2. - It isn't very useful yet while the damage layer calls us for empty operations, - mostly confuses users. - (cherry picked from commit 3948b523893d3d44b6a088340c4252e969613769) + Reflects the ability of clients to create/destroy device objects. - Signed-off-by: Keith Packard + Signed-off-by: Eamon Walsh -commit 5198ff6f86e4796819c0d1f742de57a69228c93b -Author: Michel Dänzer -Date: Fri Feb 27 16:37:28 2009 +0100 +commit 1f4ea22a20fdd30412a259a87eee133155e1163d +Author: Rémi Cardona +Date: Mon Aug 17 11:33:24 2009 +0200 - EXA: Handle separate alpha maps properly in Composite fallback, take two. + configure: fix help string after Xnest default build was changed - Preserve the EXA ABI by introducing a new driver flag EXA_SUPPORTS_PREPARE_AUX. - If the driver doesn't set this flag, we have to assume any Prepare/FinishAccess - driver hooks can't handle the EXA_PREPARE_AUX* indices, so we move out such - pixmaps at PrepareAccess time. + Signed-off-by: Rémi Cardona + +commit 7e37e7c50e4fdcb53296a99d90af3d90081ce9bb +Author: Peter Hutterer +Date: Tue Aug 11 14:39:50 2009 +1000 + + test: add protocol tests for DeviceChangedEvents - Fixes https://bugs.freedesktop.org/show_bug.cgi?id=18710 . + Signed-off-by: Peter Hutterer + +commit 2851f04cb2f6e5c30267f733d867c86d4e69a485 +Author: Peter Hutterer +Date: Tue Aug 11 16:27:40 2009 +1000 + + dix: rework DeviceChangedEvents a bit. - Signed-off-by: Michel Dänzer - (cherry picked from commit 4cfb36f6ad2df01215028fec48d99239a0e4496b) + DCEs are now processed when sent throught the master device, not when sent + through the slave device. This includes a removal of some un-used (or partly + used) fields in the DCE itself to something more self-explanatory. - Signed-off-by: Keith Packard + TODO: if a device has events queued and its attachment is changed, the DCE + is silently dropped now. Instead, it should be generated as soon as the + first event after the attachment is sent. + + Signed-off-by: Peter Hutterer -commit 199bb367152d68e784dfbec79ab9b70540d83fc2 -Author: Ian Romanick -Date: Mon Apr 20 18:20:52 2009 -0700 +commit c1d901d723c3bee523736eacc15b44a7dff484fe +Author: Richard Hughes +Date: Fri Aug 14 11:44:35 2009 +0100 - DRI2: Send the version the code actually supports + Don't reset the lastDeviceEventTime when doing DPMS actions - This prevents building an older server with a new dri2proto.h from - resulting in a DRI2 extension module that lies about the version it - supports. + When we change the DPMS mode, don't play games with the last event time as + this breaks applications using IDLETIME to turn the backlight off after a + preset time. - Signed-off-by: Ian Romanick - (cherry picked from commit 44227ef1b77467c76147b9bf79bdd0e6305a522a) + This patch fixes gnome-power-manager and xfce-power-manager - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 11db545a86c8933c638a0bc1fcd4f2c65279f617 -Author: Eamon Walsh -Date: Thu Apr 16 22:48:11 2009 -0400 +commit dca4de72d375c4b9bcdd25b151e291a77a5b06cc +Author: Peter Hutterer +Date: Sat Aug 15 20:17:20 2009 +1000 - security: Grant untrusted windows remove access on all windows. + dix: fix potential use of unused variable 'mask'. - This allows untrusted clients to destroy their own windows when they - have been reparented by a trusted window manager. - (cherry picked from commit 4559d2ace6ac55fe361f572ded0769cdd1f3b545) + Signed-off-by: Peter Hutterer -commit 5d48f288eccb57dfe1751946c9d60dbd3ac0129a -Author: Eamon Walsh -Date: Thu Apr 16 22:39:54 2009 -0400 +commit 978b65bce14598f2d42ca0177ea58fef71fc12c5 +Author: Peter Hutterer +Date: Fri Aug 14 15:05:42 2009 +1000 - security: Fix a crash caused by wrong ordering of format arguments. - (cherry picked from commit 3481b32ab971c41cb972f6819ae049f3e9f7033b) + include: XInputExtensionInit doesn't need to be exported. + + Signed-off-by: Peter Hutterer -commit 93e0a648138e569087fc6e07c1a28abfa92a4dde -Author: Eamon Walsh -Date: Thu Apr 16 22:33:12 2009 -0400 +commit 3989dc1d34f116f30915632cc5286937392e180a +Author: Peter Hutterer +Date: Tue Aug 11 16:20:20 2009 +1000 - security: Revert behavior of extension access for compatibility. + dix: use the XI2 defines for class types. - Previously, three extensions were defined as "trusted" by the extension: - BIG-REQUESTS, XC-MISC, and XPrint. No other extensions were permitted - to be used by untrusted clients. + Doesn't matter really, they have the same values anyway. - In commit 8b5d21cc1d1f4e9d20e5d5eca44cb1e60a419763 this was changed for - some reason. Return to the old, compatible behavior. - (cherry picked from commit 6045506be0cebca4ebbe943ae77f020aafa703d4) - -commit cd321badf0c185854b3e0c038f3851a3ac7864ea -Author: Adam Jackson -Date: Tue Apr 14 16:08:11 2009 -0400 - - xserver 1.6.1 + Signed-off-by: Peter Hutterer -commit fcffa3a178683ede0d15656c432fccf23ebca5ef +commit fe045820f1fb33991e8bff5c6e192097caa85727 Author: Peter Hutterer -Date: Wed Mar 25 15:51:43 2009 +1000 +Date: Mon Aug 10 15:34:00 2009 +1000 - Xext: set POINTER_SCREEN flag in XTestFakeInput if necessary. (RH #490984) + input: move DeviceChangedEvent conversion into eventconvert.c - The POINTER_SCREEN flag must be set explicitly for XTest core events to avoid - out-of-range events when the lastSlave was an SD with an explicit axis range. - Device events sent through XTest don't need this flag, they are expected to be - in the valuator range of the device anyway. - - Red Hat Bug 490984 + The version in eventconvert.c was half broken and for some reason we ended + up with a second version in exevents.c (which works). Move it over to where + it belongs and call EventToXI2 instad of having a custom function for it. Signed-off-by: Peter Hutterer - (cherry picked from commit 603db34337a61754e0c5f71525011d10eab78411) -commit 6ff60c626141771b365d557fe64bc93f200d1c9c -Author: Kristian Høgsberg -Date: Thu Apr 9 13:16:37 2009 -0400 +commit 33eb6f70816921abc5da62e434f40e78d672274e +Author: Peter Hutterer +Date: Fri Aug 14 13:49:58 2009 +1000 - glx: Fix drawable private leak on destroy + include: Unexport most symbols from exevents.h. - When a drawable goes away, we don't destroy the GLX drawable in full, - since it may be current for a context. This means that when the drawable - is destroyed in full later, the backend doesn't get a chance to - destroy resources associated with the drawable (the DRI2Drawable). + And shuffle them around so that the part used by drivers is up the top and + commented. - With this patch, we destroy the GLX drawable in full when it goes away - and then track down all contexts that reference it and NULL their - pointers. + Also, woo, the sdksyms script doesn't like declarations with return type and + function name on the same line... - (cherry picked from commit 7b6400a1b8d2f228fcbedf17c30a7e3924e4dd2a) - -commit 50894e13545538ed93f20823f7ba81d1ab8906bb -Author: Eamon Walsh -Date: Thu Apr 9 02:48:04 2009 -0400 - - xselinux: Don't BadAlloc in List* requests if there are no items to list. - (cherry picked from commit 682d7b55699cacbb2dbcd84a5e816bf6e2d2f02a) + Signed-off-by: Peter Hutterer -commit feb01d7d6e98fa77f9069b08aaa9727368ef3aaf -Author: Eamon Walsh -Date: Wed Apr 8 15:10:16 2009 -0400 +commit 1373c2ea877b562076a4a3c97e8e4f999c977217 +Author: Dave Airlie +Date: Sat Aug 15 12:14:26 2009 +1000 - xselinux: Don't require incoming context strings to be null-terminated. - (cherry picked from commit e8b324102f6e21ae2b8292a6f50d016dd6254dd6) + exa: clarify createpixmap2 new pitch return -commit b7dc7374bbcb708eee6eec26ff141619f914d8eb -Author: Eamon Walsh -Date: Mon Mar 9 17:28:40 2009 -0400 +commit 48ee5558333bd324463b6994735cabb23de262ec +Author: Aaron Zang +Date: Mon Aug 3 23:21:39 2009 -0700 - Correct access mode in call to dixLookupWindow() within RRSelectInput. + Add new VT support for OpenSolaris & future Solaris releases - Reported by Alan Coopersmith. - (cherry picked from commit 6544490700051b3b5e88ac1890d71b35634c9100) + Signed-off-by: Aaron Zang + Signed-off-by: Alan Coopersmith -commit 38b0f9edca80e43b4c8c364cfebc25c835d24cbd -Author: Adam Jackson -Date: Fri Mar 27 16:49:45 2009 -0400 +commit 613e76ff9055d8ac2b1af1130668180646a9e14c +Author: Alan Coopersmith +Date: Mon Aug 3 21:38:51 2009 -0700 - Fix distcheck with dolt + Remove support for Solaris x86 releases older than Solaris 8 + + If you want to run a pre-1999 kernel, you'll need a pre-2009 X server + + [Some pre-Solaris 8 VT support is left by this patch to allow reuse by + the new Solaris VT support that follows in the next patch.] + + Signed-off-by: Aaron Zang + Signed-off-by: Alan Coopersmith -commit 509986d6f0255588f37446173ea76640895c9886 -Author: Adam Jackson -Date: Fri Mar 27 16:28:26 2009 -0400 +commit 6c292d17053eb2a7e7054e51210f423dbc0cb7e8 +Author: Peter Hutterer +Date: Fri Aug 14 11:48:54 2009 +1000 - Fix Changelog generation + dix: update the sprite trace for all masters && floating slaves (#23257) + + When the windows are restructured, CheckMotion needs to be called for all + masters and floating slaves to update the spriteTrace. + + X.Org Bug 23257 + + Tested-by: Thomas Jaeger + Signed-off-by: Peter Hutterer -commit 0f9fd02dacf4363c0efd69825bc804bfb55599bf -Author: Adam Jackson -Date: Mon Mar 16 13:24:48 2009 -0400 +commit 49046088f10cceaea7da97401d742d3fb59371f5 +Author: Peter Hutterer +Date: Fri Aug 14 09:48:45 2009 +1000 - selinux: Add support for avc_acquire_netlink_fd() + config: don't shutdown the libhal ctx if it failed to initialize (#23213) - Requires libselinux 2.0.79 or newer. Without this, libselinux will - check for policy updates on the netlink socket on basically every policy - lookup. Statistically speaking, they never happen, and the check - translates to at least one more syscall on basically every operation. + Regression introduced by b1c3dc6ae226db178420e3b5f297b94afc87c94c. + Shutting down the libhal_ctx if the init failed may cause an abort. + This can happen if hald is not yet running at server startup. - Instead, take control of the fd from the library, and check it in - WakeupHandler if it polls readable. - (cherry picked from commit 3992dd38caf33b343affd8d732c94880d1099dcf) + X.Org Bug 23213 + + Tested-by: Stefan Dirsch + Signed-off-by: Peter Hutterer -commit b3484c848040ae29b8ca1327d3b142db30c38387 -Author: Adam Jackson -Date: Sat Feb 21 19:56:20 2009 -0500 +commit 1545a120df6dffb5b84fe96c5a992357520b7c8d +Author: Dave Airlie +Date: Tue Aug 11 15:00:36 2009 +1000 - selinux: Don't bother relabeling resources that are being destroyed + exa: fix CreatePixmap2 to be useful for tiling. - Makes window destroy about 40x faster in Xvfb. - (cherry picked from commit b030f858f2f1ce1fd27a73ebf7f9ec5db541a668) + This adds a pitch return so that the driver can align the pitch to any + value it wishes and not just the one it gave to EXA at startup. -commit 08ac9140f6042ce3faa618613fc8443c92924a21 +commit db568f9eabf3450d8a023597ff007df355b13ea8 Author: Peter Hutterer -Date: Fri Feb 13 09:56:22 2009 +1000 +Date: Thu Aug 13 16:37:00 2009 +1000 - xkb: Fix wrong colour reference in XKB geometry copying. #20081 + Xext: fix up wrong conditions for negative sync transitions. - base_color and label_color need to reference the color in the destination, not - in the source. + If the counter had a value higher than the trigger value for a negative + transition, the trigger value did not get set. - X.Org Bug 20081 + The correct sequence of checks is: + if (positive transition) + if (counter value < trigger value) + set up trigger + if (negative transition) + if (counter value > trigger value) + set up trigger + + Red Hat Bug 501601 Signed-off-by: Peter Hutterer - Signed-off-by: Daniel Stone - (cherry picked from commit f5bf1fdaf36163d5c2f1b9b51df96326ebbb0e9c) -commit 124da1da5cd3ac85edd7f0e30f61ee842d49a6a4 +commit a3e50b05747cab92090ae7d7f4475cd61d3fcadf +Author: Pierre-Loup A. Griffais +Date: Mon Jul 27 20:58:44 2009 +0200 + + Add 4 missing 10bpc picture formats to the server format list. + + Signed-off-by: Aaron Plattner + +commit 36e24a6d93bd5aced4e566b80bf2d03555fab9ca Author: Peter Hutterer -Date: Mon Mar 9 08:59:01 2009 +1000 +Date: Thu Aug 13 10:50:26 2009 +1000 - Xi: don't crash on a NULL property name, just return None. + Xext: add missing return code check to ProcSyncDestroyAlarm - strcmp doesn't like NULL-pointers. + Introduced with 57aff88c7d0761e590806d07bee1c9410680c89f. Signed-off-by: Peter Hutterer - (cherry picked from commit 0d9a42dc0380d1583889b6b6521bd5a2451735d4) -commit ea9f51fd42ae0285aed707b7efa44c558dad4e4b +commit ce69a06aff934b2dcded8606cab079ac6465007c Author: Peter Hutterer -Date: Fri Dec 5 16:24:57 2008 +1000 +Date: Tue Aug 11 16:16:54 2009 +1000 - Xi: add XATOM_FLOAT to server-defined properties. + Xi: fix up broken DeviceChangedEvent swapping code + +commit b44d34d5fd0d8aaacb89121e8b4afba04f1dcc80 +Author: Peter Hutterer +Date: Mon Aug 10 15:32:59 2009 +1000 + + include: correct a copy/paste error in a comment. - This property is used to denote type float for input properties. Such - properties can be accessed easily through the XIPropToFloat() function. + Signed-off-by: Peter Hutterer + +commit c477ac8a151a28ba23f5c87e8f7affd6e82cfaa9 +Author: Peter Hutterer +Date: Mon Aug 10 15:15:02 2009 +1000 + + dix: re-name internal eventToClassesChanged to eventToDeviceChanged - Code originally written by Simon Thum. + DeviceClassesChangedEvents (where this name comes from) have been replaced + with DeviceChangedEvents. - Signed-off-by: Peter Hutterer - (cherry picked from commit a48c81dcdf569a3f634ac23e08d2491354de6a36) + Signed-off-by: Peter Hutterer -commit 6ad6b6f900e9c3a55e4ee509bdc9ff0ddec9a59a -Author: Peter Hutterer -Date: Wed Nov 19 15:50:57 2008 +1000 +commit af38f6790c4c8ba1e70f5c5ba3530ded85d6e372 +Author: Peter Hutterer +Date: Mon Aug 10 14:08:59 2009 +1000 - Xi: add XIPropToInt() auxiliary function. - - Converts an XIPropertyValuePtr to an integer, provided that type and format is - right. + test: add event conversion tests for XIDeviceEvents + +commit 3f2e4b9867b6877ee7be32b151fcaf221ef0812f +Author: Peter Hutterer +Date: Mon Aug 10 14:07:54 2009 +1000 + + Xi: add event swapping for XIRawEvents. - Code originally written by Simon Thum. + Signed-off-by: Peter Hutterer + +commit f3b2f9fb734ecfff6db9ae85b0d247856ede8112 +Author: Peter Hutterer +Date: Mon Aug 10 13:12:40 2009 +1000 + + Xi: fix event swapping for XIDeviceEvents. - Signed-off-by: Peter Hutterer - (cherry picked from commit 669f6810af9a89187d6149841925fe765f3988ff) + Signed-off-by: Peter Hutterer -commit 667865b9eb98ed73461028b3d38be19619712533 -Author: Adam Jackson -Date: Mon Mar 2 14:26:23 2009 -0500 +commit e46f02fa2de79261221b42ab73f9daa2ce8ac650 +Author: Peter Hutterer +Date: Mon Aug 10 09:50:16 2009 +1000 - randr: Fix thinko in xf86TargetPreferred + Xext: allocate a separate event list for XTest events (#23100) - The only-one-output case would only work right if that also happened to - be the zeroth output. Oops. - (cherry picked from commit 8a6ed44a8b2fc5f14729dc54fec17607ced03859) + XTest event processing may be interrupted by a SIGIO. If Xtest uses the same + event list as the rest of the server, this list may be overwritten + in-flight. + + X.Org Bug 23100 + + Signed-off-by: Peter Hutterer -commit f469726fec502ce29999eda6919c3c3d26c127d8 -Author: Eric Paris -Date: Tue Mar 3 13:15:39 2009 -0500 +commit 8483b08152104dc619d9e44f877d2687d866355e +Author: Peter Hutterer +Date: Fri Aug 7 16:11:37 2009 +1000 - This patch changes all places in the X code to use _raw functions. The - X server should never see, translate, or deal with a munged context. - Display managers which show contexts to the user should take care of - translating these to human readable form. - (cherry picked from commit c7cf926d25523d5d0d8e5c83671c901660070d71) + test: add focus and enter conversion testing. + + Doesn't actually convert anything, but verify the expected behaviour. -commit 60c161545af80eb78eb790a05bde79409dfdf16e -Author: Keith Packard -Date: Wed Feb 25 12:05:44 2009 -0800 +commit 59dc59a72ffd4cbc4df207bc688c92bb4863e8a9 +Author: Peter Hutterer +Date: Fri Aug 7 11:23:33 2009 +1000 - X server version 1.6.0 + test: add XI2 eventconversion test for raw events. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit e9a9f807971b8fb62b957c9d54b4aea7fb0f96ca -Author: Maarten Maathuis -Date: Fri Jan 30 19:32:33 2009 +0100 +commit 10d7948e0360860e1e9633dca39f646d492e73bf +Author: Peter Hutterer +Date: Fri Aug 7 10:17:14 2009 +1000 - exa: kill of exaImageGlyphBlt + test: fix build error introduced by new AllocDevicePair API - - It serves no obvious purpose, yet it directly accesses many fb - symbols. - (cherry picked from commit 5cc67ae94c066dcac78072ad8a819c3b602d8bab) + Signed-off-by: Peter Hutterer -commit 10e0603fb989d1e6b2675ef5ca8223777b4921e6 -Author: Jeremy Huddleston -Date: Tue Feb 24 23:12:57 2009 -0800 +commit 3f161a0aac39fbdeef393a17269486b8dace4672 +Author: Peter Hutterer +Date: Mon Aug 10 14:14:35 2009 +1000 - XQuartz: Re-enable support for capslock - (cherry picked from commit 4901b8147e593d26d7a31a9b73a201254b948916) + Xi: un-statify XI2EventSwap, it is needed for tests. + + Signed-off-by: Peter Hutterer -commit e96921ca954ff0d3de8a69cea085aac2d43b0a2e -Author: Michel Dänzer -Date: Mon Feb 23 16:49:51 2009 +0100 +commit 55e1ea08d03d89ecc0f2db7652a4d15567204696 +Author: Peter Hutterer +Date: Mon Aug 3 16:46:11 2009 +1000 - DRI1: Make DRICreateDrawable return TRUE for pixmaps. + Xi: get device changed data from DeviceChangedEvents, not the device (#23100) - GLX_EXT_texture_from_pixmap was broken since commit - a26c77ff432d2e85a2665fc36fca25143460c476 ('glx: fix retval checks when failures - occur for drawable creation.') + If a new device posts an event while the DCE is in the queue, getting the + data from the device may result in invalid memory access. - Signed-off-by: Michel Dänzer - (cherry picked from commit ef320bdd5ec3419abba77041d3a4d96a3ff87563) + X.Org Bug 23100 -commit 584f46b81c6ed126014e963cb4ea40f83014f1c7 -Author: Alan Coopersmith -Date: Thu Feb 19 13:57:28 2009 -0800 +commit 8ce4fde76fdf747fd3e00836c9c9c542e516ae9a +Author: Matt Turner +Date: Wed Aug 12 14:45:13 2009 -0400 - Add Extensions section to xorg.conf man page + Fix build on Alpha + +commit a400dbb38f93030d51afe806b4b20d5ef501c855 +Author: Colin Harrison +Date: Wed Aug 12 17:48:36 2009 +0100 + + Xming: Add NET_WM_ICON to native icon conversion - Extensions section was added in X11R6.8.0 and documented in the release notes: - http://www.x.org/archive/X11R6.8.0/doc/RELNOTES2.html#3 - but never made it into the man page. + fd.o bugzilla #4491 + originally from a patch by Joe Krahn - Also fix a bonus typo. + Convert a NET_WM_ICON to a native icon by converting to a native + bitmap and then using CreateIconIndirect() - Signed-off-by: Alan Coopersmith - (cherry picked from commit ab61033700b5383a7a15370dd054eaa80e72e811) + Don't use icon alpha on Windows 2000 or if display isn't 32-bit, convert + alpha channel to a 1-bit transparency mask using a threshold value + + Fix warning in winScaleXBitmapToWindows() about signedness of *iconData + + Signed-off-by: Jon TURNEY -commit ae7991b16d7fefb4335b05f951a3228cef102683 -Author: Keith Packard -Date: Wed Feb 25 10:14:36 2009 -0800 +commit 211511f150ce189681aff176c311fa8f312517e9 +Author: Soren Sandmann Pedersen +Date: Tue Aug 11 14:40:42 2009 -0400 - Pre-clip panning coordinates to keep crtc within panning region + Disable the out-of-bounds workaround in pixman. - There is a separate panning region check, but that doesn't work under - transformation, so just pre-clip the mouse coordinates when computing the - panning offsets. This leaves the case where panning constants are changing - unresolved. + This workaround was necessary for older X servers, but now fb should + be using correct coordinates. - Signed-off-by: Keith Packard - (cherry pick from commit c090f5514d28e1602a6ebbe7c909e98a0e3374d7) + Also bump pixman requirement to 0.15.20. -commit e59aa804df8a39541aa3224a5e688b819196070c -Author: Keith Packard -Date: Wed Feb 25 10:12:23 2009 -0800 +commit 93d9646c713336e03c135204c061f561d3654e23 +Author: Tiago Vignatti +Date: Sat Aug 8 20:18:10 2009 +0300 - RandR rotations and reflections offset by one pixel + mi: fix indentation - The matrix computation for rotation and reflection resulted in dropping a - row or column of pixels as the offsets used in the matrix computations used - width and height rather than width-1 and height-1. + mieq.c looks indented-wise now. Let's see how long it will take to someone + mess it again. - Signed-off-by: Keith Packard - (cherry picked from commit 77c7a64e8885696665556c9fbcb3cffb552e367a) + Signed-off-by: Tiago Vignatti -commit 9a59e7f304ab19c91738311bb9f1ec7709fc8847 -Author: Keith Packard -Date: Sun Feb 8 15:08:15 2009 +0100 +commit f9a2fff2248d7254958857677cabfea914ed4853 +Author: Tiago Vignatti +Date: Wed Aug 5 21:02:29 2009 +0300 - xf86CrtcShadowClear is unused. + mi: fix cursor warping screens - Remove this now that clearing is done by repainting with appropriate extend - modes. - Signed-off-by: Keith Packard - (cherry picked from commit 60a4f1368785d26a49a3ef6df829723ca154c154) + The server was processing ET_RawMotion type when the cursor was wrapping to + another screen and getting wrong valuator values. This fix such issue + considering only ET_Motion, ET_KeyPress, ET_KeyRelease, ET_ButtonPress and + ET_ButtonRelease types when the cursor detects a new screen, keeping the + "normal" processing of device events. - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti + Signed-off-by: Peter Hutterer -commit 369d7b22a9009e7054bd121484fa128c7a6d21f6 -Author: Keith Packard -Date: Thu Feb 5 11:36:54 2009 -0800 +commit d573cc46d3106824902ab4f926617bd9959af57c +Author: Maarten Maathuis +Date: Sat Aug 8 11:32:05 2009 +0200 + + exa: more safety + +commit 8b652435cd42929e2d187b353b3b20e798569356 +Author: Maarten Maathuis +Date: Sat Aug 8 10:35:01 2009 +0200 + + exa: minor cleanup + +commit a73f95aa93634cf7ba2c7c9274a046b2df2adca7 +Author: Maarten Maathuis +Date: Sat Aug 8 02:31:23 2009 +0200 - Make panning+transform be correctly driven by mouse + exa: Fix the broken upload fallback for "mixed" - Figuring out how to adjust the crtc origin to keep the mouse pointer within - the crtc is a bit of a trick + - Replace it with something wfb friendly while i'm at it. + +commit c029678bf7883b91f3c8095e764abbee246c9bb6 +Author: Julien Cristau +Date: Fri Aug 7 20:42:30 2009 +0200 + + configure.ac: drop dependency on fontenc - Signed-off-by: Keith Packard - (cherry picked from commit 63810aca31b962c93be4796883bde6ccb653e3a9) + Nothing in the server uses this anymore. + +commit 3943df7f76c1b4930fb7370d9d145cee96dd562b +Author: Maarten Maathuis +Date: Fri Aug 7 20:36:03 2009 +0200 + + exa: one can never be too careful + +commit 1548e8ae5d28ed0eb6057a1a19cfc84a78ef34dc +Author: Maarten Maathuis +Date: Fri Aug 7 20:04:53 2009 +0200 + + exa: fix a potential loophole in "mixed" - Signed-off-by: Keith Packard + - Always free sys_ptr before setting the pixmap to pinned. -commit c0a36197002c7de98f14dc98969409778d29dc50 -Author: Alan Coopersmith -Date: Tue Feb 17 18:48:52 2009 -0800 +commit e94c7c42ce9d6a194b76ba7e8b3904b180f93a1b +Author: Dave Airlie +Date: Fri Aug 7 13:36:52 2009 +1000 - Make RgbPath keyword in xorg.conf a non-fatal error + parser: make libxf86config_internal.la not installed. + +commit 5fb188b547e0ef4e2f2c59b66a3183b004e4a1e9 +Author: Dave Airlie +Date: Fri Aug 7 12:05:51 2009 +1000 + + ddx: fix xf86Config.a generation - Xorg shouldn't refuse to run just because the user has an xorg.conf that - had the previously-used RgbPath keyword in it. + We were generating a shared library, but this lib is foobar, the parser + requires some symbols from the X server or from the program its being linked + into. If the program its being linked into (say a python .so) has symbol + visibility enabled then it will fail to dynamic link, also if this .so has + symbol visiblity enabled it will fail to dynamic link. - Signed-off-by: Alan Coopersmith - Acked-by: Peter Hutterer - (cherry picked from commit d2cf562bbad553d7f09b70202134f5b6ada0114e) + Screw it go back to a .a file really unless someone cleans it up properly. - Signed-off-by: Keith Packard + Signed-off-by: Dave Airlie -commit 6b8bbbd21b39181e9dd3175fa73c05008c361ba2 -Author: Peter Hutterer -Date: Wed Feb 4 11:50:18 2009 +1000 +commit 3047bd067464efb9857960d3fa6324b947faa970 +Author: Maarten Maathuis +Date: Wed Aug 5 18:39:47 2009 +0200 - config: if we can't connect to HAL, listen for a startup notification. + exa: delay malloc for "mixed" + +commit e8ac2ed5dc4c2ac0a5e1e1f371f94c15b1c729dd +Author: Maarten Maathuis +Date: Wed Aug 5 16:12:16 2009 +0200 + + exa: implement exaMoveInPixmap for "mixed" - If HAL isn't available when we try to connect, the registered NameOwnerChanged - signal handler waits until HAL is available. Once we connected to HAL, we - unregister the signal handler again. - This allows HAL to be started in parallel or after the server has started. + - This can be used to force creation of driver pixmap. + - Not for 1 or 4 bpp. + - Driver can still fail (driver) pixmap creation. + +commit 9d2a7128d3e66b8c076a714d69f84bcad49391b9 +Author: Maarten Maathuis +Date: Sun Aug 2 02:35:46 2009 +0200 + + exa: Use damage to optimise away useless copies. + +commit 03ecb164f2592c954aa408bf121e0c67b604d854 +Author: Maarten Maathuis +Date: Sat Aug 1 19:19:19 2009 +0200 + + exa: A simple 3rd backend implementation. - Signed-off-by: Peter Hutterer - (cherry picked from commit 4844bff58f296b2851be4e6b955c3a68d02437a9) + - Based on driver pixmaps with some changes (completely transparent to driver). + - It helps with the problem of known software fallbacks, such as trapezoids. + - exaDoMigration is now called for all cases that provide a do_migration hook. + - exa_migration.c is renamed to exa_migration_classic.c + +commit ac7ac913fd98ea359c05c89968ab53a3223615b4 +Author: Maarten Maathuis +Date: Wed Jul 22 21:35:41 2009 +0200 + + exa: Split out some classic and driver allocated pixmap code into seperate files - Signed-off-by: Keith Packard + - Create a few seperate functions and a few private function pointers. + - Replace a few if conditions with a check for pExaPix->pDamage instead. + - This is in preperation of a third scheme that lies somewhere in between. + - Code clarity would have suffered (i started working on it and didn't like the mess). -commit 4557b3f6c4273cd83b701beaf7a150c806fed298 -Author: Keith Packard -Date: Wed Feb 18 16:43:23 2009 -0800 +commit 3abbd327f4a732408119de1f8e9ecba4812772a5 +Author: Peter Hutterer +Date: Wed Aug 5 15:47:46 2009 +1000 - X11/fonts/fontmod.h doesn't exist anymore. + dix: detach SD during XI2 grabs only. - libXfont 1.4.0 no longer includes this file, nor do we need it, so just stop - attempting to reference it. + XI1 grabs on slave devices leave the device attached - just like in earlier + versions of XI. - Signed-off-by: Keith Packard + Tested-by: Thomas Jaeger + Signed-off-by: Peter Hutterer -commit 3575cb86a18d9e80e239acefdab18694e4ca2098 -Author: Ian Romanick -Date: Tue Feb 17 08:27:32 2009 -0800 +commit c29aa7da220661532b05972cacd3dbaff29408b5 +Author: Peter Hutterer +Date: Wed Aug 5 15:33:07 2009 +1000 - glx: Inialize best_score before calculating visual scores - - This bug was pointed out by Peter Åstrand. + Revert "dix: Remove temporary detachment of slave devices." - Signed-off-by: Ian Romanick - (cherry picked from commit 4039603413f9f46d7f725463a70b4a51838e0049) + This reverts commit 0c0ef42292f4c910c73b308cd75d77637312da53. - Signed-off-by: Keith Packard + Tested-by: Thomas Jaeger + Signed-off-by: Peter Hutterer -commit d050a61747bd5e5f9c96f5df5a191598c3f01e5c -Author: Jeremy Huddleston -Date: Tue Feb 17 23:25:14 2009 -0800 +commit 931160fcf378120e58849d801dbc62bed6e65a4b +Author: Jeremy Huddleston +Date: Tue Aug 4 02:53:17 2009 -0700 - configure.ac: Support version strings like W.X.Y.Z-XXXXXX - (cherry picked from commit 43967514cd57ad836d7fb85c8c9e58ada07e0232) - - Signed-off-by: Keith Packard + XQuartz: AIGLX: Provide empty __glXAquaDrawableResize to avoid crashing in DoMakeCurrent + (cherry picked from commit 60a1d2c2764f2f02c0751940a264588717afce79) -commit 8e16e64f55deb8e94007915333ba356c0f7e4f73 -Author: Jeremy Huddleston -Date: Tue Feb 17 23:24:25 2009 -0800 +commit 0fe639a314de397516cf00d8100da8086abdd97a +Author: Jeremy Huddleston +Date: Sat Aug 1 22:53:45 2009 -0700 - XQuartz: Don't need GlxSetVisualConfig any more + XQuartz: Only save lastpt on mouse/tablet events + (cherry picked from commit 552be074e5be6492df3e290e8b7d9daff1a2cb34) + +commit 491c2d74c12d7f9bc9f31b427ece04e049ebff5a +Author: Jeremy Huddleston +Date: Sat Aug 1 12:43:14 2009 -0700 + + XQuartz: Use mouseLocation rather than locationInWindow when setting lastpt - Fixes build failure resulting from 516f8e2cad1311a09764e2633644188d1e3c31bb - (cherry picked from commit 066b17028a35956a089815716e38571f305469c5) + I don't understand the *why* ... I just see that it works better this way for games like Quake2 through wine. It *should* be better the other way, but somehow it's not. - Signed-off-by: Keith Packard + I guess this will go in my list of puzzles to unravel. + (cherry picked from commit 65ae2d00e1a53f97f2ff9522406ab69d50bf3199) -commit eca523e7e7e267728a1f5ec346e012ada3debba0 -Author: Keith Packard -Date: Tue Feb 17 22:17:41 2009 -0800 +commit df6c01352471c766e9f71ceac03cab8c3911faa1 +Author: Jeremy Huddleston +Date: Sat Aug 1 02:32:52 2009 -0700 - Bump to 1.5.99.903 + XQuartz: Don't use location delta for tablets since NSEvent does not give a precise delta. + (cherry picked from commit 6c5bf756a7f5389cdfe2e43a339d7c31a3e522e9) -commit a86dd529e9deaba61aee220d79ac607abbcb5dc6 -Author: Keith Packard -Date: Fri Feb 6 06:34:35 2009 +0100 +commit e360104880e6e2e666aa05dfd56e2ef3880f38ef +Author: Jeremy Huddleston +Date: Wed Aug 5 16:17:01 2009 -0700 - Eliminate the shadow clear on transform change - - When the crtc transformation changes, the entire crtc must be repainted. - This was being done by clearing the shadow and then painting the rectangle - containing the screen image; the clear being required as the screen image - may not fill the crtc. When changing the transform rapidly, this leads to - flashing. Eliminate the clear by painting the entire crtc instead of just - the screen rectangle. - - (cherry picked fom commit 5394b7e66224d20888dd4020f5cb8ca930720fb4) - - Signed-off-by: Keith Packard + XQuartz: Purge redundant QuartzBell + (cherry picked from commit de14a63d20095e1537fd74352850c734d900031d) -commit 5946d603e76206105bea41a8beeb116daf209964 -Author: Keith Packard -Date: Mon Feb 9 13:09:11 2009 +0100 +commit 8c0085c715effdc450d78eec14bc32e6214c78af +Author: Tiago Vignatti +Date: Wed Aug 5 22:50:44 2009 +0300 - Add XkbDir to Files config file section + xfree86: remove more RAC junky - The XKB base directory was not configuable through the config file. + Signed-off-by: Tiago Vignatti + +commit 444723273ae82fc3f6707ed1a461aaaa8fdae39c +Author: Tiago Vignatti +Date: Wed Aug 5 22:48:02 2009 +0300 + + xfree86: reorganize pci code - (cherry picked from commit 76f18b94bd2719a8199334742d021c4d0806187d) + - xf86PciVideoInfo doesn't need to be global + - remove unused macros - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti -commit 39eb1bee84b520cfa45981c445c4c976d63ab03d -Author: Adam Jackson -Date: Tue Feb 17 12:59:16 2009 -0500 +commit 20169414e1afd5d1d02cb1b57866b1c158b2fc6c +Author: Tiago Vignatti +Date: Wed Aug 5 22:44:19 2009 +0300 - RANDR: Fail softly on GetPanning if the screen can't do it. + xfree86: remove unused functions - Just return a zeroed-out reply in that case. This is unambiguous, and - distinguishes "you didn't name a CRTC" from "you named a CRTC that can't - do panning". - (cherry picked from commit 123093996507c4d3b6dc457240ce00f8ac42f410) + RAC trash. - Signed-off-by: Keith Packard + Signed-off-by: Tiago Vignatti -commit 69a2728891e92da6a674d7313f3bb47c629b38dc -Author: Paulo Cesar Pereira de Andrade -Date: Wed Jan 7 19:37:03 2009 -0200 +commit 95b678e6dc41f2524ada4eb11289687fafce7588 +Author: Alan Coopersmith +Date: Mon Aug 3 23:49:56 2009 -0700 - Default to use standard bitmap fonts, with builtins as fallback + Correct modifier map built when ProcSetModifierMapping is called - The builtin-fonts configure option was removed, as it at best should - have been a runtime option. Instead, now it always register all "font - path element" backends, and adds built-ins fonts at the end of the - default font path. - This should be a more reasonable solution, to "correct" the most - common Xorg FAQ (could not open default font 'fixed'), and also don't - break by default applications that use only the standard/historical - X Font rendering. - (cherry picked from commit 49b93df8a3002db7196aa3fc1fd8dca1c12a55d6) + Fixes xmodmap changes to modifiers to stop corrupting modifier maps - Signed-off-by: Keith Packard + Previous code had two bugs: + - the code to increment mod was after the code to continue if no + modifier was set, so mod wouldn't be incremented for modifiers + with no keys mapped to them (such as if you called + xmodmap -e 'clear Lock') + - the value it set in the modifier map was the raw modifier number, + not the bitmask value for that modifier + + Signed-off-by: Alan Coopersmith + Signed-off-by: Peter Hutterer -commit 444127f9f408d2f517fdfab0092bd67b29073373 -Author: Alan Hourihane -Date: Fri Feb 13 13:45:08 2009 +0000 +commit 4ca305956e5ea6f606b22ef62aa462186a7b95f0 +Author: Peter Hutterer +Date: Wed Aug 5 09:40:56 2009 +1000 - glx: fix retval checks when failures occur for drawable creation. - (cherry picked from commit a26c77ff432d2e85a2665fc36fca25143460c476) + Re-enable Xnest by default. - Signed-off-by: Keith Packard + Xnest was disabled in 82fc102568b3d6b0daeb6c5f5b3a1310a7f14fcd due do build + errors. These errors have since been fixed. + + Re-enable Xnest by default to increase coverage by default builds. + + Signed-off-by: Peter Hutterer -commit 4302a32391711ea86051c7e6d90cb018c79beec2 -Author: Keith Packard -Date: Fri Jan 30 14:19:41 2009 -0800 +commit 5904ef2ccd6056b187ca76f104c21e2d686bfc1d +Author: Peter Hutterer +Date: Wed Aug 5 10:40:20 2009 +1000 - Make crtc_notify wrap/unwrap code do nothing unless mode code is inuse + xnest: restore xnestUpdateModifierState - Drivers not using the new hw/xfree86/modes code would crash in DRI due to - that code trying to monitor CRTC changes. + The meat of xnestUpdateModifierState was ifdef'd out in + 6ef46c40e62def4841a4cff4e0b443516a2ed782. This resulted in stuck modifiers + when a modifier key release event wasn't sent to Xnest (e.g. Alt-Tab away). - (cherry picked from commit ea309e47457156b60aadbf113f04e5b6851029c8) + See X.Org Bug 3664 for + the original bug report. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 09cde70a6b3a05c28014862db19b36650412c2f3 -Author: Jeremy Huddleston -Date: Mon Feb 2 12:30:57 2009 -0800 +commit 44afc7577ad3339f4555438388f6a4d8e220c991 +Author: Peter Hutterer +Date: Wed Aug 5 09:51:05 2009 +1000 - XQuartz: GLX: Fix accidental exclusion of glBlitFramebufferEXT on Leopard - (cherry picked from commit 8620579483789ae9ac1b68dbefe32b70011386de) - (cherry picked from commit 0dbc356795bbab3889b5f1684f55bd193757d0c9) + xnest: include exevents.h for XIGetKnownProperty. - Signed-off-by: Keith Packard + Silences compiler warning. + Pointer.c: In function ‘xnestPointerProc’: + Pointer.c:64: warning: implicit declaration of function ‘XIGetKnownProperty’ + Pointer.c:64: warning: nested extern declaration of ‘XIGetKnownProperty’ + + Signed-off-by: Peter Hutterer -commit 6bbe98f01d1fc66b6b853bdcdd62edf87a91627d -Author: Jeremy Huddleston -Date: Sun Feb 1 15:08:02 2009 -0800 +commit 32ce5c83a504d2cafd2b19ed628b40032ef347b3 +Author: Peter Hutterer +Date: Wed Aug 5 09:39:47 2009 +1000 - XQuartz: GLX: OpenGL.framework on Tiger doesn't have glBlitFramebufferEXT - (cherry picked from commit e01662cadcaa4052e3a5aa82cbaed00a2d0220e0) - (cherry picked from commit fd08be749e0b3c5de02a6ae8b3d21f92c5986157) + xnest: use AllocDevicePair for Xnest device initalization - Signed-off-by: Keith Packard + All Xnest needs is a single pointer+keyboard pair. AllocDevicePair sets them + up nicely with the name assigned etc. + + Signed-off-by: Peter Hutterer -commit 519ae3a0dfc61fc55370abec75551aa2615643fb -Author: Jeremy Huddleston -Date: Sun Jan 11 16:34:36 2009 -0800 +commit 26b83ad4a29dc180b336a19d9e97589814e93e37 +Author: Peter Hutterer +Date: Wed Aug 5 09:32:50 2009 +1000 - XQuartz: Don't need explicit Activate/EnableDevice in InitInput - (cherry picked from commit b1d29784410b3b93037e5636f336ba608d8ad6e3) + dix: require PointerProc and KeyboardProc to be passed into AllocDevicePair. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit ada58b69ca0a5e674cda7ae59a53061645621888 -Author: Alan Coopersmith -Date: Fri Jan 30 21:44:08 2009 -0800 +commit 6a500fdd4d717947b86f6d6844ebf9a8603eb8e2 +Author: Peter Hutterer +Date: Tue Aug 4 15:29:55 2009 +1000 - x11-input.fdi: Add options needed to handle adding USB devices on Solaris - (cherry picked from commit 60bcdd687040db76490851d4b459284ce37020e0) + dmx: include exglobals.h for DeviceKeyPress and friends. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 516f8e2cad1311a09764e2633644188d1e3c31bb -Author: Eric Anholt -Date: Sun Feb 8 13:00:15 2009 +0100 +commit 591c2f8758ef36a9bbf1625f9c207a1804f58ef4 +Author: Peter Hutterer +Date: Tue Aug 4 15:30:52 2009 +1000 - glx: Replace broken GLX visual setup with a fixed "all" mode. - - With trying to match depths so that you didn't end up with a depth 24 - fbconfig for the 32-bit composite visual, I broke the alpha bits on the depth - 24 X visual, which angered other applications. But in fixing that, the - pickFBconfigs code for "minimal" also could end up breaking GLX visuals if - the same FBconfig was chosen for more than one X visual. - We have no reason to not expose as many visuals as possible, but the old - "all" mode didn't match any existing X visuals to GLX visuals, so normal - GL apps didn't work at all. - - Instead, replace it with a simple combination of the two modes: Create GLX - visuals by picking unique FBconfigs with as many features as possible for - each X visual in order. Then, for all remaining FBconfigs that are - appropriate for display, add a corresponding X and GLX visual. + dmx: typecast to InternalEvent* before mieqEnqueue. - This gets all applications (even ones that aren't smart enough to do FBconfigs) - get all the options to get the visual configuration they want. The only - potential downside is that the composite ARGB visual is unique and gets a - nearly full-featured GLX visual (except that the root visual might have taken - the tastiest FBconfig), which means that a dumb compositing manager could - waste resources. Write compositing managers using FBconfigs instead, please. - (cherry picked from commit c40bad88475debf7a1774e22dce835577ad32154) + Silences a few compiler warnings. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 8b967b24690cb072fc37c463eceb2b886cef80c4 -Author: Eric Anholt -Date: Mon Feb 2 10:13:46 2009 -0800 +commit 15b425b5a6630d2ac95b490f4f4dbb9240ac9c5d +Author: Peter Hutterer +Date: Tue Aug 4 15:31:42 2009 +1000 - glx: Don't match fbconfigs to visuals with mismatched channel masks. + dmx: move 'state' around to silence compiler warning. - This fixes at least one known bug, where the depth 32 visual would end up - with a depth 24 fbconfig attached, angering compiz. - (cherry picked from commit 5100d829a4d71ce4a9fbc2b81694a1fb90066ccf) + Signed-off-by: Peter Hutterer + +commit e10f802c8260e034fd6c835d3f2622d9cd5085f8 +Author: Peter Hutterer +Date: Tue Aug 4 15:40:17 2009 +1000 + + dmx: include inputstr.h for inputInfo.pointer - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 7fdaca64463951ab9199dc12910379c90dba3fac -Author: Keith Packard -Date: Tue Feb 17 18:57:13 2009 -0800 +commit 9ffc3f280b3907a672f99101548a5eb52027b37f +Author: Peter Hutterer +Date: Tue Aug 4 15:40:31 2009 +1000 - mi: force the paired kbd device before CopyKeyClass. (#19574) + dmx: don't call InitKbdFeedbackClassDeviceStruct - doesn't exist anymore. - Some multi-media keyboards send the key events for multimedia keys through - the device file used by the mouse. Sending a key event through the VCP - however will fail. The VCP doesn't have a key class so the server crashes or - (with an appropriate fix) the event is simply swallowed. + InitKeyboardDeviceStruct does the same task now. - Thus, for key events if the master does not have a key class, get the device - paired with the master (i.e. the VCK) before processing the event any - further. + Signed-off-by: Peter Hutterer + +commit 63155cf985ee15bf6aad95066b076ab680cf5a31 +Author: Peter Hutterer +Date: Tue Aug 4 15:28:30 2009 +1000 + + dmx: pass button/axis labels into the init functions (currently unset) - X.Org Bug 19574 + Currently only None labels are passed in, in the future these labels should + be whatever the respective buttions/axes are. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit a9dd41de7eecc1dcede4ccc8d5cf25f10343bb1a +commit af85973a0a80fb0ba82330ae3ec897f65d618003 Author: Peter Hutterer -Date: Mon Feb 2 10:20:13 2009 +1000 +Date: Tue Aug 4 15:25:06 2009 +1000 - mi: don't call UpdateSpriteForScreen if we have Xinerama enabled. #18668 - - In Xinerama all windows hang off the first root window. Crossing the screens - must not reset the spriteTrace, otherwise picking fails and events are sent to - the root window. + dmx: remove now-useless defines. - X.Org Bug 18668 + XI is mandatory so we can expect both to be defined. + +commit 654d7a02600bbafd890f555a081aa0a2fcb5e253 +Author: Peter Hutterer +Date: Tue Aug 4 15:23:01 2009 +1000 + + Revert "dmx: claim we support XI 2." - Signed-off-by: Peter Hutterer - (cherry picked from commit 9fe9b6e4ef669b192ee349e3290db5d2aeea273c) + XQueryInputVersion doesn't exist anymore and for now XI 1 support is good + enough. - Signed-off-by: Keith Packard + This reverts commit 8da8a0fec4b1b9d9208635dedb2f449dc99e0004. -commit 536748aa88f82e4157b86ed1a02b798b1510bdb8 -Merge: 8c6ddbf be21199 -Author: Keith Packard -Date: Tue Feb 17 18:52:04 2009 -0800 +commit e7c950c6273ff93f25c091c93d86da6e332a2277 +Author: Peter Hutterer +Date: Tue Aug 4 15:15:24 2009 +1000 - Merge branch 'panning-for-server-1.6' into server-1.6-branch + dmx: fix a call to XkbSetRulesDflts, takes XkbRMLVOSet now. -commit 8c6ddbf754ff6c2eb99105f93aaa30749dced619 -Author: Paulo Cesar Pereira de Andrade -Date: Sun Nov 30 02:59:34 2008 -0200 +commit 2cde9208ff756e33d162e2324f4b99540230d743 +Author: Peter Hutterer +Date: Tue Aug 4 14:58:25 2009 +1000 - Correct static symbol XkmReadTOC and first pass on compile warning fixes. + dmx: Enable/DisableDevice take a boolean variable now. - The warnings corrected were only the ones that should correct - real problems. The most common one is 64 bit integers as - "printf %l" arguments. - Note that there is a patch related to this at: - http://bugs.freedesktop.org/show_bug.cgi?id=18204 + We want to send events here, so pass in TRUE. - Signed-off-by: Keith Packard - (cherry picked from commit 16b11cd03d8c5def07f0e598f237f71a37883a46) + Signed-off-by: Peter Hutterer -commit 0c9db0ea10d9fd2a311bca47bba3c60e74ec6eec +commit 1e210d6d10ec63d22247e5801890e024826fc861 Author: Peter Hutterer -Date: Fri Jan 9 14:32:14 2009 +1000 +Date: Tue Aug 4 14:41:46 2009 +1000 - xfree86: always force RAW mode under linux. + xkb: remove now-unused XkbGetKeysym. - The previous check for AEI on left us with the possibility that AEI is forced - off in the config, but devices are added through evdev nonetheless. A keyboard - added this way can CTRL+C the server. Even when we use kbd, we can set the - mode to RAW, so it's safer alround to to so. + XkbGetKeysyms was only used by the now-removed Keysym grabs. Signed-off-by: Peter Hutterer - (cherry picked from commit b33905234025f005819c7e2acd653a3a0ecfeb82) - - Signed-off-by: Keith Packard -commit 3284b03dca180ba2312904b1e20fe83df56f5f3c +commit a6ce6c70cff5108f0751b662b8e52c83daab0722 +Author: Michel Dänzer +Date: Tue Aug 4 23:23:21 2009 +0200 + + EXA: Simplify exaGetPixmapFirstPixel using GetImage. + +commit 842373104d08d47efc863cecbe30431d3faebef1 +Author: Michel Dänzer +Date: Tue Aug 4 23:23:21 2009 +0200 + + Add support for RENDER BGRA formats. + +commit 0bf7eaf3052ce24066b0a7c14860b4762fb81364 Author: Alan Coopersmith -Date: Thu Feb 12 18:34:22 2009 -0800 +Date: Mon Aug 3 23:14:50 2009 -0700 - Check for and report errors writing xorg.conf.new from Xorg -configure + Ansify function arguments in VTsw_noop.c Signed-off-by: Alan Coopersmith - (cherry picked from commit ac470dfb4fadaa0b28b6f8b57f4f13a20842b897) + +commit 736f2d64725c6df8413e627bd40ce7ecb011acc7 +Author: Paul Bender +Date: Tue Jun 30 15:50:46 2009 -0800 + + Bug 16832: XDMCP related build error when --disable-xdmcp is used - Signed-off-by: Keith Packard + X.Org Bugzilla #16832: http://bugs.freedesktop.org/show_bug.cgi?id=16832 + Patch #27279: http://bugs.freedesktop.org/attachment.cgi?id=27279 -commit caf503aef15865a969b881355d5e663490a6e194 +commit b159a98f7fbdb3f8b337cd1bfd2ff27312856619 +Author: Lee Leahu +Date: Thu Oct 23 21:03:57 2008 -0500 + + seg fault when initializing DMX screens + + X.Org Bugzilla #18086: http://bugs.freedesktop.org/show_bug.cgi?id=18086 + Patch #19837: http://bugs.freedesktop.org/attachment.cgi?id=19837 + +commit 5da9b255a64bec7dbf5ddb392d54dac9be5b43c0 Author: Alan Coopersmith -Date: Thu Feb 12 18:23:59 2009 -0800 +Date: Mon Aug 3 19:53:27 2009 -0700 - Correct error message if specified config file is not found + Fix blddir != srcdir builds of solaris-*.il files - By making the "Unable to open config file" header a warning, it was - not appearing with the filename when a config file was specified and - not found. Now we make it an error message again, but only issue - the error if a filename was specified - if none was specified, then - we don't even issue a warning, just the "Using autoconfig" info message. + Fixes http://bugs.freedesktop.org/show_bug.cgi?id=17509 Signed-off-by: Alan Coopersmith - (cherry picked from commit bd713794ceaa1b2890522554562103c0a2d50f04) - - Signed-off-by: Keith Packard -commit 791c6ecceec94369f6035b8845a2cb293d194eef -Author: Keith Packard -Date: Tue Feb 3 20:18:37 2009 -0800 +commit 1e69fd4a60147287b31e53bfc61543fb17bb82c8 +Author: Peter Hutterer +Date: Mon Aug 3 12:03:26 2009 +1000 - Handle the combination of panning and crtc transforms + xnest: fix up parameters to InitKeyboardDeviceStruct. - This patch gets the shadow scanout buffer repainted on panning area changes. - It does not, however, track the mouse correctly. + IKDS takes a DeviceIntPtr as first argument, and an RMVLO struct as second. + The keysyms stuff is long gone now. - Signed-off-by: Keith Packard - (cherry picked from commit 7968823cbc02615e1080c0d3f34dcebe14ea1771) + Signed-off-by: Peter Hutterer + +commit 008658049cbeea35a9f76f98037aa2f4173f3573 +Author: Peter Hutterer +Date: Mon Aug 3 12:00:45 2009 +1000 + + xnest: silence compiler warnings by typecasing properly. + + Events.c: In function ‘xnestQueueKeyEvent’: + Events.c:112: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible + pointer type + ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of + type ‘struct xEvent *’ + Events.c: In function ‘xnestCollectEvents’: + Events.c:141: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible + pointer type + ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of + type ‘struct xEvent *’ + Events.c:150: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible + pointer type + ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of + type ‘struct xEvent *’ + Events.c:160: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible + pointer type + ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of + type ‘struct xEvent *’ + Events.c:193: warning: passing argument 2 of ‘mieqEnqueue’ from incompatible + pointer type + ../../mi/mi.h:203: note: expected ‘union InternalEvent *’ but argument is of + type ‘struct xEvent *’ - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 099d74a12752808dd1972c28bdb228fc447d3774 -Author: Keith Packard -Date: Tue Feb 3 20:15:41 2009 -0800 +commit 664ac92d8bbe956dd6fd80fac5dc3161028803b2 +Author: Peter Hutterer +Date: Fri Jul 31 14:38:35 2009 +1000 - Damage re-used shadow scanout buffer using new transforms. + xfixes: backup the DisplayCursor/CloseScreen proc before restoring it (#23034) - When the shadow scanout buffer can be re-used, the underlying framebuffer - area must be damaged so that the scanout will be repainted. This patch - delays the addition of that damaged area until after the transform in the - crtc has been updated, otherwise the old transform would have been used and - the wrong area repainted. + The screen's DisplayCursor func is wrapped as + AnimCurDisplayCursor -> CursorDisplayCursor -> miPointerDisplayCursor. - Signed-off-by: Keith Packard - (cherry picked from commit 763df9eec79a867978efc5138d8635a46c1a0d17) + Calling CursorDisplayCursor while an animated cursor was currently displayed + would remove AnimCurDisplayCursor from the wrap stack. Thus, the next call + to ChangeToCursor wouldn't update the animated cursor state. The block + handler for animated cursors would then continuously overwrite the actual + cursor, leaving an animated cursor everywhere on the screen. - Signed-off-by: Keith Packard + X.Org Bug 23034 -commit 4992c31e6c09a9140d4482d762b5dfff0a4efedb -Author: Keith Packard -Date: Tue Feb 3 20:12:42 2009 -0800 +commit f48dfcc1b772a09428e328c72124ea0d46a73416 +Author: Peter Hutterer +Date: Fri Jul 31 10:44:28 2009 +1000 - Handle matrix computation overflow in RRTransformCompute - - If the computation of the composite fixed-point transform for RandR - overflows at any point, take the resulting floating point transform and - scale that back to fit in a fixed point matrix. This ensures that a matrix - will always be available, although perhaps at reduced precision. Someday we - should add floating point matrices to Render. + xfixes: allocate CurrentCursor for all devices. - Signed-off-by: Keith Packard - (cherry picked from commit 62fc98cb88e4e8b636f343453fc1168a87c58972) + XFixes requires cursor notifies to clients when the cursor changes. This + should work on the ClientPointer and then on all master pointers. Hence + change CurrentCursor to a MAXDEVICES array. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 660c2a7d4c6cb52fd0992e9d2008a24805dc8c02 -Author: Aaron Plattner -Date: Thu Feb 12 18:27:48 2009 -0800 +commit d3e5629fac224d53a972df5e2a87db02534f9b17 +Author: Peter Hutterer +Date: Thu Jul 30 16:36:56 2009 +1000 - Query the DDX in ProcRRGetScreenInfo. + dix: call SetFocusOut and LeaveWindow when disabling a device. - This is the RandR 1.1 version of GetScreenResources and needs to re-query the - DDX to see if the mode pool changed. + PointerWindows[x] would be set after removing a master pointer. Destroying + this window then crashed the server. - Fixes Launchpad bug #325115. - Signed-off-by: Adam Jackson + Signed-off-by: Peter Hutterer -commit 34724f45512e311377b972da480d35304fd687cc -Author: Paulo Cesar Pereira de Andrade -Date: Fri Jan 30 03:56:39 2009 -0200 +commit 97ed946cfccec5979c47b5fabf1ad56cc7d33ef4 +Author: Peter Hutterer +Date: Thu Jul 30 16:36:30 2009 +1000 - Work around inclusion of + dix: SetFocusOut and LeaveWindow don't need parameters other than dev. - The X Server build only needs the macros PANORAMIX_MAJOR_VERSION - and PANORAMIX_MINOR_VERSION from that header. - Addition of extra prototypes to - caused a X Server build failure. - -commit 1f729b42d567ae9533ac0e467afc9fbc83390776 -Author: Keith Packard -Date: Fri Jan 30 21:12:17 2009 -0800 - - Version to 1.5.99.902 + Signed-off-by: Peter Hutterer -commit 56f6fb8c8652c85e522e42557f8969987069076b -Author: Eric Anholt -Date: Fri Jan 30 19:06:17 2009 -0800 +commit d040940efe041b57e6323921c380ceb2bb43f41e +Author: Peter Hutterer +Date: Thu Jul 30 16:16:51 2009 +1000 - randr: Avoid re-querying the configuration on everything but GetScreenResources. + xfixes: FixesHideCursor should work on all master pointers. - The new path should only re-query on the other requests when we haven't - gathered the information from the DDX yet (such as with a non-RandR 1.2 DDX). + Presumably, a client calling HideCursor doesn't want any cursor displayed. - Bug #19037. - (cherry picked from commit 317f2b4a9fe4b606975711bc332166a82db5087d) + Signed-off-by: Peter Hutterer -commit 6d928aa227b12832dd358619722a5f4dc71b5d19 -Author: Keith Packard -Date: Fri Jan 30 20:17:10 2009 -0800 +commit b83c682e7ea4e10757eca6363de28281108591a8 +Author: Oliver McFadden +Date: Mon Aug 3 02:46:57 2009 +0300 - RandR crtcs not fetched correctly when primary output is set. + xf86Xinput: xf86PostButtonEventP must set POINTER_ACCELERATE for relative events. - Primary outputs may not have a CRTC. - Loops fetching CRTCs respecting primary output were broken. - - Signed-off-by: Keith Packard - (cherry picked from commit 4de5705545792d77c9990bbec3e263a517757a2a) + Thanks to Peter Hutterer for spotting this bug. -commit 759ed0679a030f502b74a6ff44c15d1c97d9a937 -Author: Keith Packard -Date: Fri Jan 30 19:14:36 2009 -0800 +commit ae65daab080c6269f0eadc6be1a48bf27866c542 +Author: Jeremy Huddleston +Date: Fri Jul 31 15:32:07 2009 -0700 - RandR SetOutputPrimary should work with or without a crtc connected - - The test was inverted from the protocol spec, and besides, the test is bogus - anyways. - - Signed-off-by: Keith Packard - (cherry picked from commit c1f2be1f3fd0c80cb4c85b98140b98aa9311242b) + XQuartz: Unify how we set our bitmasks for visuals + (cherry picked from commit c230b52c27ce50ac6c27011ec8e88f1b263b25f9) -commit b893dc59da9f5a7afb59ea0b2cd794872fc5727f -Author: Thomas Jaeger -Date: Sat Dec 20 16:17:02 2008 +0100 +commit 4a53fed41d711f270c7d8a7fbeaafb4fb619e879 +Author: Jeremy Huddleston +Date: Fri Jul 31 14:19:50 2009 -0700 - Don't release grabs unless all buttons are up - - Previously, only buttons <= 5 would count here, but the core protocol - allows for 255 buttons. - - http://lists.freedesktop.org/archives/xorg/2009-January/042092.html + XQuartz: Dead code removal for StaticColor visual + (cherry picked from commit e457a44e87950207f6c16bf82bf5af11c777ecf1) + +commit da0a2747803dbecb1308181176a784e14cf02809 +Author: Jeremy Huddleston +Date: Fri Jul 31 14:14:21 2009 -0700 + + XQuartz: Cleanup the bitmask setting for GLX visuals. + (cherry picked from commit d32c3df258e748958ef997c675dc4fae118c0d7b) + +commit 70ac671af2f092471022590cb7a19ef9155c51b4 +Author: Jeremy Huddleston +Date: Thu Jul 30 01:30:52 2009 -0700 + + XQuartz: no DirectColor + (cherry picked from commit df2fbc410f2c484612f65a6539a6cb069ef4a468) + +commit c0c72a866a237d3c2d9e69e1c69181ef5446e3f8 +Author: Jeremy Huddleston +Date: Thu Jul 30 00:51:55 2009 -0700 + + XQuartz: Define DDXRingBell + (cherry picked from commit 113347381289497cb2a79994d0ef5f427ae63ac5) + +commit 949811313989809f5b58424af04b9fd8e1d0bec4 +Author: Peter Hutterer +Date: Thu Jul 30 21:59:40 2009 +1000 + + test: fix build by including eventstr.h Signed-off-by: Peter Hutterer - (cherry picked from commit 717a961528ec69a6e630d536e15568670e0b398a) - - Signed-off-by: Keith Packard -commit 191161a24a80dc553b11f829c12cfffb76ec7185 -Author: Thomas Jaeger -Date: Mon Dec 22 00:55:09 2008 +0100 +commit 0565f4ed4519962bed40a0bbcf0b409471f4de40 +Author: Peter Hutterer +Date: Wed Jul 29 16:46:45 2009 +1000 - Count the number of logically down buttons in buttonsDown + Xi: set the sourceid for focus devices to the device id. - This fixes the following bug. Assuming your window manager grabs - Alt+Button1 to move windows, map Button3 to 0 via XSetPointerMapping, - then press the physical button 3 (this shouldn't have any effect), press - Alt and then button 1. The press event is delivered to the application - instead of firing the grab. + Unlike Enter/Leave events generated by a device pushing the pointer around, + a device doesn't change focus all by itself. It's a result of a + SetInputFocus call, a window becoming unviewable or a grab activating. As + such, the sourceid for focus events is always the deviceid itself. Signed-off-by: Peter Hutterer - (cherry picked from commit f7f85f696570541e2dd43462675de9e6ee46f545) -commit fe0abdceb483ec0bb2977016b00aef57033aa449 +commit de4dd5848cab90b0f8b8243ca0b49985ef047124 Author: Peter Hutterer -Date: Thu Jan 8 11:53:30 2009 +1000 +Date: Wed Jul 29 16:45:34 2009 +1000 - mi: ensure chained button mappings from SD -> MD (#19282) - - After copying the master event, flip the detail field to the mapped button of - the SD, not the physical button. This way if the SD has a mapping 1:3 and the - MD has a mapping of 3:4, a press on button 1 on the SD results in a core event - on button 4. - - X.Org Bug 19282 + include: DeviceFocusEvent is not to be exported. Signed-off-by: Peter Hutterer - Signed-off-by: Keith Packard -commit 3441917008021fb13cf86fa619970859d927e17a -Author: Thomas Jaeger -Date: Mon Jan 5 01:26:42 2009 -0500 +commit 46ac9f92416f3cb99b5d84a9d200237dc33a3bb7 +Author: Peter Hutterer +Date: Wed Jul 29 16:34:28 2009 +1000 - Don't alter device button maps in DoSetPointerMapping + dix: pass the sourceid around for enter/leave events. - Currently, if a device map differs from the core pointer map, then the - request may return MappingBusy, even though all the affected core - buttons are in the up state. + The sourceid for enter/leave events as a result of pointer motion is the ID + of the slave device. The sourceid for those as a result of a grab activating + is the device itself. Signed-off-by: Peter Hutterer - (cherry picked from commit 3d549438c29004d78032ecc50ab45ca0e3f49623) -commit 364bcbea48d5e49bef8d36f6724e2702cbb4b1af -Author: Thomas Jaeger -Date: Fri Jan 9 02:02:38 2009 -0500 +commit c299b2228fb63c192b72851c90e14ceaceb67bbc +Author: Peter Hutterer +Date: Wed Jul 29 16:00:19 2009 +1000 - Xext: Send out correct events in ProcXTestFakeInput + dix: remove obsolete comment, parameter described doesn't exist. Signed-off-by: Peter Hutterer - (cherry picked from commit b2756a71a432f7cf7c870a48676c98625512558d) -commit e789789196db8271a83c751fdf990ceb70164cb1 +commit 5085ac09a50721d87196bd9f2607dc76200ca399 Author: Peter Hutterer -Date: Mon Jan 5 15:48:45 2009 +1000 +Date: Wed Jul 29 13:45:32 2009 +1000 - xkb: fix typo - missing negation when checking button state. + input: switch internal event types to enums. - Introduced with a85f0d6b98237d8a196de624207acf1983a1859a. - - Reported by Thomas Jaeger. + Use enum EventType instead of ints. This requires a load of default + cases in various switch statements to silence compiler warnings. + Reported-by: Aaron Plattner Signed-off-by: Peter Hutterer - (cherry picked from commit 515ce3e4ba42605a1ee9979e8bb5acd3cf6470a3) -commit ebb3872925315d705e47e1ccd7f193ac90ac33dc -Author: Peter Hutterer -Date: Wed Nov 26 11:15:05 2008 +1000 +commit 1ae8332d643299a3ee9a9f45a8e25b8c87c751e1 +Author: Peter Hutterer +Date: Wed Jul 29 13:39:38 2009 +1000 - Xi: fix use of button->down - bitflags instead of int arrays. - - The device's button down state array was changed to use DOWN_LENGTH and thus - bitflags for each button in cfcb3da7. - - Update the DBSN events to copy this bit-wise state. - Update xkb and Xi to check for the bit flag instead of the array value. + include: fix enum EventType declaration. - Reported by ajax. + Having EventType after the enum declares a variable. silly me. - Signed-off-by: Peter Hutterer - (cherry picked from commit a85f0d6b98237d8a196de624207acf1983a1859a) + Reported-by: Aaron Plattner + Signed-off-by: Peter Hutterer -commit 50d80c25525a691472e3fc5859fb303a3ffe1ef2 +commit 994f7a1c814a89e90f710dac5bf6b2445fb64712 Author: Peter Hutterer -Date: Mon Jan 12 15:29:36 2009 +1000 +Date: Wed Jul 29 12:11:13 2009 +1000 - dix: fix WarpPointer calls for devices with custom valuator ranges (#19297) + record: silence some compiler warnings. - If the MD's lastSlave was a devices with custom axes ranges, then a - WarpPointer would position the cursor at the wrong location. A WarpPointer - request provides screen coordinates and these coordinates were scaled to the - device range before warping. + warning: passing argument 4 of ‘RecordAProtocolElement’ discards qualifiers + from pointer target type + note: expected ‘pointer’ but argument is of type ‘const void *’ - This patch consists of two parts: - 1) in the WarpPointer handling, get the lastSlave and post the event through - this device. - 2) assume that WarpPointer coordinates are always in screen coordinates and - scale them to device coordinates in GPE before continuing. Note that this - breaks device-coordinate based XWarpDevicePointer calls (for which the spec - isn't nailed down yet anyway) until a better solution is found. + record.c:2745: warning: passing argument 1 of ‘SwapConnSetupInfo’ from + incompatible pointer type + ../include/swaprep.h:243: note: expected ‘char *’ but argument is of type + ‘struct xConnSetup *’ - X.Org Bug 19297 + record.c:2745: warning: passing argument 1 of ‘SwapConnSetupInfo’ from + incompatible pointer type + ../include/swaprep.h:243: note: expected ‘char *’ but argument is of type + ‘struct xConnSetup *’ Signed-off-by: Peter Hutterer - (cherry picked from commit d36adf52a2b2711d22b11105f7bd907d4493fb9b) -commit b19f12712a0762f20065282fd055fe9bc3ca975f +commit a863d636293cd7361639c1a8cf9c4f7f15da1e1d Author: Peter Hutterer -Date: Mon Jan 12 16:16:24 2009 +1000 +Date: Wed Jul 29 12:09:34 2009 +1000 - dix: EnqueueEvent and PlayReleasedEvent need to handle DeviceMotionNotifies + Xi: remove FIXME and obsolete include. - No MotionNotify events in the processing anymore, so let's have them treat DMN - instead. + Signed-off-by: Peter Hutterer + +commit fac49df08f173f091cbb77feaf373d7d465358af +Author: Peter Hutterer +Date: Wed Jul 29 12:07:22 2009 +1000 + + Xi: remove obsolete comment. - Reported by Thomas Jaeger. - (cherry picked from commit 488d45295105daf10ccd17ca93ae6a6f4d0104f1) + XI1 only uses 7 bits for deviceids, bit 8 is used for the MORE_EVENTS flag + on the wire (when DeviceValuator events are required). + + Signed-off-by: Peter Hutterer -commit ef796a95dd7ddf4ad5761a6ee2de1cdb56190ba2 -Author: Xiang, Haihao -Date: Tue Jan 6 10:03:19 2009 +0800 +commit 845e65f08059e8f4bfd37356e99b48bba9416c0c +Author: Peter Hutterer +Date: Wed Jul 29 11:54:14 2009 +1000 - avoid a potential endless loop. + xkb: move XkbFilterEvents to xkbsrv.h - Previously it is possible that creating rotation data, then cleaning - up and creating again so that pScreen->BlockHandler and - xf86_config->BlockHandler all point to xf86RotateBlockHandler. - See bug #19343. - (cherry picked from commit 5f3188228eb988bd8f08b62c84f98a8ff66ee283) + Signed-off-by: Peter Hutterer -commit 840b4da5f29e43ccadc720f075e00237d12ef01b -Author: George Staplin -Date: Mon Jan 26 03:02:38 2009 -0700 +commit 8da0ff2d51086666d10ca7330d428e8610a4a0e3 +Author: Oliver McFadden +Date: Tue Jul 28 08:20:37 2009 +0300 - XQuartz: GL: indirect.c changes to fix the build with newer OpenGL frameworks. - (cherry picked from commit 143224405ba74929c702a95de52b56df140b0d1b) + xf86Xinput: Add the xf86Post(Proximity|Button|Key)EventP helper functions. + + xf86PostKeyboardEvent also makes use of xf86PostKeyEventP to avoid code + duplication, and the valuator verification has been split into the + XI_VERIFY_VALUATORS macro. + + Signed-off-by: Peter Hutterer -commit 97a6f55f221e3d470add18db80e3769713480103 -Author: Jeremy Huddleston -Date: Fri Jan 16 19:26:07 2009 -0800 +commit a148d407429c7d13136b3fcafd2d279c5438df73 +Author: Peter Hutterer +Date: Tue Jul 28 16:53:51 2009 +1000 - XQuartz: SnowLeopard: Help system book name changed in 10.6 - (cherry picked from commit b57cb05c69acbedb00a97234099ea104309aa2cb) + xkb: restore XKB PtrBtn actions. + + Ifdef'd out since the switch to internal events. PtrBtn actions now work + again. Instead of generating the event directly, GPE generates the event and + it is then posted through the usual event processing routines + (mieqProcessDeviceEvent). + + Signed-off-by: Peter Hutterer -commit e81b4d495ba02d4fc0b05c26a53fd2a816a61eaa -Author: Jeremy Huddleston -Date: Fri Jan 16 13:54:08 2009 -0800 +commit 3d3b8babd1a5407082f1a40875ed69f62ba2153f +Author: Peter Hutterer +Date: Tue Jul 28 16:51:14 2009 +1000 - XQuartz: mieq: Wait for the server to finish initializing before letting other threads mieqEnqueue + mi: update master event copying to InternalEvents. - Avoid possible race condition whereby one thread might call mieqEnqueue before InitAndStartDevices finishes - (cherry picked from commit 94e417ac87a98cd5c6bf2d7c495d702748398931) + This is long overdue. The device events are InternalEvents now (and only one + at a time), diminishing the need for an EventList for the master event. + Furthermore, don't make masterEvent a static since this will interfere if + mieqProcessDeviceEvent is called from somewhere else (e.g. XKB actions). + + Signed-off-by: Peter Hutterer -commit d6498ea62117cce6a00ac045ddeefb7268a70f5b -Author: Eric Anholt -Date: Fri Jan 30 16:17:58 2009 -0800 +commit f85619b14d130ec54d42cabfaee15e55ced0c665 +Author: Peter Hutterer +Date: Tue Jul 28 16:48:56 2009 +1000 - Move the apple fat binary hacks back to a header file, and make it apple-only. - (cherry picked from commit dd098501d800571e71d06ffc936635a6c9d44e05) + dix: update GetMaximumEventsNum() to real value (3). + + GPE and friends now use internal events so they may generate up to 3 events. + One (optional) DeviceChanged event and one raw event plus a device event. + + Signed-off-by: Peter Hutterer -commit b530a0e97b5737375ca1f4eced26d587a16f843f -Author: Jeremy Huddleston -Date: Thu Jan 15 20:57:42 2009 -0800 +commit 0217d0370c0b0bce66a9c09092eda8e820274e2e +Author: Peter Hutterer +Date: Tue Jul 28 14:54:30 2009 +1000 - XQuartz: Fix builddir != srcdir issues and undef _XSERVER64 where appropriate on fat binary compilation - (cherry picked from commit f020900641b44a1142e5c2198e9678de2744454e) + record: ifdef out RecordExtensionInit and print a warning to the log. + + The RECORD extension is currently broken. By ifdef'ing out the content of + RecordExtensionInit the extension isn't added to the server's internal list + and it does not get advertised to the client. Clients can thus fail + gracefully with a "extension not supported" instead of waiting forever for + events that never arrive. + + Signed-off-by: Peter Hutterer -commit 82fd4d170824068dddf3b4e6581a12d48e1a069a -Author: George Staplin -Date: Mon Jan 12 19:30:43 2009 -0700 +commit 4e9b2938cd8637a5d3b0a4c9f69d6ee75faab3a0 +Author: Peter Hutterer +Date: Mon Jul 27 16:54:33 2009 +1000 - XQuartz: xpr: Cleanup some of the code and possibly fix part of the GLX Pixmap problem. + include: untangle events.h from the SDK headers. - Split DRICreateSurface into 3 functions. + InternalEvents shouldn't be used anywhere outside the X server itself. Split + up into events.h for opaque typedefs for the events needed by various + headers and eventstr.h for the actual struct definitions. - Make CreateSurfaceForPixmap use xp_configure_surface. I suspect this is partly why - GLXPixmaps never worked. + eventstr.h must only be included by code that requires internal events and + is not part of the SDK. - It will require some more work and thoughts for pbuffers, unless we fake those with - pixmaps and surfaces. - (cherry picked from commit 9cf264e67744262b9f45079e6cd752eb3e3b0e08) + Signed-off-by: Peter Hutterer -commit 9cbdcf1fa4bdcc115def9f134a57ac96a56c69dd -Author: Jeremy Huddleston -Date: Sun Jan 11 14:49:39 2009 -0800 +commit fad5f96c01811af7490a071719e4017c019a8d9b +Author: Jon TURNEY +Date: Wed Jul 29 13:50:48 2009 +0100 - XQuartz: Only call DarwinUpdateModKeys when needed + Cygwin/X: Only try to build rootless extension if multiwindow extwm mode is being built - Previously, we were calling it on almost every itteration through sendX11Event - (cherry picked from commit 6461729647ff4441d80811e73f0c0d2f108f2700) + Rootless extension still needs a bit more work to build successfully for Cygwin/X -commit 037e52fd7353410397dfd877b3e9af99f791fce1 -Author: George Staplin -Date: Mon Jan 12 14:14:44 2009 -0700 +commit de2ae521abde445daaf025a07aa01563ca5ddd41 +Author: Colin Harrison +Date: Sun Jul 26 20:48:19 2009 +0100 - XQuartz: GL: Make indirect.c build and work in the 1.6 branch. - - Reorder some header files and provide some types earlier on. + Xming: Use RegisterClassEx() instead of superseded RegisterClass() - Remove the static __GLXextensionInfo __glDDXExtensionInfo; that isn't - used in 1.4 or 1.5, and seems to have been removed from 1.6. Remove - the data structures associated with that too. + RegisterClass is supserseded by RegisterClassEx, so change to using that everywhere - Remove __glXAquaDrawableResize. The GLX structure doesn't use - it anymore, and the Apple version did nothing useful before. + Signed-off-by: Jon TURNEY + +commit 764ce6ee683db342264bbca4df6379eb6093fb85 +Author: Colin Harrison +Date: Tue Jul 28 16:16:57 2009 +0100 + + Xming: Use GetClassLongPtr() instead of superseded GetClassLong() - __glXAquaDrawableSwapBuffers(): base no longer contains a drawGlxc - member. Now provide the the Apple/Aqua context in the __GLXAquaDrawable - struct. Add the context member to the __GLXAquaDrawable struct. + GetClassLong() is superseded by GetClassLongPtr(), so change to using that + +commit a85523dc50f392a33a1c00302a0946828bc9249d +Author: Jeremy Huddleston +Date: Tue Jul 28 22:52:33 2009 -0700 + + XQuartz: Avoid a possible spinlock in applicationWillTerminate + (cherry picked from commit f430cda0fdcc1a8fc5f4795743b40f09ff0bd869) + +commit 1e49c8d340ee8d8b6f90abcc2e2e9c390cbcd622 +Author: Jeremy Huddleston +Date: Tue Jul 28 13:40:15 2009 -0700 + + XQuartz: Avoid namespace collission for BOOL in Sparkle + (cherry picked from commit 227c6e01a641b72158201553273299283cdb5599) + +commit 442967c90dd9d8483a56bdc9237c49e33d619126 +Author: Alan Coopersmith +Date: Tue Jul 28 15:02:37 2009 -0700 + + Remove hardcoded gcc -Wall option from configure.ac - Remove the fallback configs. They aren't used in 1.4 or 1.5 either. + Signed-off-by: Alan Coopersmith + +commit 52e9ef5664a697a31102e8761eaa03cff01d14d8 +Author: Tiago Vignatti +Date: Tue Jul 28 19:04:59 2009 +0300 + + xfree86: remove some RAC junk - Remove init_visuals(). It's not used in 1.4 or 1.5 either. + Signed-off-by: Tiago Vignatti + +commit d57361bb9ac08bec470f76ca8ca602d60c339502 +Author: Tiago Vignatti +Date: Tue Jul 28 18:55:32 2009 +0300 + + doc: remove outdated PCI/RAC/Domain notes - In the drawable constructor initialize the state properly, including - the new context member. + Signed-off-by: Tiago Vignatti + +commit 7b3d05ebd5a55f88098f4a763d1fa7ca110bb780 +Author: Tiago Vignatti +Date: Tue Jul 28 18:45:16 2009 +0300 + + xfree86: remove pci debug macros - Remove glAquaResetExtension() -- it's not used anymore. + there's no effect when their are enabled. - This has been tested remotely and proven to work with glxgears, - fire, various texture programs I wrote, and various Mesa demos. - (cherry picked from commit d514152195452ae11ec7769c76489651620ad380) + Signed-off-by: Tiago Vignatti -commit ee9736b26b1021fac2cf259e6b2142ea780b2af8 -Author: George Staplin -Date: Fri Jan 9 16:08:26 2009 -0700 +commit 6089d9cfde533c6a7aec7dfc08aec9c8f2f18e27 +Author: Tiago Vignatti +Date: Tue Jul 28 18:29:35 2009 +0300 - XQuartz: GL: Set the __GLXconfig renderType to GLX_RGBA_BIT. - (cherry picked from commit 3c14546f58f8a138fe67c9cacc3bd0b7fa90c29a) - (cherry picked from commit a665ed16f736cf1901b89448dc5d37f4d16dfaf4) + xfree86: remove RAC/resource doc + + Signed-off-by: Tiago Vignatti -commit b646513861aac9538ba3ebd8e2c7ef530392ff5d -Author: George Staplin -Date: Fri Jan 9 15:53:39 2009 -0700 +commit c553161e17ebc577ecb91ec6c81f0bdd0ae85e13 +Author: Tiago Vignatti +Date: Fri Jul 24 13:47:31 2009 +0300 - XQuartz: GL: Make many more fbconfigs and visuals available for the 1.5 branch. + xfree86: delete devices probe code (-probe and -probeonly options) - Use a __GLXconfig linked list struct to store the configurations for the - fbconfigs and visuals in a pGlxScreen. + Inside a windowing system, it's not the place to probe for devices. Goodbye + -probe and -probeonly. - Also, remove the __GLXvisualConfig/GlxSetVisualConfigs code that isn't used - anymore. There is more code we can remove later, but I want to do that in - separate commits. - (cherry picked from commit 94162b0f8a25267aca280d25e216cc6bde47da6e) - (cherry picked from commit 15bb6abd59fdefe7037237faaea1a39711a972ed) + Signed-off-by: Tiago Vignatti -commit 9e03178ae2da6beb939141f662d875c4c1d40354 -Author: Michel Dänzer -Date: Tue Jan 13 13:05:32 2009 +0100 +commit c09779f95a9772c0556760222dfc570dbaf8a28e +Author: Tiago Vignatti +Date: Fri Jul 24 13:37:31 2009 +0300 - EXA: Declare glyph cache picture as component-alpha when necessary. + xfree86: delete stupid video driver dump (-modalias option) - Without this, rendering component-alpha glyphs may break without a mask. + Such stupid and ugly way to dump PCI information! Oh boy... Anyway, this + doesn't belong to the X server at all. Go away! - Fixes http://bugs.freedesktop.org/show_bug.cgi?id=19233 . - (cherry picked from commit 639f289dcdbe00a516820f573c01a8339e120ed4) + Signed-off-by: Tiago Vignatti -commit 648807adb059b6cfc413d41e3a37670ce850aee6 -Author: Keith Packard -Date: Fri Jan 30 15:43:04 2009 -0800 +commit 0524420622df6cbdb3872917906f7b2a6ec02958 +Author: Colin Harrison +Date: Mon Jul 27 15:49:41 2009 +0100 - Make RandR CRTC info report panning area instead of just crtc area + Cygwin/X: use GWLP_WNDPROC, GWLP_USERDATA with Get/SetWindowLongPtr - This makes the RandR info consistent with the Xinerama info. + Missed from commit b3751454cbe02ee952bab213e8c3684d429c41b3 - Signed-off-by: Keith Packard - (cherry picked from commit 6e49fdd2c839b6244c23ce95c3ae76a1a265b986) + Signed-off-by: Jon TURNEY -commit 9b2958b3a2fb4b65d67916ba7e98dfd21ad434f3 -Author: Maarten Maathuis -Date: Fri Dec 19 18:59:27 2008 +0100 +commit b1c3dc6ae226db178420e3b5f297b94afc87c94c +Author: Rémi Cardona +Date: Mon Jul 27 12:07:51 2009 +0200 - randr: Consider panned crtc's when calculating xinerama screen sizes. + config: add HAL error checks - - This will allow window managers and applications to actually use the panned area. - (cherry picked from commit 332d65ec7a6e94d75efe95d53742f137835274de) + This patch simplifies error handling in the HAL code and fixes a + segfault if libhal_find_device_by_capability() failed. + + Fixes http://bugs.gentoo.org/278760 + + Based on a patch by Martin von Gagern + + Signed-off-by: Rémi Cardona + Acked-by: Peter Hutterer -commit 251d0d8090322b2c9dc0c8b7bef001f338d19433 -Author: Keith Packard -Date: Mon Jan 12 13:03:30 2009 -0800 +commit 8898203b0d0e9fa03453b2bcd9b88843cccc3230 +Author: Dave Airlie +Date: Tue Jul 28 19:35:37 2009 +1000 - Update version to 1.5.99.901 (1.6 RC1) + sbus: fixup for rac removal -commit 71628cabc9dddf5c9fe12d555ae0c47d5902b2f7 -Author: Keith Packard -Date: Mon Jan 12 11:51:28 2009 -0800 +commit 4b42448a2388d40f257774fbffdccaea87bd0347 +Author: Dave Airlie +Date: Tue Jul 28 14:47:42 2009 +1000 - Merge include/dix-config-post-verbatim.h into include/dix-config.h.in + xserver: remove RAC/resource handling code. - No sense having an extra include file here; it makes building stuff harder. + This changes the ABI, but since the video ABI is at 6 already + it should be fine. - Signed-off-by: Keith Packard + driver changes are in the pipeline after this. + + Signed-off-by: Dave Airlie -commit cbd977cc15cf22a3e42b10942d5657cb8b9ac335 -Author: Alban Browaeys -Date: Mon Jan 12 10:57:11 2009 -0800 +commit 0a168401c401727e49a12cae43a6a387b1f2928d +Author: Dave Airlie +Date: Tue Jul 28 14:19:24 2009 +1000 - Fix dmx compile on 1.6 branch. - - http://www.mail-archive.com/pld-cvs-commit@lists.pld-linux.org/msg166742.html - contains a patch from "arekm" which I build upon to make up this one. + pci: add support for pci is boot vga call. - Mostly fixes for Xcalloc which I previously tried to fix the same way but - without finding the right spot. + This allows us to ask the kernel for the boot VGA device + instead of nasty guessing. + +commit b3751454cbe02ee952bab213e8c3684d429c41b3 +Author: Colin Harrison +Date: Tue Jul 21 16:07:56 2009 +0100 + + Cygwin/X: Update Get/SetWindowLong() to Get/SetWindowLongPtr() everywhere - I cannot tell if dmx works though. Only that it at least compile. + Get/SetWindowLong() is superseded by Get/SetWindowLongPtr(), so change to + using that everywhere it remains - Signed-off-by: Keith Packard + Signed-off-by: Jon TURNEY -commit 6be355b8e8cabeb5832ce9970a83782ea46fd4d1 -Author: Peter Hutterer -Date: Fri Jan 9 13:46:20 2009 +1000 +commit c81595e23b48368dafc054f023c1dd16bbad9494 +Author: Jeremy Huddleston +Date: Sun Jul 26 02:32:17 2009 -0700 - dix: drop x/y back into last.valuators before updating the history (#19285) - - positionSprite needs to scale to screen coordinates and in the process of - doing so alters dev->last.valuators[0:1]. Drop the real coordinates back after - finishing and before updating the motion history. This way, we don't push the - screen coordinates into the motion history. + XQuartz: Overhaul setting up visuals - X.Org Bug 19285 - (cherry picked from commit 56efbc0986e782da45addb05ece9f456d41d7a90) + The main change is cleanup of the visualConfigs and setting up alpha correctly there to match the visuals being added earlier (so the default visual has a corresponding GLX visual) + (cherry picked from commit 7351db5c8746be30a4047469ee9b50bc19e62a89) -commit ca21fd7ddf50e6a363c6ad027d46388596872d3b +commit 7f28c555b80dda2ed4f518efdb79733647dfea80 Author: Jeremy Huddleston -Date: Sun Jan 11 01:52:42 2009 -0800 +Date: Sat Jul 25 20:19:05 2009 -0700 - XQuartz: Add locking to make mieq thread safe on OSX - (cherry picked from commit 7a8d2266861e74176b5310b83652a9c10a170494) + XQuartz: Use CopyKeyClass to copy the keymap to the virtual core keyboard. + (cherry picked from commit 9a801d1716f005c30be076fcc9ac8dbb3e74d989) -commit 671b71a6cba1c9da7797524f21ef1b27ff04adfd +commit 1031ac3a7306e7a82169c79c64607696c826c47f Author: Jeremy Huddleston -Date: Sun Jan 11 01:42:24 2009 -0800 +Date: Sat Jul 25 20:18:38 2009 -0700 - XQuartz: misc 1.6 updates (still --disable-glx) - rlAccel is not longer compatable, and it's not worth fixing - Don't override DeviceCursorInitialize with a noop - Don't do a SwitchCoreKeyboard (which wasn't even needed in the first place) - (cherry picked from commit c137f681680e1d04b1513a8be68aeda4d1c56fd5) + Revert "XQuartz: Copy the keyboard map to the core keyboard" + + This reverts commit 795de791cf18c658421d701af645718493eac51e. -commit 69ddac23281534a06c0acb3005a09e4448594925 +commit 48703083a1cf308306f254691d7c2ecda09b3812 Author: Jeremy Huddleston -Date: Sat Jan 10 01:39:08 2009 -0800 +Date: Sat Jul 25 19:34:17 2009 -0700 - Apple: Don't use DRI2 - (cherry picked from commit a1d35cee5907a76977ee43c49cb55c8f411c9794) + XQuartz: Use pDev=NULL for DarwinSendDDXEvent + + These events aren't really related to physical input devices anyways, so it doesn't make sense to use the pointer. + (cherry picked from commit bfe0b9cfa7af4a48dba849cab1eb152c409b4e08) -commit 681cc0f38b0b96c5e41c93d6944e4fa58c950eda +commit b8e0f740829d0c81324aeb59222fc8e3d22493cc Author: Jeremy Huddleston -Date: Tue Dec 23 12:39:38 2008 -0800 +Date: Sat Jul 25 20:17:25 2009 -0700 - XQuartz: Use depth=24 instead of FatalError if we can't figure out our depth - (cherry picked from commit 80b65c5b78d125c4ad3620b87b121c9e666299c3) - (cherry picked from commit 932ed6e949757926a17f7efe6b0255e38efa1152) + mieq: Protect from pDev=NULL in mieqEnqueue and mieqProcessInputEvents + (cherry picked from commit bf60ffb49700da367f7f88983b042a88fef7219b) -commit 7872669105d6d81ded5128eb4d4f1acce34c3b02 +commit dab8de036808e363a8ec9be826cff7fbcd92b953 Author: Jeremy Huddleston -Date: Tue Dec 9 23:48:18 2008 -0800 +Date: Sat Jul 25 15:21:12 2009 -0700 - XQuartz: Make debugging output for invalid depths a bit more detailed - (cherry picked from commit 609fb166b7062c76f0561df12ffe893811f6ac8f) - (cherry picked from commit 1962af7ee3bdf54cfa674187dea67b9ad36cd5a1) + XQuartz: Bump the reported version to X11R7.4 + (cherry picked from commit c83f701aa75c75433b8745f5d79bca3a7516df91) -commit fdfc70ea0a183daaabb9bb436f2b0cbf95f47e88 -Merge: 8cfb353 82f4273 -Author: Keith Packard -Date: Mon Jan 12 10:40:04 2009 -0800 +commit 9e74bb97ab6cb83dd99ec36bd3842197263b8a8c +Author: Jeremy Huddleston +Date: Sat Jul 25 10:39:26 2009 -0700 - Merge commit 'whot/server-1.6-branch' into server-1.6-branch + XQuartz: xpr: Added missing include for RootlessHideAllWindows + (cherry picked from commit 75e104730810acbb11bf8503bac24ea25243b2ac) -commit 82f42730fa268e85dc95c43a6450c35f9e57acdf -Author: Peter Hutterer -Date: Mon Dec 15 10:46:07 2008 +1000 +commit 09b024db636f5a306c9cec3806fffde9acd3c408 +Author: Jeremy Huddleston +Date: Sat Jul 25 15:22:01 2009 -0700 - mi: force CopyKeyClass for key events. (#19048) - - While we don't want to copy all other device classes into the VCK, we need to - copy the key class to transfer the layout from the SDs into the VCK. - This resembles the functionality of SwitchCoreKeyboard in server 1.5. - - Thanks to Colin Guthrie for providing the follow-up patch (#19222) + XQuartz: Use the master device in DarwinSendDDXEvent to avoid duplicate events. + (cherry picked from commit 6fa62192af937aa9656f64b516050bc099231c7a) + +commit 795de791cf18c658421d701af645718493eac51e +Author: Jeremy Huddleston +Date: Sat Jul 25 15:25:06 2009 -0700 + + XQuartz: Copy the keyboard map to the core keyboard - X.Org Bug 19048 + This still doesn't handle the modifier map... gotta figure out what to do now that SwitchCoreKeyboard is gone + (cherry picked from commit 427e1aab41dabb54354bfd30f2baae98ac8202c0) + +commit af8047f7816b39be7015e8cbbe085c57bb354592 +Author: Aaron Plattner +Date: Fri Jul 17 15:10:46 2009 -0700 + + Damage: Add devPrivates to DamageRec - Signed-off-by: Peter Hutterer + Signed-off-by: Aaron Plattner -commit 8cfb353078d9b5d03a9633304038141a60adc970 -Author: Peter Hutterer -Date: Mon Jan 12 11:07:08 2009 +1000 +commit c80ed4945ef6dd50473fab871211b71b3d20dc3c +Author: Aaron Plattner +Date: Fri Jul 17 15:10:32 2009 -0700 - dix: Fix handling of do_not_propagate_mask window attribute. + Damage: Add wrappable hooks for damage create, destroy, register, & unregister. - This was broken in 32aa252e988be8cbfd4f7e373fb7b7736ef1f5f2. + Signed-off-by: Aaron Plattner + +commit fb46474feb69d52d3dfdd61452ced5cfc38f6651 +Author: Jeremy Huddleston +Date: Fri Jul 24 09:16:37 2009 -0700 + + XQuartz: Rever the "Set can_quit to true during a Sparkle-initiated relaunch." change - Signed-off-by: Kim Woelders - Signed-off-by: Peter Hutterer - (cherry picked from commit 30d2cfcd3851870178d62e5067211aa36f87fbd2) + This was in place to work around the issue that was correctly solved with the + previous commit (changing the Windows menu behavior). Reverting this change + no longer causes crashing, so it's safe to show the dialog now. + (cherry picked from commit 023cef31bbb2ab80a241098d82fcfd35ada75fc0) + +commit 0d73893a5795d5c1d28fd0287f2f1965f2ff1bda +Author: Jeremy Huddleston +Date: Thu Jul 23 20:03:29 2009 -0700 + + XQuartz: Change handling of Windows menu to workaround a bug triggered by mixing Sparkle and X11 windows - Conflicts: + NSApplication releases the separator in the Windows menu even though it's an IBOutlet + (cherry picked from commit 27ac5135267be9cb221329ae68461117dd43a4bf) + +commit 171409cecbd848e5fec6334fc61a20e882e80fa9 +Author: Jeremy Huddleston +Date: Tue Jul 21 00:22:12 2009 -0700 + + XQuartz: Set can_quit to true during a Sparkle-initiated relaunch. + (cherry picked from commit b2e9a77111d4572402d8ca95e3368db97ba7d598) + +commit 7ff84d350b44fa40669c1d0d48a715a0bf056ece +Author: Jeremy Huddleston +Date: Mon Jul 20 22:38:25 2009 -0700 + + XQuartz: Added a "Check for X11 Updates..." menu item. + (cherry picked from commit 305144bfa4b59791123c44e869fab93a084792d6) + +commit 1e1dbd1e462f571dad2f9684fcf4cd8ae17eedd5 +Author: Tiago Vignatti +Date: Mon Jun 29 11:46:17 2009 +0300 + + xfree86: "Staticize" functions in xf86AutoConfig.c - dix/events.c + Signed-off-by: Tiago Vignatti + +commit 7c6b5458de9bc7f6cd972a36b56888aaa3d201ee +Author: Aaron Plattner +Date: Thu Jul 23 15:16:44 2009 -0700 + + Fix dist. - Signed-off-by: Keith Packard + * Makefile.am: Include the test/ subdirectory in the dist tarball. + + * include/Makefile.am: Move events.h to sdk_HEADERS and eventconvert.h to + EXTRA_DIST so they're included in the tarball. events.h shouldn't be + included in the SDK either, but for now it's needed by input.h. -commit 10c0287232eab1b93d078774f52e65efa0c03607 +commit 20daa145c437c3ba67970146f6182849f87a1b43 Author: Peter Hutterer -Date: Mon Jan 12 11:07:08 2009 +1000 +Date: Thu Jul 23 08:33:00 2009 +1000 - dix: Fix handling of do_not_propagate_mask window attribute. + mi: fix build error caused by missing xtest.h - This was broken in 32aa252e988be8cbfd4f7e373fb7b7736ef1f5f2. + xtest.h was renamed to xtestconst.h in xextproto. + Requires xextproto 7.0.99.3 - Signed-off-by: Kim Woelders Signed-off-by: Peter Hutterer - (cherry picked from commit 30d2cfcd3851870178d62e5067211aa36f87fbd2) - - Conflicts: - - dix/events.c -commit 93e510d3abaee97f5b7812abc3a2693dabf94478 -Author: Peter Hutterer -Date: Thu Dec 4 10:30:02 2008 +1000 +commit 587c3a2d1961834558193e8e14e8e381a077a253 +Author: Aaron Plattner +Date: Wed Jul 22 10:55:46 2009 -0700 - xfree86: Only use the evdev ruleset on linux. + Bug #22804: Reject out of bounds XGetImage requests - As suggested by Julien Cristau + The XGetImage man page states: - This is an follow-up to - commit 9c5dd7337fa93fb1650cc017e523b939dcbf482a - Author: Peter Hutterer - Date: Wed Dec 3 14:24:25 2008 +1000 + 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. Note that the borders of the window can be included and read with + this request. - Let the DDX decide on the XkbRulesDefaults. + However, the server was only checking the requested region against the screen + bounds, allowing XGetImage requests to read pixels outside the bounds of a + window's ancestors. Normally, this would just read other pixels from the + screen, but if one of the ancestor windows is redirected, the window's backing + pixmap may be smaller than the window itself. - Signed-off-by: Peter Hutterer - Acked-by: Julien Cristau - Signed-off-by: Daniel Stone - (cherry picked from commit 13de7511b17b57a28668e1a60b196ccfe61dbcbe) + This change checks the region against the window's bounding drawable, which is + either the screen pixmap, a redirected window's backing pixmap, or the root + window for servers that don't support GetWindowPixmap. - Signed-off-by: Keith Packard + Signed-off-by: Aaron Plattner + Reviewed-by: Keith Packard -commit cc78f04b22daabc06d9fb82f550e3f4dbbe76f28 -Author: Peter Hutterer -Date: Wed Dec 3 14:24:25 2008 +1000 +commit ecd618957ebf01cb4137f98efec3faed35f8a9f5 +Author: Peter Hutterer +Date: Wed Jul 22 15:19:34 2009 +1000 - Let the DDX decide on the XkbRulesDefaults. - - Rather than assuming rules in the CoreKeyboardProc, init the default rules in - InitCoreDevices, then re-use them later. + dix: fix null-pointer dereference on activating enter/focus grabs. - In the xfree86 DDX, set the rules to "base" or "evdev", depending on whether - we'll load kbd or evdev. + EventToXI2 returns a NULL event for enter/focus events since these events + aren't yet wrapped into internal events. This is a quickfix only, it works + thanks to the alignment of internal and XI2 event types. - If we create a new MD, use pc105,us as default and re-use the rules file used - previously. + Eventually, enter/leave events should be wrapped into internal events. - Signed-off-by: Peter Hutterer - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 02c059ea99a791b9e3643b4fb131af01306b1c23 -Author: Keith Packard -Date: Sun Jan 11 15:56:28 2009 -0800 +commit 48585bd1e3e98db0f3df1ecc68022510216e00cc +Author: Peter Hutterer +Date: Wed Jul 22 12:09:24 2009 +1000 - dix: move focus handling into enterleave.c. - - This commit moves the focus handling from events.c into enterleave.c and - implements a model similar to the core enter/leave model. - For a full description of the model, see: - http://lists.freedesktop.org/archives/xorg/2008-December/041740.html + dix: use the event mask of the grab for TryClientEvents. - This commit also gets rid of the focusinout array in the WindowRec, ditching - it in favour of a local array that keeps the current focus window for each - device. + A client that grabs for button/key events may not have the + ButtonPress/KeyPress mask set and should not receive an event in that case. Signed-off-by: Peter Hutterer - (cherry picked from commit eb2d7b3d700952ba88c77deacf687b251300e660) - - Conflicts: - - dix/events.c - include/input.h - - Signed-off-by: Keith Packard -commit c75445754e283dc2a306ef1cf2bbd94b40423640 -Author: Peter Hutterer -Date: Fri Dec 19 21:05:09 2008 +1000 +commit 4dc91b3e54503a1be555bae5b18f3e52f58be307 +Author: Pierre Willenbrock +Date: Tue Jul 21 17:21:28 2009 +0200 - dix: add a few auxiliary functions for the updated focus model. + Check if new space was actually allocated before freeing. - SetFocusIn and SetFocusOut, including the static array to keep all focus - windows. + There will be no new space allocated, if mode != PropModeReplace and + len == 0, or if mode is not one of the handled modes. + This fixes freeing data that is still in use, leading to double frees and + other memory corruption. Signed-off-by: Peter Hutterer - (cherry picked from commit 673eb23aac578dcdc04e2a99d1fa5c2987eb58b8) -commit ceeb62f892a1a4df9ef1054d2c9ba12e26d55ef2 +commit 9a1bfa5664a80f03cedeb89b9f8a86115a08e7af Author: Peter Hutterer -Date: Thu Dec 18 15:55:11 2008 +1000 +Date: Mon Jul 20 16:39:16 2009 +1000 - dix: reduce FirstPointerChild complexity + input: remove XI2 keysym grabs, use keycode grabs instead. - Instead of keeping a flag on each window for the devices that are in this - window, keep a local array that holds the current pointer window for each - device. Benefit: searching for the first descendant of a pointer is a simple - run through the array. + Keysym grabs are tricky in the details, keycode grabs are known to work. So + for now, provide keycode grabs only. + + Requires inputproto 1.9.99.15. Signed-off-by: Peter Hutterer - (cherry picked from commit 38b28dcadd0990cb43f50db4300eebb8f044db96) -commit ed90245461f33f1e0eac01c0dea4d0132364f60a -Author: Peter Hutterer -Date: Thu Dec 18 15:43:18 2008 +1000 +commit afc3e3b5955ea4a49308399820cc4c499f4312da +Author: Jon TURNEY +Date: Tue Jul 21 16:14:21 2009 +0100 - dix: remove now unused "exclude" parameter from FirstPointerChild + Cygwin/X: winInitMultiWindowClass() should be static - Signed-off-by: Peter Hutterer - (cherry picked from commit 72ad4a85cc0ffe60a90011d65ef718d5852beae4) + Signed-off-by: Jon TURNEY -commit f0025f59e85f48da5e77cd6b759a693d40b2f9ea -Author: Keith Packard -Date: Fri Jan 9 11:20:31 2009 -0800 +commit 9dc8106ba095474fea1474622b0f0162c8fd5aa4 +Author: Michel Dänzer +Date: Tue Jul 21 14:34:47 2009 +0200 - dix: re-implement enter/leave model. + randr12: Add compatibility for XF86VidMode gamma ramps. - The old model was implemented based on a misunderstanding of NotifyVirtual and - NotifyNonlinearVirtual events. It became complicated and was broken in some - places [1]. This patch wipes this model completely. + Fixes screensaver fadeout effects. - A much simplified implementation is provided instead. Rather than a top-down - approach ("we have a tree of windows, which ones need to get which event") - this one uses a step-by-step approach. For each window W between A and B - determine the pointer window P as perceived by this window and determine the - event type based on this information. This is in-line with the model described - by Owen Taylor [2]. + Also make all RandR 1.2 compatibility code for XF86VidMode operate only on the + CRTC associated with the compatibility output, not all CRTCs at once. + +commit 268e227ba06c027f5c56b1aaee5dcc6a2034403f +Author: Michel Dänzer +Date: Tue Jul 21 14:34:13 2009 +0200 + + EXA: Make Prepare/FinishAccess tracking resilient to repeated / nested calls. - [1] http://lists.freedesktop.org/archives/xorg/2008-December/041559.html - [2] http://lists.freedesktop.org/archives/xorg/2008-August/037606.html - (cherry picked from commit 0aa4460c3b4f9bb17d5412d76fa8c4c501132429) + Use reference counting and do nothing unless the reference count transitions + to/from 0. - Conflicts: + Fixes https://bugs.freedesktop.org/show_bug.cgi?id=22822 . - dix/enterleave.c + As a bonus, this avoids calling the driver Prepare/FinishAccess hooks more than + once per pixmap and operation. + + Also update the Doxygen documentation for the PrepareAccess driver hook to + better match current reality. -commit 1ffd9ec7606606570980490b91b780e02bc29d7d -Author: Peter Hutterer -Date: Tue Dec 16 17:10:03 2008 +1000 +commit de7a14ca92f99ff03c8ad204aab5be8203c86a72 +Author: Michel Dänzer +Date: Tue Jul 21 12:55:27 2009 +0200 - xfree86: don't restore the TTY mode if we didn't initialize it ourselves + EXA: Fix up some issues introduced by 00fe4a297744c81b40f0243fb56ad848a9be6a2b. - Restoring it unconditionally means we restore to whatever tty_mode has as - default value (i.e. 0). K_RAW happens to be 0x00, so we always restore to raw - mode if allowEmptyInput is off. + * Check all pixmaps involved for damage records, fixes visual corruption due to + the screen pixmap never having one. + * Fix an array size and remove a now superfluous assignment. + +commit be4dd35ffecbf49aff13aa9f604a44c9b665ae92 +Author: Jeremy Huddleston +Date: Sat Jul 18 15:51:29 2009 -0700 + + XQuartz: Initial support for automatic updates through Sparkle + (cherry picked from commit c45f1be36426bceeef9af67c26351114f14f5277) + +commit 2415e2dce918efd49d5e6d71f705255a234a866b +Author: Michel Dänzer +Date: Mon Jul 20 02:08:31 2009 +0200 + + EXA: Bail earlier from exaDoPutImage if the driver has no UploadToScreen hook. - Signed-off-by: Peter Hutterer - Signed-off-by: Adam Jackson - Signed-off-by: Julien Cristau - (cherry picked from commit aea6f19f25e13768b1d09fac4991d6a5e6c2cdac) + Also remove dead code associated with access_prepared local variable. -commit 3c11efd7dc09aa92039f851c2bbf2ca6b5a5ac8c -Author: Peter Hutterer -Date: Tue Dec 23 09:59:28 2008 +1000 +commit 00fe4a297744c81b40f0243fb56ad848a9be6a2b +Author: Michel Dänzer +Date: Mon Jul 20 02:04:40 2009 +0200 - xfree86: If an input device failed to activate, return immediately. + EXA: Completely eliminate exaDoMigration calls for drivers that manage pixmaps. + +commit 7b9915b11ed9eedd0698b4563328504d686ac4ec +Author: Gaetan Nadon +Date: Sat Jul 18 20:52:20 2009 -0400 + + dix: xserver "make dist" fails due to eventconvert.h (#27825) - Devices are only activated once - right after they've been added to the - server. If a device failes activation, it's dead. There's no reason to - continue. Return the error code from ActivateDevice() without setting up - sprite information or even sending a event to the client. + This header file is not in the /dix dir, but in the /include dir. + The makefile aborted and the xserver distribution files were not created. + The fix is to remove this header file from the libdix_la_SOURCES + in the dix/Makefile.am. - Then - in the DDX - just remove the device again. + X.Org Bug 27825 Signed-off-by: Peter Hutterer - (cherry picked from commit 27011254c4de4e573a0851bf46892fb488db6522) - -commit 2ce48363b862db134624797bc071f8c45323a075 -Author: Peter Hutterer -Date: Tue Dec 23 10:00:34 2008 +1000 - - xfree86: don't call CheckMotion if a device hasn't been enabled. #19176 - - X.Org Bug - (cherry picked from commit e1a3a1a0d85c9971aea65c2228b5fd4dbf3bf57a) -commit 3225e1378e3929e538941bbfc9ec495638bd407a -Author: Maarten Maathuis -Date: Fri Dec 26 15:51:55 2008 +0100 +commit a98acb290737ac6b0776f2ea8128a4613c0f778b +Author: Adam Jackson +Date: Wed Jul 15 13:56:19 2009 -0400 - exa: preparing as source and finishing access as mask is a bad idea - (cherry picked from commit 027b440d4f9f0cdd46addff46fd2d5c44cd5c847) + s/xf86EnableOutputs/xf86CollectEnabledOutputs/ for clarity -commit 2056c10d12f6b4facf628b861230f9b0a13f3a73 -Author: Maarten Maathuis -Date: Fri Dec 26 16:38:27 2008 +0100 +commit 8fab7f72f2cc4ac5ca415c95ccbd05d3084f10c4 +Author: Adam Jackson +Date: Wed Jul 15 14:09:08 2009 -0400 - exa: Allow drivers to set non-NULL devPrivate.ptr for !offscreen pixmaps. - (cherry picked from commit 3534a5e5d9c5af85149c799f324257f89507fa23) + randr: Add Option "Primary" to Monitor sections -commit 4130761f151ff5414694472e1dc29d8f2893f10e -Author: Jeremy Huddleston -Date: Wed Jan 7 10:56:26 2009 -0800 +commit 053bb92145045e06f8d72c3bddb75806432faa1c +Author: Adam Jackson +Date: Thu Jul 16 14:29:05 2009 -0400 - XQuartz: Fat binary buildfixes + xfree86: Remove xf86SetPriority - Let Xarch.h set X_BYTE_ORDER for us. - Determine _XSERVER64 at build time. - (cherry picked from commit f54736c2b0b537c8322ecb2db8e9cee5eea68165) - (cherry picked from commit cc677cb4f458f371a4012ce0dc1121a52a2cb699) + GetClocks is pretty doomed to begin with, this really isn't going to + make it worse. -commit 2eb09af56179d570a6d52b65c2029a2b62dcdff3 -Author: Jeremy Huddleston -Date: Sun Nov 16 15:35:17 2008 -0800 +commit 3637945a45d55385335833197b18703379892884 +Author: Adam Jackson +Date: Thu Jul 16 14:10:45 2009 -0400 - XQuartz: Use AvailabilityMacros.h instead of availability.h for Tiger - (cherry picked from commit 91f3c3435716195e6d606c8e2f6b0de52eb5972d) - (cherry picked from commit 65ab2f44ea7fc4d87e021bed548eda81fc3cbae7) + loader: Remove a useless check. -commit fecc82ee1147cb314725ac774525a18a861ed719 -Author: Jeremy Huddleston -Date: Fri Jan 2 11:06:12 2009 -0800 +commit ef2bb08c76f0d1eed0c8b3e4937648f0b6ae01e5 +Author: Adam Jackson +Date: Thu Jul 16 13:53:00 2009 -0400 - XQuartz: Copyright Update... happy new year - (cherry picked from commit 9f1c26ae8d9ccf81cba0ab946326968a701f8c36) - (cherry picked from commit df6ca888b0c04fdf4ff2d7fb4a414275b983ed34) + loader: remove dead LoaderCheckUnresolved -commit d80e35a9ae507ad742f97ca7254314ff28371688 -Author: Jeremy Huddleston -Date: Fri Jan 2 10:58:40 2009 -0800 +commit 06a2784232b5d0e2440ed365d272abdfa1f5d35f +Author: Adam Jackson +Date: Thu Jul 16 13:40:48 2009 -0400 - XQuartz: pbproxy: Push dpy init and CFRunLoop hook setup into the pbproxy thread to avoid possible deadlock - (cherry picked from commit 49e59d32b88e4fad070f230b5efaa261b47f78db) - (cherry picked from commit adbfd49da2453b58a9e13b09c62e0611ea1c3f77) + loader: Remove useless TestFree() macro -commit c32a680f2670f05902ee323c68f772f52d3a8320 -Author: Jeremy Huddleston -Date: Wed Dec 31 12:40:02 2008 -0800 +commit 087745b809b9a834751e5c719b9d87967194369b +Author: Adam Jackson +Date: Thu Jul 16 10:51:53 2009 -0400 - XQuartz: Don't use NX_SECONDARYFNMASK, NX_NUMERICPADMASK, NX_HELPMASK - - We don't have keycodes for them, so don't try to use them - (cherry picked from commit cb912aca3a2834c4ad8e386c8a0d05c1bb31b0e7) - (cherry picked from commit 0676a580fcc05d54049269028a34358935a4101c) + fbdevhw: Remove pointless OS check, this never gets built on non-Linux -commit e08ce79152b0887b8d67c4b03b33634f5ef499bc -Author: Jeremy Huddleston -Date: Wed Dec 31 11:57:49 2008 -0800 +commit 892c93553aa0b03aeb3f4c27d952cc4bb7120aff +Author: Adam Jackson +Date: Thu Jul 16 10:51:27 2009 -0400 - XQuartz: Make sure to reset the saved key state when deactivating X11.app - (cherry picked from commit 3eef78eb321f4f7dbca5a10c80666c621e28a1e0) - (cherry picked from commit 0d2621b6d4684ec62e67156a5a9dbdd3297f9cb0) + fbdevhw: Remove a #if 0 -commit bdf776b4579921dd49316c2ff32ff12cc702720a -Author: Jeremy Huddleston -Date: Wed Dec 31 12:44:50 2008 -0800 +commit e5712f2926abf33b146a4ccfcf6e89c914dad37b +Author: Adam Jackson +Date: Thu Jul 16 10:49:22 2009 -0400 - XQuartz: Honor system key repeat rate - (cherry picked from commit 4303c9be39a86f5a21de108f72b90a989435905e) - (cherry picked from commit 9faf3de7e5610af340b92acb1b86bf03b6f2241a) + fbdevhw: simplify some #if 1 -commit 16560db0c041b24346e186169f6e3bd315ca13ae -Author: Jeremy Huddleston -Date: Mon Dec 29 19:54:12 2008 -0800 +commit 8868bb4131be25340bf65ec61b998d353965bab6 +Author: Adam Jackson +Date: Thu Jul 16 10:44:24 2009 -0400 - XQuartz: Better avoid stuck keys on context switches - (cherry picked from commit e9963f1a4f4f12f253eae9d4f01694b6cabe35ad) - (cherry picked from commit 4be8d7346b9fdc014b72dd6c404ceecc0ef0d245) + randr: Un-duplicate the reduced blanking check. -commit 66f801f5a06925c303027f856c5ecea4926d80f2 -Author: Jeremy Huddleston -Date: Mon Dec 29 19:38:31 2008 -0800 +commit 453ee39bc6a6b46e6cda11512cfcd431ba32a4b7 +Author: Adam Jackson +Date: Thu Jul 16 10:39:37 2009 -0400 - XQuartz: Workaround OSX VNC server bug for modifier key state - - A better approach which ensures we have a L modifier key down if we are told neither are down and atleast one is down... =/ - (cherry picked from commit bc13dda345f716bf4de9bfe4e1d85969263b60c2) - (cherry picked from commit c67a3e2972c75c02f1aeed94bc0a3c5272422267) + xfree86: Fix some misleading comments -commit a05dbce40248159099c7a7ec7e94bff9ffb700fc -Author: Jeremy Huddleston -Date: Mon Dec 29 00:30:31 2008 -0800 +commit 2c57a7aa07fdf52be315ecb498341776268c1a10 +Author: Adam Jackson +Date: Thu Jul 16 10:32:53 2009 -0400 - XQuartz: Updated man page fullscreen_hotkeys fullscreen_menu - (cherry picked from commit 00f3a2e33c8e91da5f855821313a04ea97445656) - (cherry picked from commit d790c9dd041a2c8e3513d14a556333219d0f8d5e) + xfree86: Remove some #if 0 + + This code all lives in xf86Modes.c now anyway -commit b8877bb9e4c113a400e47a65a7d25d21f0a43116 -Author: Jeremy Huddleston -Date: Sun Dec 28 15:57:59 2008 -0800 +commit 1a8c89683ff40cdd1d33da89ab47e7bd240b9bb7 +Author: Adam Jackson +Date: Thu Jul 16 10:15:41 2009 -0400 - XQuartz: pbproxy: We explicitly need libX11 for pbproxy - (cherry picked from commit 32824120da0749b8369c2592f851bb1030dea9f1) - (cherry picked from commit 8c6e8fa811c782c85e7fefbe75fe5480098739ae) + xfree86: Remove TargetRefresh option + + This was a vestige from the days before we'd make the mode list from the + EDID data, and from CRT technology when you could reasonably assume that + higher refresh rates were better. Also it did not function as advertised, + acting as a high-pass filter instead of a band-pass. -commit de78ac43782d7f4ad4b98622e3c419c893d3eeff -Author: Jeremy Huddleston -Date: Sun Dec 28 01:53:38 2008 -0800 +commit b4ee3bf700e04ae1b3c4ec021373424ed6e5338f +Author: Adam Jackson +Date: Fri Jul 17 14:57:50 2009 -0400 - XQuartz: Try harder to get the user's login environment - (cherry picked from commit 6bedaddd78aa04bd303df434a4c49bb87bd7deaa) - (cherry picked from commit 338f096861136fb6c4f604e93ff21277252676b7) + dbe: Fix indentation -commit c391114df95109d306e172bea75781d0e9523f62 -Author: Jeremy Huddleston -Date: Wed Dec 24 00:39:06 2008 -0800 +commit 0bb9a7e1650180a24246d14493a8168487cf8914 +Author: Adam Jackson +Date: Tue Jun 9 11:49:41 2009 -0400 - XQuartz: Reposition windows when we enter fullscreen to ensure our root window - (cherry picked from commit 7617d3659b5481ef85aecc1f936e7dd2662bdfbd) - (cherry picked from commit 61ae56f97326c57dda05632ca9f4873238ee18e1) + displayid: Implement mode decoding. -commit beafebef82687d777d88152c62a351954df7c864 -Author: Jeremy Huddleston -Date: Tue Dec 23 20:33:27 2008 -0800 +commit 2f1a9c5baa367818bf017bdb72f20a2f6fa7ac21 +Author: Adam Jackson +Date: Tue Jun 9 10:13:47 2009 -0400 - XQuartz: Re-enable rlAccel + ddc: s/xf86DDCMonitorSet/xf86EdidMonitorSet/ - It was incorrectly blamed for some crashes a few months ago, and it should be - safe to use until we get an EXA driver to replace it. - (cherry picked from commit 8121f30bd29591fc74fbc680fbbf20210271fa58) - (cherry picked from commit 85347902d97f2d4937f63ae1fac62ee46a61c82f) + Since we need a second path for DisplayID. -commit 86ba5eff623336592091b990209eb07b3f74f8b9 -Author: Jeremy Huddleston -Date: Sun Dec 21 21:03:59 2008 -0800 +commit d0cb4f5a91932e901d10cac5f2a4ba12bb8a0e6f +Author: Adam Jackson +Date: Tue Jun 9 10:10:18 2009 -0400 - XQuartz: Don't use keycode 0 to determine !swallow since our most common key to swallow is actual keycode 0 (a) - (cherry picked from commit 33f43a7f03023bfbab25a957cb81fc25b4afa4ca) - (cherry picked from commit 84f0c03418bf74188596635dcac128fc05f204ad) + ddc: Refactor root window property code -commit 2a42a059ce7b259622b36fb3ef909f289d5388d7 -Author: Jeremy Huddleston -Date: Sun Dec 21 20:40:36 2008 -0800 +commit 14103b781201bc36896cbe9112a2e0d991fb785d +Author: Adam Jackson +Date: Tue Jun 9 10:05:01 2009 -0400 - XQuartz: Update our "screens" when we toggle rootless rather than when we toggle fullscreen - - This old behavior was used as a workaround for the menubar behavior in the older server, - but we handle it better now and need to update our screens when we toggle the rootless - state instead. - (cherry picked from commit 508aa95bc2cd3fdc3dff448ec090919bf807d153) - (cherry picked from commit 7e7758e1780326ad867be74dbd583a154bad017b) + ddc: Don't try to publish a root window property for DisplayID -commit 51319550c9bb079c3af1b98c96c5174bc74e0dbd -Author: Jeremy Huddleston -Date: Sun Dec 21 14:54:11 2008 -0800 +commit c302a5ff49146bff24df196fc36ed38745d42911 +Author: Adam Jackson +Date: Mon Jun 8 17:42:10 2009 -0400 - XQuartz: Run applications via '/bin/sh -c ...' to support users who expect shell parsing - (cherry picked from commit 67455e716e3ecffd528930479192785958d37988) - (cherry picked from commit 7dc0dafef1d241d396f215c506ec2d4f7d8e3a24) + ddc: Skeleton for xf86DoDisplayID() -commit b0503b69f998bcf5398ed0382a38346a3f18a77c -Author: Jeremy Huddleston -Date: Sun Dec 21 14:43:12 2008 -0800 +commit 0b36f68efb1171fcdbe53e93064394f5609b7fb5 +Author: Adam Jackson +Date: Mon Jun 8 15:36:15 2009 -0400 - XQuartz: pbproxy: Release display notification lock when not needed to avoid deadlock - (cherry picked from commit 22f664ab95a0cae981e9cefad6f075d051583ca5) - (cherry picked from commit 4c256c0e9c4fe61990343c8bcf2e352c83d76e69) + ddc: mv xf86DDC.c ddc.c -commit 6c635faa6ff0474199f4f7375022efe1e8ffa8f1 -Author: Jeremy Huddleston -Date: Sun Dec 21 14:42:26 2008 -0800 +commit 8eb82168fc5c7ea958a4f63676738510647dd203 +Author: Adam Jackson +Date: Mon Jun 8 15:34:42 2009 -0400 - XQuartz: update quoting in case X11.app is moved to a directory with a space. - (cherry picked from commit cc805dc799efa37c8dcefa3db04d87e9b835ffbd) - (cherry picked from commit ecc3a7b6090552c309fe8e264d527ddd666a5761) + ddc: Give DisplayID a place to hang its hat -commit 14d9ccf0ce4c5b4820e692cb58077c4c5918f12a +commit ace0fe09aee48d57cd0079260cd8d20d041e8eb6 +Author: Adam Jackson +Date: Mon Jun 8 14:45:47 2009 -0400 + + ddc: Yet more code motion + +commit 3a350688bbe1257feaf1502b4009f5f701d2640e +Author: Adam Jackson +Date: Mon Jun 8 14:37:38 2009 -0400 + + ddc: Code motion to eliminate forward decls + +commit 8797831f82637b0e65a08e1262d1ec57c075cc12 +Author: Adam Jackson +Date: Mon Jun 8 14:30:53 2009 -0400 + + ddc: Update a comment. + +commit bb6fa39eb3aa044ffc18632288fd59909ff06344 +Author: Adam Jackson +Date: Thu Jul 16 17:43:31 2009 -0400 + + dbe: Adapt to new headers + +commit 9965174f4f09a7de00a9569607e96226208bc2ce +Author: Pierre Willenbrock +Date: Thu Jul 16 12:12:48 2009 +0200 + + Check dev->u.master if there is a custom event handler, too + + Signed-off-by: Peter Hutterer + +commit c806162c133603a99d9cd844bb04485bb663707d +Author: Pierre Willenbrock +Date: Tue Jul 14 13:42:19 2009 +0200 + + Always update u.lastSlave + + Signed-off-by: Peter Hutterer + +commit 4db8950b1dfc3150ee2e9f1f975e9ecb4eabb1f2 +Author: Peter Hutterer +Date: Fri Jul 17 10:40:59 2009 +1000 + + Xext: include securproto.h instead of securstr.h + + Reported-by: Byeong-ryeol Kim + Signed-off-by: Peter Hutterer + +commit b5f6da1e9b37f52484948185dcf17950657cd65b +Author: Oliver McFadden +Date: Fri Jul 17 06:34:00 2009 +0300 + + Revert "Coverity Prevent: RESOURCE_LEAK in AccelSetProfileProperty:" + + This reverts commit 7333dc2969f60af0abcfb28e7182a5fff9918223. + + False positive. + +commit f517fca31d8c341ad36f3de4863adb0bc5206176 +Author: Jeremy Huddleston +Date: Thu Jul 16 17:28:56 2009 -0700 + + XQuartz: Localization updates + (cherry picked from commit 4b797fc1edf2bd963410a3133e3d2182ccfda4c3) + +commit c8a35165147a9b2f747b25da80d0fdb4279a8331 +Author: Jeremy Huddleston +Date: Thu Jul 16 17:20:16 2009 -0700 + + XQuartz: Cleanup getGlCapabilities to avoid hardcoding the number of displays + (cherry picked from commit 12f7365f1f58d648217b16f09c2152fa47dcd7a1) + +commit e7bc9ff816c1848c700a376908a1411f1e20d29f +Author: Jeremy Huddleston +Date: Thu Jul 16 16:59:47 2009 -0700 + + XQuartz: Allow more than 3 OSX displays + (cherry picked from commit 45045eb3968069a8d74ce9188890a5537dc60fec) + +commit 7333dc2969f60af0abcfb28e7182a5fff9918223 +Author: Oliver McFadden +Date: Thu Jul 16 18:57:49 2009 +0300 + + Coverity Prevent: RESOURCE_LEAK in AccelSetProfileProperty: + + Event alloc_arg: Called allocation function "XIPropToInt" on "ptr" [details] + 167 rc = XIPropToInt(val, &nelem, &ptr); + + Event leaked_storage: Variable "ptr" goes out of scope + +commit 66eabbebaf1b40fed7670b7c05337ed6226dd81e +Author: Oliver McFadden +Date: Thu Jul 16 18:50:22 2009 +0300 + + Coverity Prevent: RESOURCE_LEAK in xf86CrtcSetInitialGamma + +commit 464e8ad733fa6afee028607d6e7d4663b4c273cc +Author: Oliver McFadden +Date: Thu Jul 16 17:51:08 2009 +0300 + + Coverity Prevent: NEGATIVE_RETURNS in fbdev_open_pci: + + Event var_tested_neg: Variable "fd" tested NEGATIVE + At conditional (1): "fd != -1" taking false path + 335 if (fd != -1) { + + Event negative_returns: Tracked variable "fd" was passed to a negative sink. + 347 close(fd); + +commit 53ae6b63387e04324b23b6f8311cc22a154c1fb8 +Author: Jeremy Huddleston +Date: Wed Jul 15 23:18:35 2009 -0700 + + XQuartz: Cleanup keymap locking, fix a possible synchro bug + (cherry picked from commit 33e7437a4984ee7c1b04b87d23dee7d4739d5f12) + +commit 6a90c7b93724a2d26eae70b5806ca06c91e7df4c +Author: Peter Hutterer +Date: Thu Jul 16 09:28:42 2009 +1000 + + xkb: cosmetic fix, use TRUE instead of True. + + Rest of InitKeyboardDeviceStruct uses TRUE and FALSE. + + Signed-off-by: Peter Hutterer + +commit 52088d3c2d53a91b2690f9cc402f839d77769e37 +Author: Éric Piel +Date: Sun Jun 21 18:11:28 2009 +0200 + + xserver: remove unused code in clipValuators + + The axes variables was never used, remove it. + + Signed-off-by: Peter Hutterer + +commit d3f6b43a240eb763025b3cbf546cb7ae502c94fa +Author: Peter Hutterer +Date: Wed Jul 15 16:51:05 2009 +1000 + + Update to xextproto 7.0.99.1. + + xextproto had Xlib client headers moved into libXext. + Protocol header files are named fooproto.h, header files with constants + foo.h or fooconst.h where foo.h was already in use for client-side headers. + +commit 693babbf12cc7969c3e211c4037c7af0d41c13e9 +Author: Peter Hutterer +Date: Wed Jul 15 12:30:02 2009 +1000 + + xkb: Remove XKMformat.h include from xkbsrv.h into the files that need it. + + xkbsrv.h is used by drivers, they don't need the XKM format and shouldn't + require it. + + Signed-off-by: Peter Hutterer + +commit 35ff5cd26eb7564fefebf238e30b8d43cbb4dc25 +Author: Peter Hutterer +Date: Wed Jul 15 10:29:46 2009 +1000 + + dix: fix wrong raw valuator copy + + internal events keep valuator data at the index for the valuator, not like + the wire events that start with first_valuator. + + Signed-off-by: Peter Hutterer + +commit 975bf60a82f863e3bdc36fc3f8201b48a4d1e6f7 +Author: Peter Hutterer +Date: Wed Jul 15 10:29:19 2009 +1000 + + dix: use sizeof(FP3232) instead of 2 * sizeof(int32_t). + + Signed-off-by: Peter Hutterer + +commit d040af7fa3c7314917414a28d723bdda3c4289c3 +Author: Peter Hutterer +Date: Mon Jul 13 15:09:38 2009 +1000 + + Update to type-specific raw events - require inputproto 1.9.99.14. + + Signed-off-by: Peter Hutterer + +commit 81b64f668541217c1c9518ac4a7e3fdf0a6e4002 +Author: Fredrik Höglund +Date: Wed Jul 15 00:38:01 2009 +0200 + + Require renderproto >= 0.11 and pixman-1 >= 0.15.14 for the new blend modes. + +commit 0ce42adbf4cff9e7f049d9fc79d588ece5936177 +Author: Fredrik Höglund +Date: Wed Jul 15 00:37:05 2009 +0200 + + Render: Add support for the PDF blend mode operators. + +commit 686e4867302a741f3029c4105b997d0f0ac7c13c +Author: Tiago Vignatti +Date: Tue Jul 14 17:11:13 2009 +0300 + + xfree86: remove a bunch of unused pci headers + + Signed-off-by: Tiago Vignatti + +commit cc575a3ba4a52265e410b325c2291fe900a54f33 +Author: Peter Hutterer +Date: Fri Jun 12 11:38:46 2009 +1000 + + test: add tests for ProcXIGetClientPointer. + + Signed-off-by: Peter Hutterer + +commit 92b0065a19dcde50494d2528e19a4ee76723b3da +Author: Peter Hutterer +Date: Fri Jun 12 11:38:00 2009 +1000 + + test: add ProcXISetClientPointer tests. + + Signed-off-by: Peter Hutterer + +commit 019bc9d70c5b3beb69ca20b6bca3551e00415992 +Author: Peter Hutterer +Date: Fri Jun 12 11:37:17 2009 +1000 + + test: add ProcXIGetSelectedEvents tests. + + Signed-off-by: Peter Hutterer + +commit 8579d39c12b065d8e82fe75af4b254c8ad4f4ab8 +Author: Peter Hutterer +Date: Fri Jun 12 11:36:22 2009 +1000 + + test: add ProcXISelectEvents tests. + + Signed-off-by: Peter Hutterer + +commit 60b7a7990003a22be5cc26ebe4d8d2d0803b97de +Author: Peter Hutterer +Date: Fri Jun 12 11:33:22 2009 +1000 + + test: add ProcXIQueryDevice tests. + + Signed-off-by: Peter Hutterer + +commit 441c91be05912e2665cc1d3ee72fb70dcd32e7a2 +Author: Peter Hutterer +Date: Fri Jun 12 11:32:15 2009 +1000 + + test: Add ProcXIQueryVersion tests. + + Signed-off-by: Peter Hutterer + +commit 1d635ae65b217e124b3bd5a8844e73f447eb1b79 +Author: Peter Hutterer +Date: Fri Jun 12 11:30:53 2009 +1000 + + test: add xi2 directory, prepare for protocol testing. + + These two files provide a couple of common defines, functions and variables + that will be used in a number of protocol tests. + + Signed-off-by: Peter Hutterer + +commit 4b93413b2a490a74de2e27b37268332b609ac872 +Author: Peter Hutterer +Date: Tue Jul 14 11:20:01 2009 +1000 + + Add test subdir to base Makefile.am + + If unittests are enabled, make will build those as well - spotting potential + build errors in the tests faster. Furthermore, this allows for the tests to + be run from the top-level directory. + This patch removes the "run make check to build the test suite" message + since that'd pop up after every build now. + + If unittests are disabled, this change has no effect. + + Signed-off-by: Peter Hutterer + +commit e2226f3cd7d1bef598f9657b756a171b02f1a299 +Author: Peter Hutterer +Date: Mon Jul 6 14:57:04 2009 +1000 + + dbe: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 3e1a054423c22f0e35f25127dde1ea8263892480 +Author: Peter Hutterer +Date: Mon Jul 6 14:24:45 2009 +1000 + + record: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 089c460058c4e6814d98d40179eb4c731f37fa80 +Author: Peter Hutterer +Date: Mon Jul 6 14:22:40 2009 +1000 + + xkb: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 0dc11da57562a7bbb6830ab657a2b818c596a537 +Author: Peter Hutterer +Date: Mon Jul 6 12:41:29 2009 +1000 + + ephyr: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 34bfaa9d9ecd90cfe8413bc275179fdcc193eab3 +Author: Peter Hutterer +Date: Mon Jul 6 12:29:26 2009 +1000 + + xfixes: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit e216527107fda470b92b7e526f3db22465962a43 +Author: Peter Hutterer +Date: Mon Jul 6 12:27:23 2009 +1000 + + render: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 1d6b71b8d0dc24355e84391ba413170a03ccdbf4 +Author: Peter Hutterer +Date: Mon Jul 6 12:25:39 2009 +1000 + + os: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 86b239ff9c4d01685c357ca2b1ef761d167e3224 +Author: Peter Hutterer +Date: Mon Jul 6 12:24:27 2009 +1000 + + randr: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 7b9e84e320a6f6449fe7bc58a8d6a094ae37b86c +Author: Peter Hutterer +Date: Mon Jul 6 12:20:21 2009 +1000 + + Xext: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit aa19d355125a10b1a385c8f134d68e79d3d609c7 +Author: Peter Hutterer +Date: Fri Jul 10 14:25:22 2009 +1000 + + xwin: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 0b4e6af857bcc5513e4c19912a54656d4696e56d +Author: Peter Hutterer +Date: Mon Jul 6 12:12:57 2009 +1000 + + xquartz: switch to byte counting functions + + Signed-off-by: Peter Hutterer + Signed-off-by: Jeremy Huddleston + +commit ad508c93c239a5ba8381000c031e96caf2769265 +Author: Peter Hutterer +Date: Mon Jul 6 12:12:20 2009 +1000 + + xfree86: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit c20304226b3ca2f8d0a4f4866480b0d71913941c +Author: Peter Hutterer +Date: Mon Jul 6 12:07:00 2009 +1000 + + glx: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit 87ce93c9973067255b4197d82772f83cd4ea5d27 +Author: Peter Hutterer +Date: Mon Jul 6 12:06:47 2009 +1000 + + dmx: switch to byte counting functions + + Signed-off-by: Peter Hutterer + +commit dc82e11e509ecf586d77c3e7c1325d515509ce51 +Author: Peter Hutterer +Date: Mon Jul 6 12:03:34 2009 +1000 + + exa: switch to byte counting functions. + + Signed-off-by: Peter Hutterer + Acked-by: Michel Dänzer + +commit 2d35ea8d957a955e1200ba2b14424bddfe1f4148 +Author: Peter Hutterer +Date: Fri Jul 3 16:50:03 2009 +1000 + + dix: switch to byte-counting functions. + + Signed-off-by: Peter Hutterer + +commit 7dd415aa6a3959f15276741db168ba264948ecfe +Author: Peter Hutterer +Date: Fri Jul 3 13:57:14 2009 +1000 + + Xi: use byte-counting macros instead of manual calculation. + + Signed-off-by: Peter Hutterer + +commit 912402fd71144bcee255141efe12a78abad39240 +Author: Peter Hutterer +Date: Mon Jun 29 13:09:57 2009 +1000 + + include: introduce byte counting functions. + + This patch adds the following three functions: + bits_to_bytes(bits) - the number of bytes needed to hold 'bits' + bytes_to_int32(bytes) - the number of 4-byte units to hold 'bytes' + pad_to_int32(bytes) - the closest multiple of 4 equal to or larger than + 'bytes'. + + All three operations are common in protocol processing and currently the + server has ((foo + 7)/8 + 3)/4 operations all over the place. A common set + of functions reduce the error rate of these (albeit simple) calculations and + improve readability of the code. + + The functions do not check for overflow. + + Signed-off-by: Peter Hutterer + +commit 2c535b6f13ffbf2c4ac59834dae39bb8e172c003 +Author: Peter Hutterer +Date: Tue Jul 14 08:58:23 2009 +1000 + + dix: don't send presence events for attaching/detaching slave devices. + + The code that didn't list attached slave devices for XI1 clients doesn't + exist anymore, so there's no need for these presence events. + + Signed-off-by: Peter Hutterer + +commit 0c0ef42292f4c910c73b308cd75d77637312da53 +Author: Peter Hutterer +Date: Sun Jul 12 21:43:06 2009 +1000 + + dix: Remove temporary detachment of slave devices. + + Previously, an active grab on an attached slave device would send the device + floating for the duration of the grab. This breaks existing XI applications + (e.g. the GIMP) since they grab all devices automatically - resulting in the + loss of control over the VCP. + + The behaviour of extended input devices during a grab in relation to the + core pointer is not specified in the XI protocol specification. + The removal of the temporary detachment restores the behaviour of extended + input devices as present in currently released servers - even if a device is + grabbed, an event from this device will result in an event from the core + pointer. + + Signed-off-by: Peter Hutterer + +commit 9f1570c8f4f549cdd2fbae1234011290fcc73e18 +Author: Peter Hutterer +Date: Fri Jun 26 09:12:47 2009 +1000 + + input: include effective modifiers in device events. + + Require inputproto 1.9.99.13 + + Signed-off-by: Peter Hutterer + +commit 78aedbe609e309446e71835ed15e4557d3ccf723 +Author: Alan Coopersmith +Date: Fri Jul 10 17:35:57 2009 -0700 + + Fix int10 module build after typo in commit 2638e9899 + + Signed-off-by: Alan Coopersmith + +commit 35758544813f156eaac28844e693b2a28f6de316 +Author: Michel Dänzer +Date: Fri Jul 10 20:06:02 2009 +0200 + + EXA: Only pass CT_YXBANDED to RECTS_TO_REGION() if that is really true. + + Fixes http://bugs.freedesktop.org/show_bug.cgi?id=22642 . + +commit 2638e9899e57d9b25a2cade3247083a6e101dee2 +Author: Tiago Vignatti +Date: Fri Jul 10 16:09:08 2009 +0300 + + configure: introduce --{enable,disable}-int10-module + + Not all chipsets need to rely on the int10 scheme to do its daily work. + + Well, the ideal would be to remove all int10 module from the server. I'll try + to provide some patches "soon" for this. Something like: + http://cgit.freedesktop.org/~vignatti/libx86/ + + Signed-off-by: Tiago Vignatti + Acked-by: Oliver McFadden + +commit 42d95a38535f382bfb2ddfaef93839ab9bbf584f +Author: Tiago Vignatti +Date: Fri Jul 10 15:54:07 2009 +0300 + + configure: introduce --{enable,disable}-vbe + + Not all drivers need this kind of access as well. + + Signed-off-by: Tiago Vignatti + Acked-by: Oliver McFadden + +commit 53d64930513fecaa417bb5a922966b45c9ff8679 +Author: Tiago Vignatti +Date: Fri Jul 10 15:38:51 2009 +0300 + + configure: introduce --{enable,disable}-vgahw + + Not all drivers need this kind of access. + + Signed-off-by: Tiago Vignatti + Acked-by: Oliver McFadden + +commit aafcae79d17c1f802bc880d2142af7171fed75d8 +Author: Tiago Vignatti +Date: Fri Jul 10 14:56:31 2009 +0300 + + xfree86: remove unused vbe header file + + Signed-off-by: Tiago Vignatti + Acked-by: Oliver McFadden + +commit 3711d68f657c77b947cc4670cc4eac4f62de3af8 +Author: Peter Hutterer +Date: Fri Jul 10 11:10:30 2009 +1000 + + Revert "XKB: Sanitise * actions" commits (#19602) + + Reverts the following four patches: + + feb757f384382c7782ceac55 "XKB: Sanitise vmods for redirected keys" + b5f49382fe48f0a762d9a15f "XKB: Sanitise ctrls action" + 1bd7fd195d85681e722161f8 "XKB: Sanitise pointer actions" + 61c508fa78aa08ea2666fde9 "XKB: Sanitise vmods in actions" + + Strictly speaking, the structs used in the server are not part of the client + ABI. Practically, they are as we copy from the wire straight into the + structs. Changing the struct sizes breaks various wire/server conversions. + + Even when the structs have the same size, some internal magic causes + conversions to fail. Visible by diffing the output files of: + setxkbmap -layout de; xkbcomp -xkb :0 busted.xkb + setxkbmap -layout de -print | xkbcomp -xkb - correct.xkb + + Interestingly enough, busted.xkb is the working one although the output is + incorrect. Revert the four offending patches until the exact cause of this + breakage can be determined. + + This patch restores functionality to Level3 modifiers. + + X.Org Bug 19602 + + Signed-off-by: Peter Hutterer + +commit 08df24555cb432eb0d90a3f63275e9485e777c4c +Author: Soren Sandmann Pedersen +Date: Thu Jul 9 17:23:04 2009 -0400 + + Reserve space for two GC values in copy_drawable(). + + Pointed out by Pierre Willenbrock. + +commit 3c53f3241319d96affad2f4539f7661e02e80521 +Author: Adam Jackson +Date: Thu Jul 9 17:21:07 2009 -0400 + + Since font modules are dead, don't mention them in xorg-server.pc + +commit e812103382a7bc3f7ba8b2f44ab97960dec265a9 +Author: Adam Jackson +Date: Thu Jul 9 10:58:17 2009 -0400 + + config: drop i810. + + The driver assuredly no longer builds against 1.7. Please don't feel + compelled to fix it. + +commit 55ac155d42268009eaecbc183329a8744d6ae48c +Author: Alan Coopersmith +Date: Tue Jul 7 16:59:39 2009 -0700 + + Xephyr & Xserver man page fixes + + Xephyr(1): Fix quote formatting, add missing ' to contraction + Xserver(1): Add Xephyr(1) & startx(1) to SEE ALSO section + + Signed-off-by: Alan Coopersmith + +commit 36dc66ae79b6e1cbe0a23587e32aebb68365b0d0 +Author: Michael Lorenz +Date: Tue Jul 7 20:21:39 2009 -0400 + + The way XaceHook() mixes struct initializers and va_arg() is not portable and + gives bogus data on sparc and probably others leading to a crash. + + Fix: + Don't use initializers, instead set each member directly to enforce order. + + Signed-off-by: Eamon Walsh + +commit 48a9d65b88f56d1f8ab3bf824a4fe48c2f68725f +Author: Jon TURNEY +Date: Sun Jul 5 16:06:32 2009 +0100 + + Cygwin/X: Window positioning improvements for multiwindow mode + + A few tweaks to winUpdateWindowPosition(): + + * Don't allow window decoration to disappear off to top or left edge of the + display as a result of adjustments to add decoration for the window style + + * Honour the position requested by window geometry for the client area (so + windows which save their position don't get moved by the decoration width + every time they are created) (unless we need to bump it away from top or left edges) + + * Fix an issue with initial window placement being offscreen on multimonitor + systems when some monitors have negative coordinates (are to the left of or + above the primary monitor) + + Signed-off-by: Jon TURNEY + +commit 5aec72745232dd61e60cfbf3acc4628d4fcd0315 +Author: Keith Packard +Date: Mon Jul 6 11:53:02 2009 -0700 + + Remove old DRI2 buffer alloc/free interfaces + + These old interfaces are no longer supported by the server, removing them + requires bumping the video driver ABI. Note that this is not guaranteed to + be the last change in ABI version 6. + + Signed-off-by: Keith Packard + Reviewed-by: Eric Anholt + +commit 2e2c5b216cc1c7a9bc26bd2c68226aaed5fc52ca +Author: Keith Packard +Date: Wed Jul 1 14:01:57 2009 -0700 + + dri2: Preserve compatibility with 1.6 DRI2 API/ABI + + The old DRI2 buffer allocation API wasn't great, but there's no reason to + make the server stop working with those drivers. This patch has the + X server adapting to the API provided by the driver, using the new API where + available and falling back to the old API as necessary. A warning will be + placed in the log file when the old API is in use. + + Signed-off-by: Keith Packard + +commit 7c7f0c2c6a04f7044d5ce69e97a615735e5831f1 +Author: Søren Sandmann Pedersen +Date: Fri Jul 3 14:36:54 2009 -0400 + + Use IncludeInferiors when copying windows before compositing. + + Part of bug 22484. + +commit 43ee8d2ead862f84a4526a472519663ef27a8d6a +Author: Matthias Hopf +Date: Wed Jun 24 18:26:23 2009 +0200 + + Unclaim PCI slot if driver probing fails. + + Otherwise no subsequent driver will be able to claim this pci slot. + Example for this: fbdev tries to claim, but framebuffer device is not + available. Later on VESA cannot claim the device. + +commit 73abdc94c3cceadeda26a9b6bd3cdfecf0df8db2 +Author: RALOVICH, Kristóf +Date: Sun Jul 5 16:42:54 2009 +0200 + + glx: damage is only used with DRI + + Signed-off-by: Tiago Vignatti + +commit 7208a0f032d487bd3bc9809800f4557f1da1e2db +Author: RALOVICH, Kristóf +Date: Mon Jun 29 15:18:56 2009 +0200 + + glx: remove Xgl leftover + + GlxSetRenderTables was only used by the long gone Xgl. + + Signed-off-by: Tiago Vignatti + +commit 89cf81cd85919e3dbb5adff5e6c6056c7990b60f +Author: Peter Hutterer +Date: Fri Jul 3 10:29:40 2009 +1000 + + Xext: return BadValue for XTestFakeInput on unsupported capabilities. + + Calling XTestFakeDevice*Event on a device that doesn't allow the matching + event returns BadValue. + + Reported-by: Florian Echtler + Signed-off-by: Peter Hutterer + +commit b773b4e8e0087993406d1d8c2df895db4d6301ff +Author: Peter Hutterer +Date: Mon Jul 6 11:04:47 2009 +1000 + + s/MAX_DEVICES/MAXDEVICES/ updates. + + The number of input devices is MAXDEVICES, not MAX_DEVICES (f781a752e6) + Two comments updated to refer to MAXDEVICES. + + MAX_FUNCS in sigio.c was set to 16 if MAX_DEVICES was undefined. If more + than 15 physical input devices were present, this could result in a + failure to install the SIGIO handler for any device above 15. + + Signed-off-by: Peter Hutterer + +commit 183c075d2f9d5f6effa1ce7ab135fb0c2e46085a +Author: Peter Hutterer +Date: Fri Jul 3 13:26:48 2009 +1000 + + dix: always init the full button map to default values (#22594) + + Master devices must have the standard button map applied for all buttons to + ensure buttons larger than 7 (the default for MDs) are mapped appropriately. + + We can't copy the button map from SDs to MDs since that breaks the chained + button mapping. However, by ensuring all buttons (even non-existing ones) + are mapped, devices that send such buttons continue to work. + + X.Org Bug 22594 + + Signed-off-by: Peter Hutterer + +commit 744bb559826ede37a77e9000b6c620eaa6a3c837 +Author: Peter Hutterer +Date: Fri Jul 3 13:58:38 2009 +1000 + + Xext: remove unused variable 'it'. + + Signed-off-by: Peter Hutterer + +commit fd4eed69e40cf090aa16e79a8596eb1442461829 +Author: Eric Anholt +Date: Thu Jul 2 15:24:54 2009 -0700 + + dri2: Enable GLX_SGI_make_current_read when the DRI driver supports it. + + This matches idr's 82f150d73cc9a7d8eaf7241a51b03af05bcec159 for DRI1. + +commit d6b8205e699c0c62af76c4a9cbff1402337927b3 +Author: Jeremy Huddleston +Date: Fri Jul 3 19:25:33 2009 -0700 + + XQuartz: ProcAppleWMAttachTransient to play nice with the new Dock in SL + (cherry picked from commit ddc0242d8f291d0f961ba9aa8c883e39861ce2ad) + +commit a49ae50370ec94e08e7dec3c742d33e20e79ef36 +Author: Colin Harrison +Date: Mon Jun 29 22:55:52 2009 +0100 + + Xming: Apply window style hints in -multiwindow mode + + Remove frames from 'dock' windows and make them topmost in -multiwindow mode. + Remove frames from windows with MOTIF_WM_HINTS of no decorations in -multiwindow mode. + Apply some _NET_WM_STATE hints in -multiwindow mode. + Apply window styles overrides from .rc file + Correctly update region of shaped windows when applying styles + + Copyright (C) Colin Harrison 2005-2009 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + +commit 17e67c407d130c325d3899c18d68b8eef6a88bea +Author: Joe Krahn +Date: Wed Jun 24 23:32:09 2009 +0100 + + Cygwin/X: Change to a single native window class for all X windows + + from fd.o Bugzilla #4491: + + There is no point in having one class for every window, aside from trying to + set custom icons via the class, which we no longer do, so converted to using + a single class for all client windows. + + Signed-off-by: Jon TURNEY + +commit 2c69deb92e11542f615df0f24fdc03e3b4415475 +Author: Rémi Cardona +Date: Fri Jul 3 10:51:50 2009 +0200 + + configure: libXinerama isn't needed anymore + + since libXinerama commit 90d4d23bf2e94721149ddc0a80093b10a82e8845 and + xineramaproto commit 21477147613c28c968b5e1eb9d8aea7017dd399d, the + server no longer needs libXinerama. + + Signed-off-by: Rémi Cardona + +commit 97e29ffb5bf04a9d8280f1fa32ceced148503492 +Author: Peter Hutterer +Date: Fri Jul 3 09:52:04 2009 +1000 + + xfree86: fix wrong IsMaster() check causing crashes. + + Crashes caused by dereferencing NULL if the path was executed for a floating + slave device. + + Signed-off-by: Peter Hutterer + +commit 50a2a8dc76645d8736f7d712f0ef05f23089407e +Author: Peter Hutterer +Date: Fri Jul 3 09:18:57 2009 +1000 + + Fix IsXtstDevice - returns false positives since 0814f511d5. + + Missing check for the value of 'mid' returned false positives if master was + NULL. + + Signed-off-by: Peter Hutterer + +commit 3525d140567e0ad5f0184e4b37893c47239e1628 +Author: Felix Kuehling +Date: Tue Jun 30 20:07:30 2009 -0400 + + Export symbols needed by the RandR implementation in fglrx + + GIT change + http://cgit.freedesktop.org/xorg/xserver/commit/?id=45c8bd0fe54273039fdaa1eeeafb81b5774f2c75 + changed the default symbol visibility of the Xserver. As a result 2 symbols + that are needed by the RandR 1.2/1.3 implementation in the fglrx driver are no + longer visible: + + xf86configptr + xf86CursorScreenKey + + We would like to get these two symbols _X_EXPORT'ed before Xserver 1.7 is + released. Otherwise it will be problematic for fglrx to support RandR 1.3 on + Xserver 1.7. + + In the future, we may want to sync our RandR implementation to later versions + of the RandR implementation in hw/xfree86/modes. Therefore it would be nice if + all symbols used by the Xserver RandR implementation were _X_EXPORT'ed in the + future. + +commit e13605ea40cfc671314a0086c75b917564298b55 +Author: Peter Hutterer +Date: Tue Jun 23 14:09:22 2009 +1000 + + dix: introduce "Xtst Device" label property. + + Xtst devices get this property assigned automatically so they can be + detected easily by a client. + The property is read-only. + + Signed-off-by: Peter Hutterer + +commit 0814f511d56a89c7b1868b17eba7a89f990b9ab1 +Author: Peter Hutterer +Date: Tue Jun 23 10:50:52 2009 +1000 + + input: store the master device's ID in the devPrivate for XTest devices. + + Rather than storing a simple boolean in the devPrivate for XTest devices, + store the actual master device's id (since it is constant for the life of + the device anyway). + + Callers should use GetXtstDevice now instead of digging around in the + devPrivates themselves. + + This patch allows for a cleanup in the creation of new master devices since + GetMaster and GetXtstDevice spare the need for loops, IsPointer checks and + similar. + + Signed-off-by: Peter Hutterer + Acked-by: Benjamin Close + +commit 1bcc0d3c244ce7d9f5cbab626aa5fd5784b41a1c +Author: Peter Hutterer +Date: Mon Jun 22 16:55:12 2009 +1000 + + input: abstract Xtst device lookup + + The callers should need to use the dev privates key to look up xtest + devices. + + Signed-off-by: Peter Hutterer + Acked-by: Benjamin Close + +commit 34424fab9abd7a4ca11036be25414129980db0e0 +Author: Peter Hutterer +Date: Tue Jun 23 11:50:29 2009 +1000 + + dix: set the client's error value to the bad deviceid in check_butmap_change. + + Signed-off-by: Peter Hutterer + +commit da04e8f1354fa9d253443489dc002b16f94d6fab +Author: Eamon Walsh +Date: Tue Jun 30 17:52:42 2009 -0400 + + dix/property.c: use memcpy where appropriate. + + Signed-off-by: Eamon Walsh + +commit 65d74d93145d22b68bad5728a7ebe38dc662cb21 +Author: Jon TURNEY +Date: Wed May 20 21:32:54 2009 +0100 + + Cygwin/X: Fix multiwindow extwm mode to build again + + Build and link with rootless extension + Update Xwin code for removal of RootlessAccelInit() + Fix Xwin code which now has a collision with the type name EventType + + Based on patches from Colin Harrison, Jon Turney and Yaakov Selkowitz + + Signed-off-by: Jon TURNEY + +commit 638ca9a7a2363757dc5b5d456e10d34f6f158885 +Author: Jon TURNEY +Date: Mon May 18 18:14:41 2009 +0100 + + Cygwin/X: Avoid a visual glitch on window move in rootless modes + + Handle and ignore WM_ERASEBKGND since we repaint the entire invalidated region anyhow + (this avoids a white flickering on window resize) + + Signed-off-by: Jon TURNEY + +commit b718b2e0880cf2b969675da98d5ef8a4a01ca5d6 +Author: Jon TURNEY +Date: Mon May 18 18:14:09 2009 +0100 + + Cygwin/X: Allow pointer warping to work in rootless modes + + Mouse pointer warping in multiwindow/rootless mode was never implemented, + due to concerns that moving the mouse pointer without asking might be rude + + This patch allows X applications to move the mouse pointer in rootless modes, + Let's hope they don't abuse this privilege ;-) + + Signed-off-by: Jon TURNEY + +commit 85614946ba3d5a233eece612afc7f09572a347c2 +Author: Jon TURNEY +Date: Mon May 18 18:13:08 2009 +0100 + + Cygwin/X: Improve mouse tracking for moving/resizing undecorated windows + + In -multiwindow mode, tell Windows we wish to capture the mouse when a button + is down. This causes Windows to continue to send movement events for the mouse + even if the mouse pointer moves outside the window frame. + + This helps greatly with undecorated windows which have regions you can grab + to move (e.g. gmplayer, xine control panels) or resize (e.g. Songbird) the + window, as it means the window continues to receive the mouse motion even if the + mouse pointer moves out of the window (which presumably happens if we don't + manage to update the window fast enough to track the mouse pointer) + + Consolidate the multiple instances of the code to start the mouse position + polling timer into a new function winStartMousePolling(), and use that to + restart the polling timer when we release the mouse. + + Also, start the timer on WM_SHOW, so that xeyes will track the mouse position + when it is first shown, even if the mouse doesn't enter it's window + (You probably need focus-stealing turned off to see this problem) + + Signed-off-by: Jon TURNEY + +commit f351c10a9774cc0ea2cbb58f00f07ece7f7c6e73 +Author: Yaakov Selkowitz +Date: Mon Jan 5 16:17:30 2009 +0000 + + Cygwin/X: Man page fixes + + Correct path names in man pages, using cpprules.in + Install XWinrc man page into section 5 + + Signed-off-by: Jon TURNEY + +commit 5c1afac5eea1d8327c74342d12d082b75f0cebde +Author: Adam Jackson +Date: Mon Jun 29 11:16:16 2009 -0400 + + EDID: Fix timing class names to match the spec + +commit bf0f3b8f2bc830c8bd6f8f9410b89394b8d96257 +Author: Peter Hutterer +Date: Mon Jun 29 13:31:02 2009 +1000 + + test: fix build error introduced by XINPUT_ABI 7 + + Signed-off-by: Peter Hutterer + +commit cbeb6a73c44143674a700d36a7e9804d41003a4f +Author: Thomas Jaeger +Date: Sat Jun 20 20:17:41 2009 -0400 + + dix: report subpixel coordinates for high-resolution devices + + Acked-by: Simon Thum + Signed-off-by: Peter Hutterer + +commit 069b4c5f0a38649de73d1e0f70ee81bf862399dd +Author: Simon Thum +Date: Sat Jun 20 18:57:22 2009 +0200 + + dix: make part of ptrveloc.h internal + + Though this is a SDK header, some functions are intended solely + for the server. + + Signed-off-by: Peter Hutterer + +commit 2830e8493757e0da1253fe5ab280562b84730e77 +Author: Simon Thum +Date: Wed Jun 24 11:33:19 2009 +0200 + + dix: rename pDev->dev, pVel->vel for consistency + + Signed-off-by: Peter Hutterer + +commit 373e8c960d00e2b8c2250dd3f66859b081e14854 +Author: Simon Thum +Date: Wed Jun 24 11:16:24 2009 +0200 + + dix: improve pointer acceleration API + + This makes the ptr accel api actually sensible from a driver + perspective, since it avoids superfluous device lookups. + Also, makes independent accel contexts possible. + + Signed-off-by: Peter Hutterer + +commit b8050bb6deebdb1ee60731f63884ffca575c09ce +Author: Jeremy Huddleston +Date: Sat Jun 27 15:28:34 2009 -0700 + + XQuartz: Don't leave zombied processes at startup + (cherry picked from commit 40c1406830588fa85d880e9f4e9ca570db1db306) + +commit aaff92c8c22a47804a21010d023ef76d82e7ec7e +Author: Jeremy Huddleston +Date: Sat Jun 27 14:40:03 2009 -0700 + + XQuartz: 64bit fix for screen origin in AppleWM + (cherry picked from commit 60a757d2802a5c34acd91ca9a052937b5a169ede) + +commit cabcc1a7ad697dde915794ddcea6a300f66b0a28 +Author: Jeremy Huddleston +Date: Sat Jun 27 10:40:59 2009 -0700 + + XQuartz: More localization updates + (cherry picked from commit 5925c1f48ad05bf94195b986c1fdefc52a20ae42) + +commit 31166c2ae0ce898c96995a8b16b58b127dc85a2f +Author: Eamon Walsh +Date: Fri Jun 26 16:51:22 2009 -0400 + + xace: add a new hook for checking property content after it has been set. + + Allows security modules to enforce what property contents can be set by + clients. Uses the new DixPostAccess bit to distinguish between the + existing call made during the lookup (with the old property data) and + this new call. Note that this only applies to writes, prepends, or + appends to existing properties; for new properties the existing + DixCreateAccess hook call may be used since it includes the new data. + + Refer to the XACE-Spec document in xorg-docs, section "Property Access." + + Signed-off-by: Eamon Walsh + +commit 51105de9b0d865c4b5e5a7d9ab23c89d808d1cfa +Author: Eamon Walsh +Date: Fri Jun 26 16:50:12 2009 -0400 + + xselinux: ignore property hook calls with the new Post access mode bit set. + + Signed-off-by: Eamon Walsh + +commit 10812204b3415c969bcebd3215e84d758a0b4dd8 +Author: Eamon Walsh +Date: Fri Jun 26 16:48:24 2009 -0400 + + dix: add a new DixAccess bit, "DixPostAccess". + + This will be used for follow-up checks after a client has written something, + for security modules that enforce a set of valid values a client can set. + + Signed-off-by: Eamon Walsh + +commit 9480725af0275b72b24b9e1cb3b68d5b97288d01 +Author: Rémi Cardona +Date: Wed Jun 24 10:12:50 2009 +0200 + + xfree86: DIST_SUBDIRS should not contain variables + + This patch reverts a change done in commit + 800d4ceb946c93724661ebfd28afb658ca63b6b1 to add conditional XAA support + + Signed-off-by: Rémi Cardona + +commit 84662e40c3d4141ebb298a1ad714f75056a4ab74 +Author: Eamon Walsh +Date: Thu Jun 18 23:19:21 2009 -0400 + + Xi: check for GetAttr permission when listing or querying devices. + + If the check fails, leave the device off the returned list of info + structures. Under XI2, this may cause inconsistent views of the device + topology after a change (for example, devices disappearing from view, + or showing as attached to a master that cannot be seen). More work is + needed to deal with topology changes and device relabeling. + + Signed-off-by: Eamon Walsh + +commit 00bc8d34c68dab6c818cd1c7e03e9992d1d0cbfc +Author: Eamon Walsh +Date: Thu Jun 18 21:41:17 2009 -0400 + + Xi: check for Use permission on the device in SetClientPointer(). + + Presumably, some intelligent, XI2-aware management app will be calling + XISetClientPointer on behalf of other clients; this check makes sure + the target client has permission on the device. + + Requires changing the prototype to return status code instead of Bool. + + Signed-off-by: Eamon Walsh + Acked-by: Peter Hutterer + +commit 119b96667778391436998c76a68bf64e746c9e08 +Author: Eamon Walsh +Date: Thu Jun 18 21:02:28 2009 -0400 + + Xi: fix up access modes for calls to dixLookupDevice(). + + New access modes are being passed to the device access hook for XI2: + DixCreateAccess for creating a new master device; + DixAdd/RemoveAccess for attaching/removing slave devices to a master; and + DixListProp/GetProp/SetPropAccess for device properties. + + Refer to the XACE-Spec document in xorg-docs, section "Device Access." + + Signed-off-by: Eamon Walsh + +commit 07c36e4fdcd93df3d33bdab6cca4780ebc9c1f54 +Author: Dave Airlie +Date: Wed Jun 10 13:33:47 2009 +1000 + + dix/resource: fix use after free in resource code with DRI + + LookupClientResourceComplex is used by DRI1 code to find and free a DRI + drawable in a callback, however when the DRI code returns this->value + is now pointing at freed memory. It seemed easiest to store the value + to a temporary and return it afterwards. + + Another option might be a new FreeClientResourceComplex or one that + also returns the id, so we can free it using an alternative means. + + found using valgrind. + + amended along ajax's suggestions + +commit 184deb9bc325eb7aa7eb7b7d4f98aa917f0269cb +Author: Dave Airlie +Date: Wed Jun 24 10:40:05 2009 +1000 + + GLX: make function static. + + This function isn't called from anywhere else and I don't think it shuold be. + +commit 9d85b56078ec05da1369ca22930d8eb214c389db +Author: Dave Airlie +Date: Wed Jun 24 10:38:49 2009 +1000 + + GLX: note the implicit flushes with ReadPixels in indirect contexts. + + This just notes the flush has occured when readpixels returns, and + fixes the glean test. + +commit e341512bfa40dd98853a20596dc65dcac4dcaa37 +Author: Thomas Jaeger +Date: Sat Jun 20 21:37:59 2009 -0400 + + dix: update a comment + + Signed-off-by: Peter Hutterer + +commit 5cbd4d3d6e54d202ecdbb527b57aaefeb8435600 +Author: Thomas Jaeger +Date: Sat Jun 20 20:17:04 2009 -0400 + + dix: do away with an instance of temporary in-place modification + + Signed-off-by: Peter Hutterer + +commit 94cdc1ef0a72802573a11ba292e2f5cc8474aa8e +Author: Thomas Jaeger +Date: Mon Jun 22 13:00:37 2009 -0400 + + dix: deal with first_valuator > 0 correctly if POINTER_SCREEN is set + + Signed-off-by: Peter Hutterer + +commit 21cbb4c9de44e2629dc0fa6f647ce2d139f2cef0 +Author: Oliver McFadden +Date: Wed Jun 24 00:24:44 2009 +0300 + + xorg-server.h.in: Export the X Access Control Extension (XACE), too. + + Acked-by: Daniel Stone + +commit 800d4ceb946c93724661ebfd28afb658ca63b6b1 +Author: Tiago Vignatti +Date: Wed Jun 24 00:13:32 2009 +0300 + + configure: Provide the --enable/disable-xaa option. + + Acked-by: Daniel Stone + +commit ce3d539ac6ecab3e343cef7c4dc112bc82ea3b02 +Author: Oliver McFadden +Date: Wed Jun 24 00:08:51 2009 +0300 + + xf86Config: Avoid attempting to load non-compiled modules. + + Acked-by: Daniel Stone + +commit 33e25143361d63ea776dfafeeb6d4ffb2f6fd8d4 +Author: Yaakov Selkowitz +Date: Fri May 15 22:42:30 2009 +0100 + + Cygwin/X: add hebrew to autodetected keyboard layouts + + Signed-off-by: Jon TURNEY + +commit b079945c39e2a72220b46953352e24c3a74ef39a +Author: Jon TURNEY +Date: Fri Feb 20 15:21:35 2009 +0000 + + Cygwin/X: more warnings fixes + + More warning fixes, mainly removing casts on function return values + + Signed-off-by: Jon TURNEY + +commit e79db6a97d02c8a256a4a7e145ea0b48b5a084ab +Author: Jon TURNEY +Date: Wed Feb 11 23:00:58 2009 +0000 + + Cygwin/X: Remove an obsolete mention of xf86Config + + Signed-off-by: Jon TURNEY + +commit 3020b1d43e34fca08cd51f7c7c8ed51497d49ef3 +Author: Michel Dänzer +Date: Tue Jun 23 16:45:40 2009 +0200 + + glx: Clean up more thoroughly if the drawable of a current context goes away. + + Fixes crash when restarting compiz, due to cl->currentContexts[x] being stale. + +commit df597709d71f47b8516e27c6fb1bfffd59de5e48 +Author: Michel Dänzer +Date: Tue Jun 23 16:45:39 2009 +0200 + + dri2: Don't crash if pPriv is NULL. + +commit 048697ccfa31cf7f7a29afa90a2f702d43efb7d4 +Author: Ben Skeggs +Date: Wed May 27 15:24:34 2009 +1000 + + quirk: use first detailed timing as preferred for PEA prod 9003 (rh#492359) + + Signed-off-by: Peter Hutterer + +commit 07154db4a8b96467785fd6be93d20379acacffa8 +Author: Ben Gamari +Date: Mon Jun 22 08:59:45 2009 -0400 + + Xi: Remove redundant and incorrect butmap range check + + Maps are CARD8s, therefore checking for values above 255 is completely + unnecessary. Moreover, 0 is a valid value for maps, so the check wasn't + even correct to begin with. This fixes bug #22392, which was uncovered + by commit 280b7f92d729ec910ffa3d18dce7bbc215be7a3c. + + Signed-off-by: Ben Gamari + Signed-off-by: Peter Hutterer + +commit 96706c24bd57fbd9b11e5bd5e38d05d81b90aebe +Author: Peter Hutterer +Date: Thu Jun 18 15:40:08 2009 +1000 + + dix: fix wrong indices in set_valuator. + + Reported-by: Thomas Jaeger + Signed-off-by: Peter Hutterer + +commit ae20e748cd3a656173e1f50109bfd4af0712bb87 +Author: Jeremy Huddleston +Date: Fri Jun 19 21:16:10 2009 -0700 + + XQuartz: More localization updates + (cherry picked from commit 7fbe974246f54535c545861a57c043d80e127ee7) + +commit 30df49f54945e75f033a0eb6445c26d37eb33c5e +Author: Eamon Walsh +Date: Thu Jun 18 23:15:47 2009 -0400 + + Xi: fix 2 memory leaks. + + In ProcXIQueryDevice() and XISendDeviceHierarchyEvent(). + + Signed-off-by: Eamon Walsh + +commit eb35402d0a5290e8a73d7d1e92f173294c364cc2 +Author: Adam Jackson +Date: Fri Jun 19 12:42:07 2009 -0400 + + pci: Dump vendor/devices ids in the printed device list + +commit c733660428c0a7c1d11f7bd21e23e1bb934d352e +Author: Søren Sandmann Pedersen +Date: Fri Jun 19 08:37:18 2009 -0400 + + Use pixman_version_string() instead of PIXMAN_VERSION_STRING + + Pointed out by Julien Cristau. + +commit aef6b904ebf0d7de6259058606c7c04ea177bda3 +Author: Dave Airlie +Date: Thu Mar 13 16:16:46 2008 +1000 + + fbdev: make entity fail if PCI claimed already. + + bad kitty fbdev, been shipping this in Fedora for a while now + +commit 3efb23a6c40dc3583d083d25ada3853ecc56000d +Author: Eamon Walsh +Date: Thu Jun 18 19:35:22 2009 -0400 + + xselinux: Add new device permissions for XI2. + + Refects the ability of clients to add/remove devices and device properties. + + Signed-off-by: Eamon Walsh + +commit 1e060c3d8b13d352a58fc65980cb9a3c6cb5718f +Author: Eamon Walsh +Date: Thu Jun 18 18:48:24 2009 -0400 + + xselinux: Move the security class mapping to the header file. + + Take the mapping of DixAccess bits to Flask permissions, move it + into the header file, break up the extremely long lines, and + annotate the permission names with the bit being referenced. + + Signed-off-by: Eamon Walsh + +commit 75c51c67b340548286efd41a53882e2acaf74ab5 +Author: Alan Coopersmith +Date: Thu Jun 18 09:49:12 2009 -0700 + + Clarify use of and need for mffs vs. ffs + + Signed-off-by: Alan Coopersmith + +commit 128cd03eecacc6d5c5903d59a11966dcf3697bf1 +Author: Søren Sandmann Pedersen +Date: Sat Jun 13 10:55:04 2009 -0400 + + Fix miComputeCompositeRegion() to follow new clip rules. + + Ignore the hierarchy clip, and always apply any client clip after + transformation and repeating. + +commit e9aa61e9f0d663d5b34a397b943b4d1df44e873d +Author: Søren Sandmann Pedersen +Date: Sat Jun 13 10:28:21 2009 -0400 + + Fix clipping when windows are used as sources + + The new clipping rules: + + - client clips happen after transformation + - pixels unavailable due to the hierarchy are undefined + + The first one is implemented in pixman; the second one is realized by + making a copy of window sources (to prevent out-of-bounds access). + +commit d9b5e77a0e48a16c53653b56bc61a0b8dc4122a1 +Author: Søren Sandmann Pedersen +Date: Tue Jun 9 14:36:21 2009 -0400 + + Print the current version of pixman. + +commit 1e9907499c27321a2aa5dc8a75a375b7a82c999a +Author: Peter Hutterer +Date: Thu Jun 18 15:14:00 2009 +1000 + + record: use dixLookupResourceByClass instead of LookupIDByClass. + + Signed-off-by: Peter Hutterer + +commit 66089e9129a821cfb1983d3d35f41b975a52de5e +Author: Peter Hutterer +Date: Wed Mar 25 12:55:42 2009 +1000 + + xfree86: fix SWCursor check in xf86CursorSetCursor. + + Wrong check for inputInfo.pointer resulted in a SW cursor being rendered when + the pointer left the screen (in a Xinerama setup). + We must call the sprite rendering function if + - SW cursors are enabled, or + - The current device is not the VCP and not attached to the VCP. + + Reported-by: Gordon Yuan + Signed-off-by: Peter Hutterer + +commit afa680e495622f521cae80563511c0d284f57551 +Author: David Miller +Date: Thu Jun 11 05:15:05 2009 -0700 + + mi: ignore DGA events in ChangeDeviceID + + DGA events don't have a deviceid, so they don't need changing. + + Signed-off-by: Peter Hutterer + +commit cd8abd17abcc8ae9da6704d03b97a9f134f36aab +Author: Jon TURNEY +Date: Wed Jun 17 14:17:07 2009 +0100 + + Cygwin/X: Fix permuted args to InitPointerDeviceStruct() + + Fix a minor error in commit a30fef9956b296f59ea18a9ee38d0abafeb15a4e, + new btn_labels argument to InitPointerDeviceStruct() wasn't added in + the right place + +commit 14581afb474552716c02ca15220ca7050123c375 +Author: Benjamin Close +Date: Thu Feb 26 17:32:10 2009 +1030 + + xfree86: correctly define barriers for FreeBSD amd64 + + Previously when compiling on freebsd amd64 we'd end up at xi86 + block (line 1315) which would define mem_barrier and write_mem_barrier + to be NOP's. Instead they should be valid, as per the linux amd64 setup. + + This stops the hangs experienced by many when using the nv driver + which would hang due to out of order dma requests as noticed in + http://bugs.freedesktop.org/show_bug.cgi?id=3168 + + Signed-off-by: Benjamin Close + +commit e92dcb6ce07aa3cfb53e8bad5701481c106c4094 +Author: Peter Hutterer +Date: Thu Jun 18 14:40:20 2009 +1000 + + input: unify button numbers on master devices. + + Master devices provide the union of all attached slave devices' buttons, + i.e. the number of buttons on the master device is always the number of + buttons of the slave device with the highest number of buttons. When slaves + are attached or detached, the master device adjusts the button number to + reflect the new buttons. + + On a slave switch, this slave's button labels are copied into the master (up + to slave->num_buttons). The remaining button labels (if any) stay as they + are. Thus, if any of the higher buttons is still pressed, it reflects the + label of the last pressed device that provided this button. + + If two devices press the same button and it is differently labelled the last + pressed one will be reflected in the master device. + + Signed-off-by: Peter Hutterer + +commit 280b7f92d729ec910ffa3d18dce7bbc215be7a3c +Author: Peter Hutterer +Date: Wed Jun 17 22:38:24 2009 +1000 + + dix: reduce MDs and xtest pointers to 7 buttons by default. + + MD's will soon be the union of all devices anyway. XTest pointers are only + for the core protocol XTest stuff, so 7 buttons (lmr + 4 wheel buttons) + should do. + + Signed-off-by: Peter Hutterer + +commit 6c7d992735eebbd7a20247926e7725896348b865 +Author: Peter Hutterer +Date: Wed Jun 17 09:11:52 2009 +1000 + + Require inputproto 1.9.99.12 + + Signed-off-by: Peter Hutterer + +commit 09cef7573938e5c08007e578e1b638bc5e1796a8 +Author: Peter Hutterer +Date: Wed Jun 17 09:04:08 2009 +1000 + + Xi: valuator/button labels are called labels now, not 'names' + + Signed-off-by: Peter Hutterer + +commit bc2ff5365030ad8bc11efde430b1064080dd7098 +Author: Peter Hutterer +Date: Tue Jun 16 19:36:57 2009 +1000 + + Xi: copy the valuator mode from SD to MD. + + Signed-off-by: Peter Hutterer + +commit 87d1f44bad608507e3995e17eb84fa0a0119796c +Author: Peter Hutterer +Date: Tue Jun 16 14:18:45 2009 +1000 + + Xi: copy the button and axes labels into the XIQueryDevice reply. + + Signed-off-by: Peter Hutterer + +commit a30fef9956b296f59ea18a9ee38d0abafeb15a4e +Author: Peter Hutterer +Date: Tue Jun 16 16:38:11 2009 +1000 + + input: Add labels to buttons and valuators - ABI_XINPUT_VERSION 7 + + Signed-off-by: Peter Hutterer + +commit 17f9723f488d0470e3879c6b0dfdba61544cdd7b +Author: Peter Hutterer +Date: Wed Jun 17 08:37:44 2009 +1000 + + input: bump to ints for deviceids - XI2 requires 16-bit deviceids. + + Note: ABI break, but ABI_XINPUT_VERSION has NOT been bumped. Recompile input + drivers. + + Revert "Xi: return BadImplementation for deviceids 256 and above" + This reverts commit 2b459f44f3edaea137df9a28bc7adfeb1b9f1df7. + + Signed-off-by: Peter Hutterer + +commit cbeb73e2055f6c013c8fe6325851f2631170137d +Author: Peter Hutterer +Date: Tue Jun 16 10:51:52 2009 +1000 + + Xi: return current valuator values in XIQueryDevice. + + Signed-off-by: Peter Hutterer + +commit 80837dbefd9d5e96ab5c1f4b4c2d5c66ce17ce67 +Author: Peter Hutterer +Date: Tue Jun 16 10:51:17 2009 +1000 + + input: change axisVal from uint to double. + + With subpixel support, uint just doesn't cut it. + + Signed-off-by: Peter Hutterer + +commit 25b6fc4a42f7698e6ae0b16becb316bfd7835d05 +Author: Peter Hutterer +Date: Tue Jun 16 10:48:48 2009 +1000 + + Xi: last_valuator is used like an index, so range it accordingly. + + The previous code would always skip the last valuator due to a wrong + upper boundary in the loop. last_valuator is the index of the last set + valuator - which also means it must be initialized to -1, not 0. + + Signed-off-by: Peter Hutterer + +commit 3fc762ea9474cf92f6b7ef60dd0bb68f59385272 +Author: Peter Hutterer +Date: Tue Jun 16 10:48:03 2009 +1000 + + Xi: Motion events update the device state too. + + Without this line the device's axis values don't get updated properly for + pure motion events. + + Signed-off-by: Peter Hutterer + +commit ad6347afead500fefb34131841f4beb5b03ad4ec +Author: Peter Hutterer +Date: Tue Jun 16 16:35:40 2009 +1000 + + Xi: don't use a constant number for class sizes - use sizeof instead + + The protocol is still changing, and having to debug crap like this is + annoying. + + Signed-off-by: Peter Hutterer + +commit d5a15e65b7d2fdda14fb719e3ebf076cce896386 +Author: Peter Hutterer +Date: Sun Jun 14 09:15:42 2009 +1000 + + Xi: include button state in XIButtonInfo. + + Signed-off-by: Peter Hutterer + +commit 01241b4247a7ad0974d48412cc0d88e1a09a3c49 +Author: Peter Hutterer +Date: Sun Jun 7 19:43:11 2009 +1000 + + Xi: Add support for sourceid in the device classes. + +commit d230742ea820a21a3f1ed0c58b5e6d8680b2f2aa +Author: Peter Hutterer +Date: Tue Jun 16 12:13:16 2009 +1000 + + Xi: namespace XI2 files. + + Some files (notably those merged with MPX before XI2 came along) didn't use + a 'xi' prefix. This patch changes all of them to meaningful names. + + Signed-off-by: Peter Hutterer + +commit 4ddb002b6847e8a88e6f13ae0453a35ee79946ae +Author: Peter Hutterer +Date: Mon Jun 15 15:14:47 2009 +1000 + + xfree86: fix two compiler warnings in xf86vmode.c + + xf86vmode.c:1578: warning: pointer targets in passing argument 1 of + ‘SwapShorts’ differ in signedness + ../../../../include/misc.h:231: note: expected ‘short int *’ but argument is + of type ‘CARD16 *’ + + xf86vmode.c:1543: warning: unused variable ‘i’ + + Signed-off-by: Peter Hutterer + +commit 9d039ef5047af2dedff0c583dda11032f71bed01 +Author: Peter Hutterer +Date: Mon Jun 15 15:11:55 2009 +1000 + + xfree86: silence a few int10 compiler warnings. + + 'format ‘%#x’ expects type ‘unsigned int’, but argument 2 has type ‘CARD32’' + and similar. + + Signed-off-by: Peter Hutterer + +commit fb882b3da189847bc98b646faddac9e8b42aa435 +Author: Adam Jackson +Date: Tue Jun 16 14:58:10 2009 -0400 + + EDID: Carp about 1.4 monitors with no preferred refresh rate + + You would think, having finally tightened down the spec, that + monitor vendors would bother to implement what the spec actually + mandates. You would be wrong. + +commit b2d2af76b43d62c9be07fd736652f904632fd663 +Author: Adam Jackson +Date: Mon Jun 15 10:10:26 2009 -0400 + + randr: fix typo in swapped dispatch + +commit 1d20b9021e8aecb942619f64017cc2fc31f02d29 +Author: Peter Hutterer +Date: Sun Jun 14 18:21:43 2009 +1000 + + Xi: fix wrong bit->byte conversion in ProcXIQueryPointer + + Signed-off-by: Peter Hutterer + +commit 9afc3241c157e5c19454f913c7105b3fdfa97ec4 +Author: Peter Hutterer +Date: Sun Jun 14 18:24:22 2009 +1000 + + Xi: remove un-used IsOn macro. + + Signed-off-by: Peter Hutterer + +commit b67ff1d6e0d01170c6261593ca36d0f51d917393 +Author: Peter Hutterer +Date: Fri Jun 12 16:45:36 2009 +1000 + + Xi: XISelectEvents/XIGetSelectedEvents use 'win' instead of 'window' now. + + Signed-off-by: Peter Hutterer + +commit b2bf67b61c564a4b92a429ca9ad455403161f33a +Author: Benjamin Defnet +Date: Wed Jun 10 21:35:48 2009 -0700 + + randr: fix operation order so that rotation+transform works + + The matrix multiply to combine rotation and projective transforms was being + done in the wrong order. + + Signed-off-by: Keith Packard + +commit 3627215e97734cc367fff273a3358c2a95a53481 +Author: Peter Hutterer +Date: Fri Jun 5 16:38:46 2009 +1000 + + Add check for GNU ld in the test suite. + + The GNU linker supports a -wrap option to wrap function calls at link-time. + This allows for easy overriding of functions in the X server with stubs in + the test suite. This functionality is only supported on the GNU linker and + will be used extensively in the tests. Disable the tests if GNU ld is not + available. + +commit 5e0ca6fabd12401de6bd6039484be2079e28851e +Author: Peter Hutterer +Date: Thu Jun 11 16:26:23 2009 +1000 + + input: remove un-used "setter" argument from SetClientPointer. + + It's obsolete, not likely to come back, let's drop it. + + Signed-off-by: Peter Hutterer + +commit e6a18762ef113296c6a09833be70cb4b45aa3940 +Author: Peter Hutterer +Date: Thu Jun 11 15:40:56 2009 +1000 + + Xi: fix XISetClientPointer return values. + + If SetClientPointer fails, the only reason may be that the device is not a + pointer or that the device is an SD. Return BadDevice instead of BadAccess. + (BadAccess is a leftover from the early times of the ClientPointer + implementation when only one client was allowed to set it). + + If the window parameter doesn't name a valid window or client, return + BadWindow. + + Finally, allow both master keyboards and master pointers as deviceid. + + Signed-off-by: Peter Hutterer + +commit ae7dab2a136d15b976b956f68feec53886951dd6 +Author: Peter Hutterer +Date: Thu Jun 11 15:40:38 2009 +1000 + + Xi: Fix XISetClientPointer swapping. + + Signed-off-by: Peter Hutterer + +commit 98e8ec8deb09db1c56e06afed267e6f517042fa3 +Author: Peter Hutterer +Date: Fri Jun 12 10:54:55 2009 +1000 + + Xi: sanitize ProcXIGetClientPointer. + + This was quite old code and can be streamlined a bit. The new code is + essentially the same as in ProcXISetClientPointer. + + Signed-off-by: Peter Hutterer + +commit 96ea82fdac7c28feb2748cd4ff7faa7c00dbbdcd +Author: Peter Hutterer +Date: Fri Jun 12 10:54:25 2009 +1000 + + Xi: fix reply swapping in XIGetClientPointer. + + Signed-off-by: Peter Hutterer + +commit 76f25086a9ab6f0c27d9e2cd745f8463c6b3df5f +Author: Adam Jackson +Date: Thu Jun 11 16:34:06 2009 -0400 + + vfb: Re-enable 30bpp support + +commit 0de58c88aba7ddd69b04f24ab5b2967c359aa69e +Author: Dave Airlie +Date: Thu Jun 11 14:21:53 2009 +1000 + + xfree86: move didLock assignment down to where the function pointer is valid. + + crtc->funcs->lock is NULL, so it's no use calling it here. Move it down so + it's actually defined before we use it. + + Introduced with 6f59a8160042ea145514fdcb410f17f33fd437c2. + + Tested-by: Peter Hutterer + +commit a85d210b34506cb39e7f08d81d4586fe4448b47f +Author: Peter Hutterer +Date: Thu Jun 11 12:32:39 2009 +1000 + + Xi: store mask_len before swapping in ProcXIGetSelectedEvents. + + Swapping the mask_len and then advancing the pointer by the swapped length + is just a bad idea. + + Signed-off-by: Peter Hutterer + +commit 5043f42f36b49e7702d0c496dd4fe157788d44ee +Author: Peter Hutterer +Date: Thu Jun 11 12:31:40 2009 +1000 + + Xi: correct return buffer size for XIGetSelectedEvents. + + The maximum number of bytes is calculated by the mask len, and the mask len + is always in 4-byte units. XI2MASKSIZE however is in bytes. + + Signed-off-by: Peter Hutterer + +commit fcf0c0b8f332cbde03c1a479376b7913197962eb +Author: Peter Hutterer +Date: Thu Jun 11 13:23:40 2009 +1000 + + Xi: XISetEventMask needs to clear the mask if len is 0 + + zero-length masks are supposed to clear the device's mask. + ProcXISelectEvents passes these masks through directly, so we need to clear + the bits here if such a mask is supplied. + + Signed-off-by: Peter Hutterer + +commit 22b4ac44e4fdbe8f36f2b570cdbdaf237062e33e +Author: Peter Hutterer +Date: Thu Jun 11 11:08:07 2009 +1000 + + Xi: XIGetSelectedEvents mustn't returned masks from non-existing devices. + + Or devices the client doesn't have XACE permissions for. + + Signed-off-by: Peter Hutterer + +commit 7868956b0fb50a9d77506f01413612e429f540a3 +Author: Peter Hutterer +Date: Wed Jun 10 17:09:39 2009 +1000 + + Xi: ProcXIGetSelectedEvents must use WriteToClient for swapped data. + + The data is already swapped before, so we just post it to the client as-is, + without attempting to swap it again. + + Signed-off-by: Peter Hutterer + +commit aa2babf11c30be4d289b58212dc330b84efb4053 +Author: Peter Hutterer +Date: Wed Jun 10 14:41:11 2009 +1000 + + input: remove dependency on XI2 protocol for XI_LASTEVENT. + + inputstr includes XI2proto.h for the sole purpose of XI_LASTEVENT. + However, using XI_LASTEVENT in the server is prone to errors, if the server + is recompiled against a newer version of the protocol it would bump this + variable and associates bits, including potential ABI. + + This patch defines an XI2LASTEVENT for use in the server and removes the + XI2proto.h require. XI2LASTEVENT is the current value of XI_LASTEVENT. + + This patch is required by components that require access to inputInfo + (currently xf86-video-geode and xf86-video-cirrus) but should not have a + require for the XI2 protocol. + + Signed-off-by: Peter Hutterer + +commit 77cc816da4a2777110182ed01c22c0f5e6ac3b65 +Author: Peter Hutterer +Date: Wed Jun 10 13:54:42 2009 +1000 + + Xi: rename ProcXISelectEvent to ProcXISelectEvents. + + The request name has the plural, so let's do it here too. Purely cosmetic + change. + + Signed-off-by: Peter Hutterer + +commit 9563feeeb5309de1971c4e04419fed6fd31c9cea +Author: Peter Hutterer +Date: Tue Jun 9 16:57:09 2009 +1000 + + Xi: start checking for invalid mask bits _after_ LASTEVENT. + + Two issues that combined to false positives and false negatives. + - The checking for invalid bits must be performed when there are enough bits + that an event outside of LASTEVENT may be selected. + - The first invalid bit is LASTEVENT + 1, not LASTEVENT. + + Signed-off-by: Peter Hutterer + +commit e2fbaebb87e18198143a4b8e6be7e650563cd819 +Author: Peter Hutterer +Date: Tue Jun 9 14:38:25 2009 +1000 + + Xi: XISelectEvents returns BadValue for num_masks == 0. + + Signed-off-by: Peter Hutterer + +commit cfeb65ac45b194dead5e8b9b26192df2619d6f9c +Author: Peter Hutterer +Date: Tue Jun 9 14:37:42 2009 +1000 + + Xi: XISelectEvents needs to be at least size 3, not exactly. + + Signed-off-by: Peter Hutterer + +commit f3c26034eca98924d93a004aaa1f61ffe5d344ca +Author: Peter Hutterer +Date: Tue Jun 9 13:04:56 2009 +1000 + + Xi: XIQueryDevice should use XIFooClass instead of the old FooClass defines. + + This is merely a cosmetic change, the actual values are the same anyway. + + Signed-off-by: Peter Hutterer + +commit 482cc72aa790bd2f23c7ebe12c5549b47c737115 +Author: Peter Hutterer +Date: Tue Jun 9 13:03:38 2009 +1000 + + Xi: get the class length before swapping. + + Advancing by the already-swapped length lets our pointers point into + nirvana. + + Signed-off-by: Peter Hutterer + +commit 9974249980894f74f3ead466655da87958a43670 +Author: Peter Hutterer +Date: Tue Jun 9 13:02:52 2009 +1000 + + Xi: Swapping 32 bit keycodes requires swapl, not swaps. + + Signed-off-by: Peter Hutterer + +commit 918923e285f4e269a257bb5be4d3c8a50174aad0 +Author: Dave Airlie +Date: Wed Jun 10 15:13:45 2009 +1000 + + glx: fix open-coded linked list removal function + + OMG stab stab stab, YALL. + + removal function was made of crack, actually truncated the list from + the one after the find point. + + However fixing this makes glean makecurrent fail with a GLX error. + +commit 3ea747c0dbbec0db6761d66d4f6c680d2e9ddeaf +Author: Daniel Stone +Date: Sat Apr 25 16:53:18 2009 +1000 + + KDrive: Warning fixes + + xEvent vs. InternalEvent confusion still reigns though. + + Signed-off-by: Daniel Stone + +commit 754be1e2ec9be2486bf45000d7244d217556de07 +Author: Daniel Stone +Date: Sat Apr 25 16:51:48 2009 +1000 + + KDrive: Xephyr: DRI: Warning fixes + + Signed-off-by: Daniel Stone + +commit f534e6bea17746db952feb563ffea7320846b49d +Author: Daniel Stone +Date: Sat Apr 25 15:38:54 2009 +1000 + + OS: Fix compile warnings + + It's a marvel the sigaction() ever actually worked. + + Signed-off-by: Daniel Stone + +commit 305ab237f666936cd812c464bf43f86f6079838e +Author: Daniel Stone +Date: Sat Apr 25 22:11:19 2009 +1000 + + KDrive: Warning fixes and cleanups + + Signed-off-by: Daniel Stone + +commit 4c8812b544ce5f319a7ac59a131c626c8509ef99 +Author: Adam Jackson +Date: Wed Jun 10 12:46:19 2009 -0400 + + kdrive: undef PSEUDO8 + + Whatever that was, it no longer is. + +commit 973ef5a7086aedb88a4888e73630c96ae68078e4 +Author: Adam Jackson +Date: Wed Jun 10 12:44:45 2009 -0400 + + kdrive: Remove a lie about PCMCIA support. + +commit e3c65cf1df9bbfb126f07b4b8e1254a855fb70c7 +Author: Adam Jackson +Date: Wed Jun 10 12:05:09 2009 -0400 + + xephyr: Add -title option. + +commit 803522300344265f11ea9f869ee6747730413bae +Author: Colin Harrison +Date: Wed Jun 10 13:32:54 2009 +0100 + + Xming: Simplify logic for ownership release of native clipboard + + Make the logic simpler for the ownership release of the Windows + clipboard in winclipboardwrappers.c - We've already marked a selection + as unowned if it is owned by our clipboard window + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit c7d3965bee0edb72468a0135aaa8a081348c2b23 +Author: Jon TURNEY +Date: Wed Jun 10 13:09:44 2009 +0100 + + Cygwin/X: Add a needed inputstr.h + + Add a needed inputstr.h to fix build after commit 6d4ffcc9e066bc7c49dcff1a2ff0111801c7286d + + Signed-off-by: Jon TURNEY + +commit 6d4ffcc9e066bc7c49dcff1a2ff0111801c7286d +Author: Dave Airlie +Date: Wed Jun 10 11:15:24 2009 +1000 + + input: move inputstr.h to where its needed. + + This stops inputstr.h being needed to be included by output drivers. + + Signed-off-by: Dave Airlie + Signed-off-by: Peter Hutterer + +commit 6f59a8160042ea145514fdcb410f17f33fd437c2 +Author: Benjamin Defnet +Date: Mon Jun 8 21:45:42 2009 -0700 + + hw/xf86/modes: Set crtc mode/rotation/transform before calling set_mode_major + + This moves code out of each implementation of set_mode_major and back into + the X server. The real feature here is that the transform is now available + in the crtc for use by either xf86CrtcRotate or whatever the driver wants to + do. Without this change, the transform was lost for drivers providing the + set_mode_major interface. + + Note that users of this API will want to stop smashing the transformPresent + field, and could also stop setting mode/x/y/rotation for new enough X servers, + but there's no reason to make that change as it will break things when + running against older X servers. + + Signed-off-by: Keith Packard + Acked-by: Daniel Stone + +commit e244a5991e2cc55f5aa2f6e5255f1dabf56f0235 +Author: Federico Mena Quintero +Date: Tue Jun 9 13:28:28 2009 -0700 + + dix/randr: Add missing fields to SRR*NotifyEvent() + + Also, remove redundant field swaps and make others match the order in which + they are declared in the xRR*NotifyEvent structs. + + Signed-off-by: Federico Mena Quintero + Signed-off-by: Keith Packard + +commit 746e7b22e1bdfbf2363040367209b4c46b1c2fec +Author: Adam Jackson +Date: Tue Jun 9 16:20:11 2009 -0400 + + cvt: Allow multiple-of-60Hz refresh rates for reduced blanking. + + Doing so generates the same timings as given in the DMT spec for + 120Hz RB, so we should be set there. Other rates might be legal + too but why push our luck. + +commit 2a035600e603117a4dbe668cad50bae2a5609094 +Author: Peter Hutterer +Date: Tue Jun 9 10:14:18 2009 +1000 + + require inputproto 1.9.99.11 + +commit c5bebca46f63bcd7279f3c73ed1d50e0c1b6051f +Author: Peter Hutterer +Date: Mon Jun 8 17:54:05 2009 +1000 + + Xi: hierarchy events have a num_info now instead of num_devices. + + Signed-off-by: Peter Hutterer + +commit 810b74dbbcc1db3e68e0f95ff5223ddb2032e46d +Author: Peter Hutterer +Date: Mon Jun 8 13:41:22 2009 +1000 + + Xi: change from XICreateMaster to XIAddMaster for consistency. + + add/remove is used for slave devices and hierarchy flags. + + Signed-off-by: Peter Hutterer + +commit 0e66a443a09176b8a875195d5473b953c07aa1d7 +Author: Peter Hutterer +Date: Mon Jun 8 12:24:24 2009 +1000 + + Xi: return BadValue for XI_HierarchyChangd mask on devices. + + This mask may only be selected for XIAllDevices. + + Signed-off-by: Peter Hutterer + +commit 69a9545d1f8110841538410818df19fd960412c5 +Author: Keith Packard +Date: Thu May 28 14:43:27 2009 -0700 + + Make RANDR 'set' timestamps follow client specified time. Bug 21987. + + The lastSetTime value which indicates when the configuration within the + server was last changed was not getting set in the appropriate RandR + requests. + + Signed-off-by: Keith Packard + +commit fa18c569ed26d2fdd442af6b39723cb7e30d18df +Author: Ben Skeggs +Date: Tue Jun 9 09:14:50 2009 +1000 + + exa: driver pixmaps enabled if either CreatePixmap or CreatePixmap2 present + +commit 9390b7a133b0e17bc4d303effb7a99ff7017b086 +Author: Peter Hutterer +Date: Sun Jun 7 19:26:45 2009 +1000 + + dix: protect against missing ButtonClasses in GetPointerMapping. + + Signed-off-by: Peter Hutterer + +commit 21ee53060b4dcb5c12074cf6682ff0067021ff60 +Author: Simon Thum +Date: Fri Jun 5 19:14:48 2009 +0200 + + config: fix build after XI2 API changes to RemoveDevice. + + This patch fixes the build with --enable-config-dbus is enabled + (default disabled). + + Signed-off-by: Peter Hutterer + +commit 07c59411817c10fbcbf1bc9efa6a0818b6e0493a +Author: Jeremy Huddleston +Date: Sat Jun 6 22:17:44 2009 -0700 + + GLX: Make sure the types match for ALIAS in indirect_reqsize.c + (cherry picked from commit 31a20a573b867fb7f02da15425e3b92eccb82dec) + +commit 077556ec39ca7e0d14486005cf71fcdc70a350cf +Author: Jeremy Huddleston +Date: Sat Jun 6 21:55:51 2009 -0700 + + XQuartz: AIGLX dispatch table cleanup + + This should fix some issues when building on different versions of OSX. We only use extensions that OpenGL.framework's glext.h header tells us are supported. + (cherry picked from commit e10c53421f1e780573c8adcd8ea89d6ffa28a819) + +commit fc24a97a8d022e986082b65b5133701e7273805c +Author: Jeremy Huddleston +Date: Tue Jun 2 19:12:41 2009 -0700 + + Rootless: Use serverClient instead of NullClient + + This seems to stop the crash when moving windows. + (cherry picked from commit fcd857c8e295ac7a2b0e58d49dc71251ed9d1266) + +commit 2429daa7e6b35a4a8032e1c3736e417a5a9baa07 +Author: Jeremy Huddleston +Date: Tue Jun 2 10:19:10 2009 -0700 + + XQuartz: Default to forcing the dialog box before quitting. + + This will show the quit dialog for users of other WMs. + (cherry picked from commit 639ee0c913f892776fee40bcfb856e0d7c2c7db5) + (cherry picked from commit 09432fa5b1cc8e9a30fb8aa4f8dfcd08c4a91e30) + +commit 6d75606a9e7f6ebc4d79a5261a48f5915c71cbcd +Author: Jeremy Huddleston +Date: Mon Jun 1 22:58:40 2009 -0700 + + Rootless: Despite its name, we still want to allow rootless to mess with the root window + + This fixes the bug whereby we weren't drawing the contents of the root window since RootlessPaintWindowBackground got nuked. + (cherry picked from commit bb0e208787739fbed2a0c8d1af6b6504319fdf88) + +commit 7348bf4935c8887bca33fe340ac2548f551bee3d +Author: Peter Hutterer +Date: Sun Jun 7 09:25:51 2009 +1000 + + dix: set the generic event type for implicit XI2 grabs. + + event->type is always GenericEvent for XI2 events. Instead, XI_ButtonPress + (the generic event's evtype must be stored. + + Signed-off-by: Peter Hutterer + +commit bb1c131b785ca30cf40041db3513b7669beebd34 +Author: Simon Thum +Date: Thu May 28 16:43:42 2009 +0200 + + dix: suppress pointer acceleration on xtest devices + + Signed-off-by: Peter Hutterer + +commit 707b1241681c33401d5e0eb6edeace33f5970928 +Author: Simon Thum +Date: Sat Mar 21 18:19:19 2009 +0100 + + dix: prefer lroundf() over roundf() in axis scaling + + it's unclear whether there actually is a problem, but in a very similar + case there is (bug#21456). Also, integer addition is generally faster. + + Signed-off-by: Peter Hutterer + +commit ef0e435052b2fda598cd8670c8e892aea685a102 +Author: Jon TURNEY +Date: Fri Jun 5 19:29:02 2009 +0100 + + Cygwin/X: Remove obsolete ChangeLog files + + Remove obsolete ChangeLog files, this information is preserved in + the automatically generated top-level ChangeLog + +commit 752508577aa7f0a1d80c8b7f4d3fe2b098e5e5fa +Author: Jon TURNEY +Date: Tue Jun 2 21:19:46 2009 +0100 + + Cygwin/X: Tidy up a bodge to avoid collision between X header and native Win32 API definitions of ATOM + + Remove a bodge to avoid collision between X header and the native + Win32 API definitions of ATOM, use X11/Xwindows.h which exists to + address these issues + + Consequently, include misc.h in winclipboardthread.c, to provide the + max() macro, as previously we were using the the one from windows.h + + Signed-off-by: Jon TURNEY + +commit 7f3456a4080d1bf21010170e126128cb653bd2ac +Author: Peter Hutterer +Date: Fri Jun 5 16:44:46 2009 +1000 + + Revert "test: check from INT_MIN to INT_MAX for core type conversion" + + This patch requires extra special casing to check if the linker supports the + -wrap option. Patches to do so will follow, in the meantime, revert this + commit. It shouldn't have been pushed in the first place anyway. + + This reverts commit d979f443946011158b6a183582728a6899c33b85. + + Signed-off-by: Peter Hutterer + +commit d770d57f1701c7c8272b4ca3714f64c46c47adf1 +Author: Peter Hutterer +Date: Fri Jun 5 11:38:22 2009 +1000 + + Xi: fix XIQueryVersion major/minor swapping. + + Signed-off-by: Peter Hutterer + +commit 17d62306b602ec61e8c5b44a79c9e1edc6b63a21 +Author: Peter Hutterer +Date: Fri Jun 5 11:31:39 2009 +1000 + + Xi: add XIQueryVersion reply swapping hook. + + Signed-off-by: Peter Hutterer + +commit 2b459f44f3edaea137df9a28bc7adfeb1b9f1df7 +Author: Peter Hutterer +Date: Fri Jun 5 10:15:04 2009 +1000 + + Xi: return BadImplementation for deviceids 256 and above + + The protocol allows for 16 bit device ids, but the implementation doesn't + yet. We need to break the input ABI once more to shift the DeviceIntRec's + CARD8 to a CARD16, along with some changes in the privates. + + Once that is done, revert this patch. + + Signed-off-by: Peter Hutterer + +commit c77f42049f77417c68bbce3dd2e995d916fc5af5 +Author: Peter Hutterer +Date: Fri Jun 5 09:29:04 2009 +1000 + + Xi: fix flawed logic in XIQueryVersion return value. + + The server returned the minimum of major/minor each instead of the lower of + the combined major.minor version. As a result, a client reporting + 3.0 and a server supporting 2.7, the return value would be 2.0 (the minimum + of both). + + Reported-by: Rémi Cardona + Reported-by: Simon Thum + Signed-off-by: Peter Hutterer + +commit b6c16fc7dad5a4ffcaaab647c0e3fabed372efd5 +Author: Dave Airlie +Date: Fri Jun 5 11:57:39 2009 +1000 + + crtc/gamma: check xf86_config is valid before using it. + + If you have multiple cards, some that support randr 1.2 and some that don't + you can get a null dereference in here. + + Signed-off-by: Dave Airlie + +commit da682abc78c9a5b2fde09679061fe08a3a1c665b +Author: Eamon Walsh +Date: Thu Jun 4 17:33:25 2009 -0400 + + xace: fix a bad send access hook call. + +commit 993daf06497c85bb6a1e70592df380503d721dfb +Merge: 2bda50e e1e8c7d +Author: Peter Hutterer +Date: Thu Jun 4 15:11:36 2009 +1000 + + Merge branch 'master' into xi2 + +commit 2bda50ee14b7873f28081bf0f1bfb7b79a40de43 +Author: Peter Hutterer +Date: Thu Jun 4 14:51:32 2009 +1000 + + Xi: memset the device name padding to 0. + + Signed-off-by: Peter Hutterer + +commit 09266d6cd1bb6e09fe8bcb875c63fb4011c05b55 +Author: Peter Hutterer +Date: Thu Jun 4 13:36:29 2009 +1000 + + Xi: XIQueryVersion requires major_version of 2 or higher. + + Otherwise, a BadValue error is sent. + + Signed-off-by: Peter Hutterer + +commit 6ea76c550486bd17eb31f43757030fb44bdf1eab +Author: Peter Hutterer +Date: Thu Jun 4 12:55:59 2009 +1000 + + Xi: reply with the XI2 version that is supported by both client and server. + + XIQueryVersion must return the client's version if the client's version is + lower than or equal to the server's version, or the server's version + otherwise. + + Signed-off-by: Peter Hutterer + +commit d979f443946011158b6a183582728a6899c33b85 +Author: Peter Hutterer +Date: Thu Jun 4 11:40:14 2009 +1000 + + test: check from INT_MIN to INT_MAX for core type conversion + + ErrorF is link-wrapped to reduce the run-time of the test. + + Signed-off-by: Peter Hutterer + +commit d8471bc7a19cd248fa3e0845496c86cd9251c013 +Author: Peter Hutterer +Date: Thu Jun 4 11:32:15 2009 +1000 + + test: adjust event_to_core_conversion test for new ProximityIn/Out behaviour + + Both types now return BadMatch. + + Signed-off-by: Peter Hutterer + +commit e1e8c7ddd7fdbfd674361364295fb1cbd5f28b45 +Author: Adam Jackson +Date: Wed Jun 3 13:37:32 2009 -0400 + + s/MIN/min/, s/MAX/max/ (#2968) + +commit 6de67d3206f0a4e307070714564c19efc84da2ec +Merge: af4e658 7f027d9 +Author: Peter Hutterer +Date: Wed Jun 3 16:11:33 2009 +1000 + + Merge branch 'master' into xi2 + +commit 7f027d9dc0146e229802aeac342ea2dbab63264a +Author: Olivier Blin +Date: Thu May 28 20:57:36 2009 +0200 + + kdrive: add protocol mouse option + + kdrive probes a lot of PS/2 protocols for the mouse device, which + makes the mouse unusable for some seconds after X startup. + This new "protocol" option allows forcing the mouse protocol. + It can be used this way: + Xfbdev -mouse mouse,,protocol=ps/2 -keybd keyboard + + Signed-off-by: Olivier Blin + Signed-off-by: Peter Hutterer + +commit af4e658401ecbd2f2cc515b4c5abe5aee361cd64 +Author: Peter Hutterer +Date: Wed Jun 3 12:30:49 2009 +1000 + + mi: only process master if the SD is still attached. + + SDs may be detached during event processing (e.g. if a passive grab + activates). In this case, the event must not be processed through the master + device. + + Reported-by: Thomas Jaeger + Signed-off-by: Peter Hutterer + +commit eb3fe72cbcd10c3953d713e07c321e695194a328 +Author: Peter Hutterer +Date: Wed Jun 3 12:19:14 2009 +1000 + + dix: set the device state for XI valuator events. + + Valuator events need to include the device's state, while other device + events need to include the state of the core devices. + + Reported-by: Thomas Jaeger + Signed-off-by: Peter Hutterer + +commit 08fd2ae652883393ecff6f3d5177e63408d58472 +Author: Peter Hutterer +Date: Wed Jun 3 12:18:17 2009 +1000 + + dix: use GetMaster in PickPointer and PickKeyboard. + + GetMaster is more reliable than GetPairedDevice, it always returns the + keyboard/pointer if desired, even if the wrong device was passed in. + + Signed-off-by: Peter Hutterer + +commit a1d1dd06f8f91c1d1802f7581a4ba7568b85e69c +Author: Peter Hutterer +Date: Wed Jun 3 12:15:32 2009 +1000 + + Xi: fix wrong grabtype from ProcXGrabDevice. + + 86077f0058ce88ee9b3df5d1ab854eeca43 switched from a boolean to a grabtype + enum. ProcXGrabDevice didn't switch with it. PickPointer during an XI grab + on a slave device would thus return a wrong (or NULL) device and crash the + server. + + Signed-off-by: Peter Hutterer + +commit a371a47205fc58a587e1aecaa4a031184481c228 +Author: Peter Hutterer +Date: Wed Jun 3 11:47:16 2009 +1000 + + dix: EventToCore must return BadMatch for proximity events. + + Signed-off-by: Peter Hutterer + +commit 2bd5ea80652358565346071bed829070ddf5de85 +Author: Peter Hutterer +Date: Wed Jun 3 10:45:28 2009 +1000 + + dix: rework event conversion before delivery. + + Let EventToXI, EventToCore and EventToXI2 return BadMatch if there's no + matching event for this protocol spec. + Adjust the delivery paths to cope with BadMatch errors (and clean them up on + the way). + + As a side-effect, this fixes server crashes on proximity events for a + grabbed device. + + Signed-off-by: Peter Hutterer + +commit 20d9f11abfd6ce3630d9640ff7383c7e27196956 +Author: Adam Jackson +Date: Tue Jun 2 13:49:09 2009 -0400 + + Typo fix + +commit ab5d1ae82ebd467d0a5b39821b286097865ba0c7 +Author: Adam Jackson +Date: Tue Jun 2 13:39:55 2009 -0400 + + Remove some libXfont leftovers + +commit 75795637c7160f1579dbe81c2d7600e85b1d141f +Author: Matthias Hopf +Date: Tue Jun 2 18:41:36 2009 +0200 + + Revert "randr: Setting gamma: inverse logic looks more sane" + + This reverts commit 79138eec1b49cbaca6a16f2bdd8579b5828aeb28. + +commit fc8cfc3a055d8af4ac809799c71f3db0d5246433 +Author: Peter Hutterer +Date: Tue Jun 2 16:53:41 2009 +1000 + + dix: ensure EventIsDeliverable has inputMasks set at all times. + + For proximity events, the XI2 type is 0 and inputMasks never got set in the + preceding condition. As a result, proximity events got never delivered. + + Signed-off-by: Peter Hutterer + +commit ef9f851057a0ae214be02cb919f1e6634cead20b +Author: Peter Hutterer +Date: Tue Jun 2 16:38:32 2009 +1000 + + dix: protect against floating slaves in DeviceFocusEvent. + + Signed-off-by: Peter Hutterer + +commit b40289c8766c5e7543b7288009b4d965f57c04ce +Author: Peter Hutterer +Date: Tue Jun 2 16:22:58 2009 +1000 + + xkb: allow pointer events to pass through for floating SDs without a key class. + + Signed-off-by: Peter Hutterer + +commit 38df8351b36195070232b42a6d3369a9a08df8df +Author: Peter Hutterer +Date: Mon Jun 1 17:09:03 2009 +1000 + + test: add grab matching tests. + +commit a66686a83edd6093ed540414dda2b8700e4696fe +Author: Peter Hutterer +Date: Mon Jun 1 15:37:53 2009 +1000 + + input: add support for XIAllDevices and XIAllMasterDevices passive grabs. + + These grabs are suported through two fake devices inputInfo.all_devices and + inputInfo.all_master_devices. These devices are not part of the device list + and are only initialised for their device id, nothing else. + + Signed-off-by: Peter Hutterer + +commit f00cf7675164bb984ef310412f9e09582813adb1 +Author: Peter Hutterer +Date: Tue Jun 2 10:23:21 2009 +1000 + + dix: get the current MD keyboard for the grab modifier state. + + A passive XI2 grab always uses the paired master device as a modifier + device. After issuing a passive grab, the slave may be reattached to a + different master and hence the modifier device may change. + +commit 1cd7ed75d903b5b0ea4e1070d2c26f4f931bdec7 +Author: Peter Hutterer +Date: Mon Jun 1 17:14:06 2009 +1000 + + dix: float slave devices during passive grabs. + + Exception - implicit passive grabs. + + Signed-off-by: Peter Hutterer + +commit f56ad22e8fabdca6d9fd1e408a539696e30599fc +Author: Peter Hutterer +Date: Tue Jun 2 15:07:53 2009 +1000 + + dix: rework temporary slave detachment. + + Rework addresses two issues: + - storing the master device's pointer is a bad idea, we need to store the ID + of the device in case it disappears during the grab. + - restoring the old master did not actually reattach the device. Fixed now. + +commit b3e4810a283d369e1a2c6c3288177fe79fd1e9d6 +Author: Peter Hutterer +Date: Mon Jun 1 16:22:51 2009 +1000 + + dix: move grab device assignment into GRABTYPE_CORE block. + + grab->type is the device type and XI2 types overlap with core events (being + less than GenericEvent). Thus, for passive grabs the grab device would be + overwritten with whatever device was activating it. + + Signed-off-by: Peter Hutterer + +commit 18ae48670859d05788776e164c65985e3b35a723 +Author: Peter Hutterer +Date: Mon Jun 1 16:06:52 2009 +1000 + + dix: don't try to activate a passive grab on top of an active one. + + If we already have an active grab on the device, don't attempt to activate + another one. + + Signed-off-by: Peter Hutterer + +commit ff1d6244eba8367a8d50601ca393e951c60c94b6 +Author: Peter Hutterer +Date: Mon Jun 1 12:14:52 2009 +1000 + + input: add support for XIAnyModifier in passive XI2 grabs. + +commit eb97d125fb6adeff0b23c2460108e0587ba253e2 +Author: Adam Jackson +Date: Mon Jun 1 15:50:50 2009 -0400 + + doc: Remove mention of Kerberos replay cache from Xserver.1 + + The krb5 auth code is gone, so this is no longer relevant. + +commit 80ea32ad4f1440d068e18db65830f45498703b09 +Author: Peter Hutterer +Date: Mon Jun 1 12:06:55 2009 +1000 + + dix: switch grab checking from unsigned shorts to unsigned ints + + XIAnyModifier is outside of the unsigned short range. + + Signed-off-by: Peter Hutterer + +commit 080b0331b3d1de1bab5ecdc9df175b287616e358 +Author: Peter Hutterer +Date: Mon Jun 1 13:33:29 2009 +1000 + + input: Add grabtype to GrabParameters. + + Signed-off-by: Peter Hutterer + +commit bee02a154145c0e4ed97980f7857dc27a96b1be8 +Author: Peter Hutterer +Date: Mon Jun 1 10:49:22 2009 +1000 + + dix: ensure implicit passive grabs have the xi2mask zeroed. + + This fixes XI2 enter/leave events being sent to non-xi2 clients. + + Signed-off-by: Peter Hutterer + +commit cc71dfb4f5cf6113053951070dec9556e941aa24 +Author: Peter Hutterer +Date: Mon Jun 1 13:19:30 2009 +1000 + + test: Add GetSelectedEvents request to struct size tests. + + Signed-off-by: Peter Hutterer + +commit de5c5b1db083a9c1d23e1461b54ca646c5a5c445 +Author: Peter Hutterer +Date: Mon Jun 1 13:21:36 2009 +1000 + + test: fix compile error introduced with the removal of isMaster. + + Signed-off-by: Peter Hutterer + +commit 45530d16097459a756696e255ab4e72d6e51fbc4 +Author: Adam Jackson +Date: Fri May 29 18:07:48 2009 -0400 + + vfb: Fix depth setup. + + Initialize the depth corresponding to the root window before the + pixmap-only depths. Otherwise you end up with the root window depth in + the depth list twice, which is mildly confusing for clients and + catastrophically confusing for PanoramiXConsolidate(). + +commit 5c8540d8cf9ebaafa7572cb2879d057cd3d15190 +Author: Jeremy Huddleston +Date: Fri May 29 14:49:05 2009 -0700 + + GLX: Purge glxint.h usage + + Change (__GLXvisualConfig *) to (void *) in ABI compat stubs. + (cherry picked from commit b4adec886c65e49cfbd30d1e3a41b613f75ed3f5) + +commit da2aaee24a77e1a9e5c623af4a78ca6acc94b787 +Author: Jeremy Huddleston +Date: Fri May 29 14:27:46 2009 -0700 + + GLX: Purge some glxint.h usage in glxcmds.c + + This is related to d0b249f1c5df81f3941d638f9625fe1e16c31807 in the 1.4 branch and was ok'd by krh + (cherry picked from commit e587436cf2e1237813c924699ff0662aea3034e1) + +commit f2151f1b0dcbfec23488a29a7ea887fbfd53b760 +Author: Jeremy Huddleston +Date: Fri May 29 14:09:41 2009 -0700 + + XQuartz: Fix alpha to be 1 (screenshots bug, etc) + + This is related to: bc964ff1e3f3d8ab590eab40d767c6241e9480d4 + XQuartz: Stab at fixing the alpha 0/1 bug (screenshots, etc) by pulling in some old code that got gutted from rootless. + + which was on the 1.4 branch and implemented in fbPaintWindow. Now that fbPaintWindow is gone, this is now in miPaintWindow(). + (cherry picked from commit 032173f693f75385a40e144ffe2b7b5378bb6f2e) + +commit 7c158ffa766d32f5e8ec459693da5ef6ba0892e9 +Author: Jeremy Huddleston +Date: Fri May 29 14:02:18 2009 -0700 + + Rootless: Use miPaintWindow since PaintWindowBackground is dead + (cherry picked from commit 4975c087257b52189a5c21d93d121f183e60e4c8) + +commit ec9e9c8bcabda3d8b4ea07037fb247f7408dde8b +Author: Jeremy Huddleston +Date: Thu May 28 15:30:20 2009 -0700 + + XQuartz: English localization update for HIG compliance + (cherry picked from commit 86b319d7cbef39d75b54c9d321e5dbf097b552be) + +commit 0f47203a9a690498c18b1c04e2ef5deed23a0e3e +Author: Peter Hutterer +Date: Fri May 29 14:18:37 2009 +1000 + + dix: Set the implicit passive grab flag for keyboards too. + + Extension devices have ActivateKeyboardGrab as their grab activation + function, hence we need to ensure the implicit passive grab flag is set + accordingly in the grab for further event delivery. + + Signed-off-by: Peter Hutterer + +commit 75c73ddbfc9236bb80a3709d467674c4a1018fa3 +Author: Peter Hutterer +Date: Fri May 29 14:01:21 2009 +1000 + + dix: fix broken XI event delivery during grabs. + + If an implicit passive grab is active, the XI event mask is in + grab->deviceMask. Otherwise, for explicit grabs, the XI event mask is in + grab->eventMask. + + Reported-by: Thomas Jaeger + Signed-off-by: Peter Hutterer + +commit 29ec14625111a8ae600d245ccbf905f059074d7c +Author: Peter Hutterer +Date: Fri May 29 13:26:22 2009 +1000 + + Xi: passively grabbing slave devices must be possible. + + There's use-cases where this is useful, so take the check out preventing + that. + + Reported-by: Thomas Jaeger + Signed-off-by: Peter Hutterer + +commit 235cf5713a8fe535c439628cce9ea5b42432acb1 +Author: Peter Hutterer +Date: Fri May 29 12:09:06 2009 +1000 + + dix: init inputMasks to NULL to avoid random value dereference. + + If the XI2 type is 0 (e.g. proximity events), inputMasks is never set and + may not be NULL. + + Signed-off-by: Peter Hutterer + +commit 68d86adbe8d84099812c37a04098f2ab5ccc45a4 +Author: Peter Hutterer +Date: Fri May 29 12:03:07 2009 +1000 + + dix: GetProximityEvents doesn't need extra events for valuators. + + With internal events, we only have one event for all the data, no need to + calculate for extra events. + + Reported-by: Thomas Jaeger + Signed-off-by: Peter Hutterer + +commit 71a83d68ab07536c0778d13cd7da974c40e77457 +Author: Adam Jackson +Date: Thu May 28 15:32:03 2009 -0400 + + xfree86: Remove loader symbol list ABI stubs + + These have been nops since 2006 and functionally void since 7.0. + +commit d7aef3f663f4b5d861799f8615dcd301d5ce2906 +Merge: a25f248 e08c6a0 +Author: Peter Hutterer +Date: Thu May 28 17:20:58 2009 +1000 + + Merge branch 'master' into xi2 + + Conflicts: + Xext/geext.c + Xi/chdevcur.c + Xi/extgrbdev.c + Xi/xiproperty.c + configure.ac + dix/ptrveloc.c + hw/xfree86/common/xf86Config.c + mi/mipointer.h + test/input.c + xkb/xkb.c + +commit a25f248fc3bd0375d91ca8a44320200d445ecfbb +Author: Peter Hutterer +Date: Wed May 27 15:47:12 2009 +1000 + + Xi: Send Enter or Leave events with XIPassive(Un)grabNotify + + If a passive enter or focus in grab activates, send additional enter or + focus events with mode XIPassiveGrabNotify to the grabbing client. + Likewise, if the grab deactivates, send additional leave or focus out + events. + + Signed-off-by: Peter Hutterer + +commit 2107becb0ce2ffda001be65728c53563496d8d50 +Author: Adam Jackson +Date: Wed May 27 17:13:19 2009 -0400 + + xfree86: Remove unused (and useless) xf86{Add,Delete}ModuleInfo + +commit e08c6a0752772745f35f7afcf6c2b1c927b91cc0 +Author: Federico Mena Quintero +Date: Mon May 4 19:24:47 2009 -0500 + + randr: bug #21554 - re-probe outputs when coming back from laptop unsuspend + + Make xf86 RANDR wrap the EnterVT call chain, so it can re-probe the + outputs when a laptop comes back from suspend/unsuspend (actually, any + time that we enter our VT again). The X server should then send RR* + events to clients, so they can cope with a monitor that was unplugged + while the laptop was suspended. + + Signed-off-by: Federico Mena Quintero + +commit b12010e10f38951358b48ff1076c026f943b7cc3 +Merge: 6e69272 f1441b8 +Author: Søren Sandmann Pedersen +Date: Wed May 27 15:56:30 2009 -0400 + + Merge branch 'master' of git+ssh://sandmann@git.freedesktop.org/git/xorg/xserver + +commit 6e6927247359cc0db47501fb9d2dbb7856aa5985 +Author: Søren Sandmann Pedersen +Date: Wed May 27 15:51:19 2009 -0400 + + Fix alpha map computation in miComputeCompositeRegion() + + According to the RENDER spec, the origin of the alpha map is + interpreted relative to the origin of the drawable of the image, not + the origin of the drawable of the alpha map. + + This commit fixes that and adds an alpha-test.c test program. + + The only use of alpha maps I have been able to find is in Qt and they + don't use a non-zero alpha origin. + +commit f1441b83538e6ba8a1557c4f5ea50c53a1837087 +Author: Jeremy Huddleston +Date: Wed May 27 10:06:10 2009 -0700 + + XQuartz: More localization updates + (cherry picked from commit b107390c9f30571902556b4d053e62d345a06254) + +commit 996357e905c1082479bb238110b93bc170b8cb84 +Author: Joe Krahn +Date: Tue Jan 20 17:22:37 2009 +0000 + + Cygwin/X: Update icons directly, rather than modifying the window's class + + Update icons directly to windows rather than modifying + the window's class. Respect custom icons overriden via + the configuration file. + + fd.o bugzilla #4491 + + Signed-off-by: Jon TURNEY + +commit 9955252564922e7e8702dfb6eab52c75605a4a22 +Author: Jon TURNEY +Date: Mon Jan 5 17:55:58 2009 +0000 + + Cygwin/X: Remove unused TimeSinceLastInputEvent() + + Signed-off-by: Jon TURNEY + +commit 966c8fccf16e77c8f459b4b175718323f2ec5d58 +Author: Colin Harrison +Date: Mon Jan 5 18:04:42 2009 +0000 + + Xming: Fix some log spam + + Signed-off-by: Jon TURNEY + +commit d190a27a33ab8c198568baf184fad14df57e24ad +Author: Colin Harrison +Date: Thu Jan 29 11:00:49 2009 +0000 + + Xming: Various tidy ups in winClipboardFlushXEvents() + + Signed-off-by: Jon TURNEY + +commit fbe9ed27b1c2f932a74aa6ced134ecec021a3fc1 +Author: Colin Harrison +Date: Thu Jan 29 10:55:38 2009 +0000 + + Xming: Avoid log spam if a windows application won't provide CF_UNICODETEXT clipboard format + + Signed-off-by: Jon TURNEY + +commit 71ba9856a4f01aa7a42f1178c8da98a2e5ac23ae +Author: Colin Harrison +Date: Tue Feb 3 15:52:11 2009 +0000 + + Xming: Cache atom lookups in clipboard integration code + + Cache the CLIPBOARD atom lookups in winClipboardWindowProc() + Cache atom lookups in winClipboardFlushXEvents() + Recache on server regeneration + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit 4ec110327bdc4f2617cd6116539812d7ef96b24e +Author: Colin Harrison +Date: Sun Jan 4 18:29:49 2009 +0000 + + Xming: Prevent the mouse wheel from stalling when another window is minimized. + + Fix internal WM to prevent the mouse wheel from stalling when another window is minimized. + + Signed-off-by: Jon TURNEY + +commit 888e6961a4ee76d05d212cfb946f089caafb3f69 +Author: Jon TURNEY +Date: Wed Jan 7 18:29:16 2009 +0000 + + Cygwin/X: Correctly allow for the native window frame width in ValidateSizing() + + Fix internal WM to correctly calculate the native window border when validating window sizing + + Signed-off-by: Jon TURNEY + +commit a72865868f03b675f86990476fcee601822894b3 +Author: Joe Krahn +Date: Tue Jan 20 17:09:43 2009 +0000 + + Cygwin/X: Consolidate dialog initialization in winInitDialog() + + Changed windialogs.c to set icons via window properties rather than class + properties, and use LoadImage() for small icons, because LoadIcon() can only open + large icons. Since this code is redundant across the dialogs, I put it in the + winCenterDialog function, along with a few other redundant instructions, and + renamed in winInitDialog(). + + Also, don't bogusly put our dialogs at the center of the virtual desktop if we + are on a multimonitor system (this causes the dialog to end up split across two + monitors in a dual-monitor side-by-side setup) + + Corrections to use HWND_TOPMOST instead of HWND_TOP and not to use SWP_NOZORDER + from Colin Harrison + + fd.o bugzilla #4491 + + Signed-off-by: Jon TURNEY + +commit d053c3b35f730fdaddd22f841dbd8ed6c9a07ba0 +Author: Colin Harrison +Date: Sun Nov 2 20:46:40 2008 +0000 + + Xming: Slightly adjust the formatting of the logged command line + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit a7e23a79c1fc429aedbf9b6c0e78b1c8d7e02238 +Author: Peter Hutterer +Date: Tue May 26 14:42:25 2009 +1000 + + Xi: Add support for Enter and FocusIn grabs. + + Enter grabs are checked for in CheckMotion(), each time the sprite window + changes the current grab is deactivated (if applicable) and the new grab is + activated (if applicable). Exception - if the grab is on a parent window of + the current window since we keep the grab across descendants. + + Since CheckMotion() may change the grab status of a device, we mustn't get + "dev->deviceGrab.grab" in ProcessOtherEvents until after CheckMotion(). + FocusIn grabs are checked in much the same manner. + + The event delivery for grabs replaces the NotifyNormal on window change with + a NotifyGrab on window change. Note that this happens before the grab + activates, so the EnterNotify(NotifyGrab) is still delivered to the window, + not to the grabbing client. This is in line with the core protocol semantics + for NotifyGrab events. + + Signed-off-by: Peter Hutterer + +commit ec2fe9660dbc0c16cdaca33b3b878011857e0fe2 +Author: Peter Hutterer +Date: Wed May 27 14:18:51 2009 +1000 + + Require inputproto 1.9.99.10 + +commit 0ef6ba9f3b9e68307bfe623e12dc36d44440e97f +Author: Peter Hutterer +Date: Wed May 27 11:15:54 2009 +1000 + + Xi: set the passive grab's resource mask on ungrabbing. + + Not having the resource mask set means we never match an existing grab, + hence we never actually ungrab. + + Signed-off-by: Peter Hutterer + +commit 1f61d6647f25add487c5cae9739643a6e33ef9f7 +Author: Peter Hutterer +Date: Wed May 27 11:15:23 2009 +1000 + + Xi: use GetMaster(MASTER_KEYBOARD) to get the modifier device for XI2 pgrabs + + Signed-off-by: Peter Hutterer + +commit bedfd561fa534b65baea45c4a05b1752ef172c45 +Author: Peter Hutterer +Date: Tue May 26 15:28:33 2009 +1000 + + dix: fix wrong cast of eventMask into CreateGrab. + +commit 023f959e6812e63f9d38b078158275bc13d238ae +Author: Peter Hutterer +Date: Tue May 26 14:56:40 2009 +1000 + + Xi: fix an inputInfo.keyboard usage. + + In the case of a RevertToFollowKeyboard, the master device should be used + (since this is the closest equivalent to the VCK as before). Only if the + master keyboard is the same as the device, revert to the VCK itself. + +commit b3463fd6a89d50a7562c357aeb5314ce0fd6493d +Author: Peter Hutterer +Date: Tue May 26 13:58:04 2009 +1000 + + dix: convert window coordinates to FP1616 before calculating event_x/y + + root_x/y are already in FP1616 at this point. + + Signed-off-by: Peter Hutterer + +commit 10b5e981016132b4c07ad9ce68646d0a385e9c24 +Author: Peter Hutterer +Date: Tue May 26 11:12:08 2009 +1000 + + dix: take grabs into account for XI_Enter/Leave events. + +commit 1815defdb2f19e79f4ec0354d188ea763ff1f15e +Author: Peter Hutterer +Date: Tue May 26 10:08:46 2009 +1000 + + Xi: only deactivate passive grabs if the event type matches. + + This didn't use to be a problem when devices could only be pointers or + keyboards, not both. Nowadays, slave devices may have both buttons and + keyboards, and in this case we don't want to deactivate a passive keyboard + grab when a button release is detected. + +commit 9dc3810a034f067096d71b33b0b47f1f52092fe6 +Author: Peter Hutterer +Date: Wed May 27 13:53:46 2009 +1000 + + dix: set the right grab event type for implicit passive grabs. + + Signed-off-by: Peter Hutterer + +commit 6583477035234e23ead2fad9db7a07e5862447a4 +Author: Nicolai Hähnle +Date: Sat May 23 13:35:24 2009 +0200 + + Remove reference to non-existing requestLog and requestLogIndex + + These fields were removed in 252ec504817e05b185e4896a2d899e9c00b8aeef. + + Signed-off-by: Nicolai Haehnle + Signed-off-by: Peter Hutterer + +commit 53e821ab4a41cbc85932b9ef150311bc2e57c108 +Author: Peter Hutterer +Date: Mon May 25 12:20:37 2009 +1000 + + Xi: add request processing for XIGetSelectedEvents. + + Signed-off-by: Peter Hutterer + +commit b377994336c978875dd786f4a4a9ef2e7b20e32c +Author: Peter Hutterer +Date: Mon May 25 13:55:17 2009 +1000 + + Xi: advance by the right number of bytes when trawling XI2 event masks. + + The wire layout is [struct xXIEventMask][mask bytes]. So the pointer needs + to not only be advanced by the mask bytes, but also by the size of the + struct. + + Signed-off-by: Peter Hutterer + +commit 12e725d08b4cf7dbb7f09b9ec09fa1b621156ea9 +Author: Julien Cristau +Date: Fri May 22 09:54:38 2009 +0200 + + randr: fix server crash in RRGetScreenInfo + + We don't return rates to randr < 1.1 clients, so don't allocate space + for them. This fixes a FatalError due to not all allocated space being + used. + + X.Org bug#21861 + + Reported-by: Guillaume Quintin + Signed-off-by: Julien Cristau + +commit c70511931ca1441878daed50e710fa514f6ab077 +Author: Matthieu Herrb +Date: Sun May 24 10:32:38 2009 +0200 + + fix typo in cabff9007 which led to an unintialized memory read and a crash. + +commit 99e22b86c5f1a3653f3caaf01368a777d2b208d0 +Author: Adam Jackson +Date: Fri May 22 12:01:55 2009 -0400 + + EDID: Add modes from Established Timings III descriptor to mode pool + + EDID 1.4, section 3.10.3.9 + +commit b395da91c531d633ec47bb5a51e361d321663a3a +Author: Peter Hutterer +Date: Fri May 22 15:24:46 2009 +1000 + + Xi: silence two compiler warnings + + chdevcur.c:97: warning: ‘SecurityLookupIDByType’ is deprecated (declared at + ../include/resource.h:269) + + xiproperty.c:200: warning: passing argument 2 of ‘GetEventFilter’ from + incompatible pointer type + + Signed-off-by: Peter Hutterer + +commit 4a36db60774640ea92a2133a3b08fbf1a48d2cb6 +Author: Peter Hutterer +Date: Fri May 22 15:16:43 2009 +1000 + + os: fix compiler warning "too few arguments to format" + + Signed-off-by: Peter Hutterer + +commit da0d3baf71b34657cc235d6c6b37d548541f9449 +Author: Peter Hutterer +Date: Thu May 21 14:11:27 2009 +1000 + + dix: protect event generation against single-valuator devices. + + If we have a single-axis device and it sends events it should not access + non-existant memory. + +commit b25e29e8014927815d4fdc9c416bfeb2af0ecba6 +Author: Peter Hutterer +Date: Thu May 21 12:07:03 2009 +1000 + + mi: use GetMaster() from MIPOINTER and MISPRITE. + + Both may in some cases be called for a SD attached to a master keyboard. In + this case, we need to get the right master device (i.e. the pointer). + + Signed-off-by: Peter Hutterer + +commit c9df51b070377ca33e48644dbc842b6e3dfb2975 +Author: Peter Hutterer +Date: Thu May 21 11:16:37 2009 +1000 + + input: allow for master pointers to not have a button class. + + There's devices (e.g. some barcode readers) that have axes but no buttons. + When such a device sends a motion event, the valuator and button class is + copied into the master pointer (i.e. removing the button class). + So we need a couple of extra sanity checks for the button class to exist. + + Signed-off-by: Peter Hutterer + +commit ebe45e1a72a81ad22413e5d0514869b3e45a69b7 +Author: Peter Hutterer +Date: Wed May 20 10:55:03 2009 +1000 + + input: introduce partial class copying depending on the event. + + Copying all classes into the master device has drawbacks for hybrid devices + (devices that are both mice and keyboards). If such a device posts an event, + it's key classes are moved into the VCP. The key event itself is unaffected + by keyboard grabs and the like. + + Partial class copying copies depending on the event and copies the classes + into the right master device (i.e. the VCK for key events, the VCP for + pointer events). + + Signed-off-by: Peter Hutterer + +commit bc63c8a4570c989f19a036965854bceb9800ce19 +Author: Peter Hutterer +Date: Wed May 20 12:33:49 2009 +1000 + + dix: introduce GetMaster() + + For hybrid devices (keys + buttons/axes) the attached master device is + generally the wrong one. One shouldn't post a button event through a + keyboard and vice versa. + + GetMaster(dev) returns the right master device for the given type needed. + This may be the MD paired with this device's MD. + + Signed-off-by: Peter Hutterer + +commit d79318f269d959d566ec66239b4c985afd61b259 +Author: Peter Hutterer +Date: Thu May 21 08:29:14 2009 +1000 + + dix: Add a deviceid to the DeviceChangedEvent. + + ChangeDeviceId would actually overwrite the flags field if deviceid wasn't + present. Aside from the event of course not telling which device generated + it in the first place. + + Signed-off-by: Peter Hutterer + +commit eeffb83fea2d87070beafd0c692fb0ad54d8335b +Author: Peter Hutterer +Date: Mon May 18 20:03:07 2009 +1000 + + ABI_XINPUT_VERSION 6 + +commit 438a4eafa5d14ab676827f7a5d58782c02db4cdf +Author: Peter Hutterer +Date: Wed May 20 17:02:50 2009 +1000 + + input: remove nested union from InternalEvent. + + There's no need for internal events to be a struct with a single nested + union, we might as well make the union itself the InternalEvent. + + Signed-off-by: Peter Hutterer + +commit b12d302df8283186ce87882c29b2b0294adb2770 +Author: Peter Hutterer +Date: Wed May 20 15:28:16 2009 +1000 + + Input: rename DeviceIntRec->isMaster to ->type. + + isMaster is not enough as long as we differ between master pointers and + keyboard. With flexible device classes, the usual checks for whether a + master device is a pointer (currently check for ->button, ->valuators or + ->key) do not work as an SD may post an event through a master and mess this + check up. + + Example, a device with valuators but no buttons would remove the button + class from the VCP and thus result in the + IsPointerDevice(inputInfo.pointer) == FALSE. + + This will become worse in the future when new device classes are introduced + that aren't provided in the current system (e.g. a switch class). + + This patch replaces isMaster with "type", one of SLAVE, MASTER_POINTER and + MASTER_KEYBOARD. All checks for dev->isMaster are replaced with an + IsMaster(dev). + +commit add2defac7e59dcb355e5ba12b6de60861741d57 +Author: Peter Hutterer +Date: Wed May 20 16:23:38 2009 +1000 + + Split the signal-handler's lastSlave out into a separate variable. + + dev->u.lastSlave was not signal safe since it was accessed by the DIX and + during signal handling. + Replaced with: + 'dev->last.slave' for the signal handler's lastSlave (used to generate + DeviceChangedEvents), . + 'dev->u.lastSlave' for the DIX lastSlave (currently only used in + change_modmap) + + Signed-off-by: Peter Hutterer + +commit 525aa17f804d37d1cfcbbf6b8e6cddb45e999b20 +Author: Tomas Janousek +Date: Wed May 20 15:03:01 2009 +0200 + + Bug #6428, #16458, #21464: Fix crash due to uninitialized VModMap fields. + + In ProcXkbGetKbdByName, mrep.firstVModMapKey, .nVModMapKeys and + .totalVModMapKeys were not initialized, contained random values and caused + accesses to unallocated and later modified memory, causing + XkbSizeVirtualModMap and XkbWriteVirtualModMap to see different number of + nonzero values, resulting in writes past the end of an array in XkbSendMap. + + This patch initializes those values sensibly and reverts commits 5c0a2088 and + 6dd4fc46, which have been plain non-sense. + + Signed-off-by: Tomas Janousek + Signed-off-by: Peter Hutterer + +commit 7db55a0806c82bd4143c8bf1b8eb2b62e456ad9a +Author: Peter Hutterer +Date: Wed May 20 14:38:25 2009 +1000 + + dix: remove superfluous loop in change_modmap. + + A device can only be attached to a single master device. So instead of + looping and searching for the master device, we can just use dev->u.master + directly. + + Signed-off-by: Peter Hutterer + +commit cabff9007a4abad026b450a4aab155c7bcd94326 +Author: Peter Hutterer +Date: Fri May 15 08:33:07 2009 +1000 + + xfree86: treat other drivers as mouse drivers in the config. + + Historically, if no input device was referenced in the ServerLayout, + the server would pick the first "mouse" device found in the xorg.conf. + This patch gives evdev, synaptics, vmmouse and void the same status. If + there is a section in the config file using this driver - use it as the core + pointer. + + Device selection is in driver-order, not in config-order. If a "mouse" + device is listed after a "synaptics" device, the "mouse" device gets + preference. This replicates the original behaviour. + + This code only takes effect if AllowEmptyInput is off and there is no core + pointer in the server layout. + + Signed-off-by: Peter Hutterer + +commit a8bd1e1f96d8e5380972a7dce8d9940cd912aa09 +Author: Søren Sandmann Pedersen +Date: Thu May 21 18:21:28 2009 -0400 + + Fix alpha map computation in miComputeCompositeRegion() + + According to the RENDER spec, the origin of the alpha map is + interpreted relative to the origin of the drawable of the image, not + the origin of drawable of the alpha map. + + The only use of alpha maps I have been able to find is in Qt and they + don't use a non-zero alpha origin. + +commit a2c5ee36b21c2ee5c0468f1b251e74c1412dbecb +Author: Adam Jackson +Date: Thu May 21 10:20:48 2009 -0400 + + EDID: Be more cautious about finding vendor blocks. + + Many old monitors zero-fill the detailed descriptors, so check for that + to avoid a useless warning like: + + (WW) RADEON(0): Unknown vendor-specific block 0 + +commit 73c7398aafa050950ad614483fb0d5bdd2da97ca +Author: Peter Hutterer +Date: Wed May 20 10:40:50 2009 +1000 + + dix: 'namespace' HAS_OLD_SLAVE and HAS_NEW_SLAVE. + + We need more flags for this in the near future, so let's namespace them now. + +commit 5b5e3fa2771383a85afff679be34df19d3a4e290 +Author: Peter Hutterer +Date: Fri May 15 08:33:07 2009 +1000 + + xfree86: treat other drivers as mouse drivers in the config. + + Historically, if no input device was referenced in the ServerLayout, + the server would pick the first "mouse" device found in the xorg.conf. + This patch gives evdev, synaptics, vmmouse and void the same status. If + there is a section in the config file using this driver - use it as the core + pointer. + + Device selection is in driver-order, not in config-order. If a "mouse" + device is listed after a "synaptics" device, the "mouse" device gets + preference. This replicates the original behaviour. + + This code only takes effect if AllowEmptyInput is off and there is no core + pointer in the server layout. + + Signed-off-by: Peter Hutterer + +commit 1cce55cc0333a026474100cf2f784e220fd96473 +Author: Peter Hutterer +Date: Wed May 20 14:07:03 2009 +1000 + + input: rename device->type to device->xinput_type. + + This type is only used in XI to give a hint of what type this device may be. + Call it xinput_type for clarity. + + Signed-off-by: Peter Hutterer + +commit 81b3b0cce088866dc3cda099d7c8d6655849fd43 +Author: Tomas Janousek +Date: Wed May 20 15:03:01 2009 +0200 + + Bug #6428, #16458, #21464: Fix crash due to uninitialized VModMap fields. + + In ProcXkbGetKbdByName, mrep.firstVModMapKey, .nVModMapKeys and + .totalVModMapKeys were not initialized, contained random values and caused + accesses to unallocated and later modified memory, causing + XkbSizeVirtualModMap and XkbWriteVirtualModMap to see different number of + nonzero values, resulting in writes past the end of an array in XkbSendMap. + + This patch initializes those values sensibly and reverts commits 5c0a2088 and + 6dd4fc46, which have been plain non-sense. + + Signed-off-by: Tomas Janousek + Signed-off-by: Peter Hutterer + +commit 79138eec1b49cbaca6a16f2bdd8579b5828aeb28 +Author: Matthias Hopf +Date: Tue Mar 10 16:03:26 2009 +0100 + + randr: Setting gamma: inverse logic looks more sane + +commit c2785ae7eb6197bbfc75e92e99fffbb8ad8064da +Author: Peter Hutterer +Date: Wed May 20 14:29:13 2009 +1000 + + dix: refuse events from disabled devices. + + If the device is disabled ("off"), it must not send events to a client. + The driver shouldn't send events in that case anyway, but just to make sure + we simply drop events coming while the device is disabled. + + Signed-off-by: Peter Hutterer + +commit 47a89b1cba67a43ddc225f7e1b5d2b3b217c8d93 +Author: Peter Hutterer +Date: Wed May 20 13:59:22 2009 +1000 + + Xi: remove DeviceIsPointerType + + This approach is broken anyway. DIPT only checked for the XInput type + "MOUSE" and the only user of this is xf86ActivateDevice when it sets the + Activate/DeactivateGrab functions. + Since synaptics and wacom set their own types, evdev only sets MOUSE for, + well, mice half the devices didn't have this set correctly anyway. + + Instead, ActivatePointerGrab should be merged together with + ActivateKeyboardGrab. + + Signed-off-by: Peter Hutterer + +commit 7b891e450f005688e41e61016e6d3e394c67f6be +Author: Peter Hutterer +Date: Wed May 20 14:23:16 2009 +1000 + + dix: Remove obsolete comment + + CorePointerProc doesn't back up the device classes anymore. + +commit 42719ce5c715ac3a0c74c3c7df872d35e0449de9 +Author: Peter Hutterer +Date: Wed May 20 14:38:25 2009 +1000 + + dix: remove superfluous loop in change_modmap. + + A device can only be attached to a single master device. So instead of + looping and searching for the master device, we can just use dev->u.master + directly. + + Signed-off-by: Peter Hutterer + +commit 22a33a7250f879b8f3eb84176a1dc57027d8cfe4 +Author: Eamon Walsh +Date: Tue May 19 22:45:26 2009 -0400 + + glx: Register names for the GLX resource types. + + Signed-off-by: Eamon Walsh + +commit 77e90261d5f91e1170c6e02deb0e3739c24fcfd6 +Author: Eamon Walsh +Date: Tue May 19 19:36:03 2009 -0400 + + dix/events.c: Use wClient where appropriate. + + Signed-off-by: Eamon Walsh + +commit 3cea176d5abcb0f14eefbdcbe17fed0847524dd4 +Author: Eamon Walsh +Date: Tue May 19 19:30:33 2009 -0400 + + xace: Fix a bad device access hook call. + + Add a proper access mode, and reverse the logic of the return value. + Zero ("Success") is returned on success from the hook calls. + + Signed-off-by: Eamon Walsh + +commit 4addfcd633484b7f53b5df1b92bfca8bfc1dc48d +Author: Michel Dänzer +Date: Mon May 18 17:53:35 2009 +0200 + + EXA: Allocate from the end of free offscreen memory rather than from the start. + + This way we don't always need to scan over previously allocated areas when + looking for an available one, and there might be less fragmentation. + + Signed-off-by: Michel Dänzer + +commit 510cbd43cd4e34bd459e8f74ab2855714b4ca95d +Author: Michel Dänzer +Date: Mon May 18 17:48:57 2009 +0200 + + EXA: Defragment offscreen memory. + + At most once per second, under the following circumstances: + + * We can't satisfy an offscreen memory allocation, but there seems to be enough + offscreen memory available in total. + + or + + * The server has been idle for at least 100ms, and there is more than one + available offscreen area. + + Signed-off-by: Michel Dänzer + +commit 8331bde0adeccefb275c4d707e7b2cb1d95b1581 +Author: Jon TURNEY +Date: Fri May 15 21:36:28 2009 +0100 + + Cygwin/X: link with libmain.a to provide main() + + Commit 987579c930bda803427a28cb82773c389f5110d6 moves main.c into + into libmain.a to be linked with separately, and updates the various + DDXs to link with libmain.a, except Xwin, which also needs this change. + + Signed-off-by: Jon TURNEY + +commit b89dcfbfbd38f8d19f1d4f24f81820b1048572c8 +Author: Peter Hutterer +Date: Mon May 18 16:07:08 2009 +1000 + + Xi: fix length field in XIQueryPointer. + + Buttons append 4-byte units to the end of the reply, they need to be + included too. + + Reported-by: Mark Dokter + Signed-off-by: Peter Hutterer + +commit 33fcaaaea5ea30be39156ecfdbcca891b47ab465 +Author: Peter Hutterer +Date: Mon May 18 16:05:35 2009 +1000 + + Fix missing parentheses in FP1616 macro. + + Missing parens led to interesting results if an expression instead of a + constant was passed in (ProcXIQueryPointer for example). + +commit 6258bb1d6bb78ea4ba1d33470bc74e9b711b25f2 +Author: Peter Hutterer +Date: Fri May 15 20:24:45 2009 +1000 + + require inputproto 1.9.99.9 + +commit e7e94ff5fd3205a9ee01fe9342d9f3a426b0a068 +Author: Peter Hutterer +Date: Thu May 14 21:45:35 2009 +1000 + + Xi: set the right length for the XIQueryPointer reply. + +commit 00ab04d73c612ac85e024105015bdb7c79e8874b +Author: Peter Hutterer +Date: Thu May 14 16:24:35 2009 +1000 + + Include full modifier + button state in XIQueryPointer. + +commit de78aab0ea3a2d8c61222493f25911258dea65a2 +Author: Peter Hutterer +Date: Thu May 14 15:53:00 2009 +1000 + + Xi: take the paired device as modifier device only for MD pointers. + +commit 0e089f9ceb0edec532078c6e1df7f642f7220fb5 +Author: Peter Hutterer +Date: Thu May 14 15:29:20 2009 +1000 + + Xi: check cursor and grab_window before attempting the passive grab. + + Both values need to fail the request, so we need to check them before + looping through the modifier masks. Otherwise, a wrong grab_window will + show up in each failed modifier return but not actually cause an error. + +commit 5c3e7289801c958e2c70e2629d3ebee122449be8 +Author: Peter Hutterer +Date: Wed May 13 16:54:01 2009 +1000 + + dix: count must be 1 if we have an XI2 match. + + This fixes events that activated a grab not getting delivered to the client. + + Signed-off-by: Peter Hutterer + +commit e083a0d0692baa5ad29002a38293b36dd3b4c06f +Author: Peter Hutterer +Date: Wed May 13 15:23:42 2009 +1000 + + Xi: set the correct length for XIPassiveGrab replies. + +commit 8ff1bff8f462b15e05fc298704e1e40cb244587f +Author: Peter Hutterer +Date: Tue May 12 21:33:54 2009 +1000 + + Update to new XI2 names + +commit bbf266cca3de726e3eb932d66dd5b8981ef40fa8 +Author: Peter Hutterer +Date: Tue May 12 21:24:07 2009 +1000 + + mi: un-deprecate miPointerWarpCursor + + This function was deprecated in ef68273f5bdb27a492ec0b69548ec4fbede46c08 + because it didn't take a device argument. The device argument was added in + 1c7568b8a1417257fa67c7fca69aa253099b9461 though, so the deprecation is + obsolete. + + Signed-off-by: Peter Hutterer + +commit 273890924b8ed6f8b7949c0322c8258b9e6f8630 +Author: Peter Hutterer +Date: Tue May 12 16:09:35 2009 +1000 + + input: reduce the number of superfluous hierarchy events + + There's only two reasons for hierarchy events: + - device is added, removed, etc. In this case we want to send the event as + it happens. + - devices are added in a XIChangeDeviceHierarchy request. In this case we + only want one event cumulating all changes. + +commit 033a2b12fcd02fa9a2c2f20a352bec0a43074512 +Author: Peter Hutterer +Date: Tue May 12 14:50:57 2009 +1000 + + Xi: set per-device hierarchy changed flags. + + Rather than have one field per hierarchy change, XI2 has two fields - one + generic one and one per-device that include the device-specific flags. + This requires some funky handling for removed devices, but oh well. + +commit 8fb51feae222ff4f4aa1c440b6973ce7383bbc09 +Author: Peter Hutterer +Date: Mon May 11 15:45:46 2009 +1000 + + xkb: if kbd init failed, NULL out the pointers after freeing them (#21278) + + Reproducible: + Configure a server that uses the keyboard driver with an invalid ruleset, + e.g. (Option "XkbRules" "foobar"). Ensure that Option "AllowEmptyInput" is + "off" in the ServerFlags or ServerLayout section. Start the server. + After failing to init the keymap, the server will try to clean up after the + device, double-freeing some xkb structs that have not been reset properly. + + X.Org Bug 21278 + + Signed-off-by: Peter Hutterer + +commit 0cea199eda772a5ddd5ce72e4823a00ec14405d4 +Author: Peter Hutterer +Date: Mon May 11 12:53:56 2009 +1000 + + kdrive: set Activate/Deactivate grab for input devices (#21591) + + X.Org Bug 21591 + +commit fdce58ca822f4e43e03e84076190f52d7104e01f +Author: Peter Hutterer +Date: Mon May 11 12:51:40 2009 +1000 + + dix: ensure Activate/DeactivateGrab has a valid value. + + Xephyr doesn't manually set Activate/DeactivateGrab for new devices, + resulting in a NULL-pointer dereference later when a grab is activated. + Avoid the segfault by ensuring that the pointer is always valid. + + Signed-off-by: Peter Hutterer + +commit e3f296d91dfe6b827195e1d387e1a04aa73b85c3 +Author: Peter Hutterer +Date: Sat May 9 10:33:20 2009 +1000 + + xfree86: Remove superfluous ifdef DEBUG checks. + + Signed-off-by: Peter Hutterer + +commit 2f6253376df45e508284ff7766b186cfc7fb56d6 +Author: Peter Hutterer +Date: Sat May 9 10:33:20 2009 +1000 + + xfree86: Remove superfluous ifdef DEBUG checks. + + Signed-off-by: Peter Hutterer + +commit b1b5ec45c1cb650ccb8c659218f9481379c777d9 +Author: Peter Hutterer +Date: Mon May 11 12:53:56 2009 +1000 + + kdrive: set Activate/Deactivate grab for input devices (#21591) + + X.Org Bug 21591 + +commit 6f9e22049862ee9ac7f604411d005d8bb1b2dd1c +Author: Peter Hutterer +Date: Mon May 11 12:51:40 2009 +1000 + + dix: ensure Activate/DeactivateGrab has a valid value. + + Xephyr doesn't manually set Activate/DeactivateGrab for new devices, + resulting in a NULL-pointer dereference later when a grab is activated. + Avoid the segfault by ensuring that the pointer is always valid. + + Signed-off-by: Peter Hutterer + +commit c1d7deca9281ba1df0df1abcea1fe7f841b10ff9 +Author: Peter Hutterer +Date: Tue May 12 21:24:07 2009 +1000 + + mi: un-deprecate miPointerWarpCursor + + This function was deprecated in ef68273f5bdb27a492ec0b69548ec4fbede46c08 + because it didn't take a device argument. The device argument was added in + 1c7568b8a1417257fa67c7fca69aa253099b9461 though, so the deprecation is + obsolete. + + Signed-off-by: Peter Hutterer + +commit 8609a4e883e917d1652fbe810108de99bc89713f +Author: Alan Coopersmith +Date: Fri May 15 11:22:32 2009 -0700 + + Correct some Sun license notices to Sun's standard X11 license format + + Signed-off-by: Alan Coopersmith + +commit 94c5093c8bc803d0b14429b5a4d07ae46da865cc +Author: David Marx +Date: Wed May 13 17:53:30 2009 -0700 + + Solaris: Make sure non-inline versions of asm routines end with ret statements + + Signed-off-by: Alan Coopersmith + +commit 2be5eecb4b0f30b9c55d12521415edbb88c1f8fd +Author: Julien Cristau +Date: Fri May 15 17:29:32 2009 +0200 + + xfree86: add edid quirk for Samsung Syncmaster 2333HD + + It reports vertical size in cm in the detailed mode. + X.Org bug#21750 + + Reported-by: Peter Poklop + Signed-off-by: Julien Cristau + +commit 7c8327f0a75087a85864256a9cea80dd4b86def5 +Author: Michel Dänzer +Date: Fri May 15 15:48:37 2009 +0200 + + EXA: Always damage glyph cache pixmap manually after uploading a glyph. + + Signed-off-by: Michel Dänzer + +commit 850675d4de4373e5df95507dbf2cd9affaaf54bc +Author: Michel Dänzer +Date: Fri May 15 15:48:37 2009 +0200 + + EXA: Take GC client clip type into account for migration. + + Fixes http://bugs.freedesktop.org/show_bug.cgi?id=18950 . + + Signed-off-by: Michel Dänzer + +commit c1bbac2708bbccbec01cecba39cdf2b8160c1dc3 +Author: Matt Turner +Date: Thu May 14 16:00:41 2009 -0400 + + Fix mem_barrier() on Alpha + + Some clean up (da086901c6579e41b28b1418fa12ebb82744fc87) went a bit + too far and removed the needed mem_barrier macro on Alpha. + + This re-adds it. + +commit 4cc33190ef1a4207c0f8fc74201cdf2568a239c2 +Author: Julien Cristau +Date: Thu May 14 17:01:57 2009 +0200 + + xfree86: Add two __FreeBSD_kernel__ checks + + Debian bug#525475 + + Reported-by: Petr Salinger + Signed-off-by: Julien Cristau + +commit ed9aecf8517bf74e943e3e5dd3ee407524e8f122 +Author: Adam Jackson +Date: Thu May 14 06:55:21 2009 -0400 + + COPYING: SGI FreeB 1.1 -> SGI FreeB 2.0 + +commit c66cf7b64c7bcb31c81b9fde27cc40c1440753b1 +Author: Adam Jackson +Date: Thu May 14 06:53:08 2009 -0400 + + Update several of my and/or Red Hat's licenses to standard form. + +commit 2075d4bf9e53b8baef0b919da6c44771220cd4a5 +Author: Michel Dänzer +Date: Thu May 14 11:46:41 2009 +0200 + + glx: If a destroyed window is bound to the current context, make it not current. + + Avoids subsequent crashes due to stale pointers to the DrawableRec, see + https://bugs.freedesktop.org/show_bug.cgi?id=21132#c15 and previous comments. + + Signed-off-by: Michel Dänzer + +commit 2c1190f888515292de01e60fe74657c34b99fd9e +Author: Michel Dänzer +Date: Thu May 14 11:39:16 2009 +0200 + + randr12: Initialize and keep track of updates to VidMode extension gamma value. + + This way clients querying the gamma value via the VidMode extension at least + get the last value set via the same, rather than always something bogus. + + Signed-off-by: Michel Dänzer + +commit fc3ce861cdab8606610726ce7c53f57d950c2407 +Author: Michel Dänzer +Date: Thu May 14 11:35:25 2009 +0200 + + randr12: Fix calculation of gamma ramp values. + + The reciprocal gamma value was missed in the first copy and this mistake was + propagated to the second one. + + Signed-off-by: Michel Dänzer + +commit f3c61377d0fb1f09a18833995556fc9ffd11e097 +Author: Alan Coopersmith +Date: Tue May 12 18:27:09 2009 -0700 + + COPYING file updates for git master changes since 1.6 branch + + Signed-off-by: Alan Coopersmith + +commit b746a00cffca5c553b607a8e9c1074294a23b443 +Author: Alan Coopersmith +Date: Tue May 12 16:11:01 2009 -0700 + + Resync COPYING file with notices in code base as of xorg-server-1.6.1 + + Remove notices for code no longer in tree (Xprint, Xgl, kdrive, cfb, etc.) + Add/update notices for new/changed code in tree + + Signed-off-by: Alan Coopersmith + +commit f250eea2e90fc50bec5214c2f41132b95edc2c46 +Author: Jerome Glisse +Date: Mon May 11 22:52:46 2009 +0200 + + DRI2: update DRI2 private drawable width & height according to X drawable + +commit 0952d12717031e9dda9e48123bb922d0f4e81834 +Author: Eamon Walsh +Date: Mon May 11 15:27:46 2009 -0400 + + xselinux: Relax ownership restriction on SetSelectionUseContext. + + Instead, clients should keep track of the selection instances they use. + +commit ac13145dbcb284293582435409d8a90f276785c5 +Author: Peter Hutterer +Date: Mon May 11 15:45:46 2009 +1000 + + xkb: if kbd init failed, NULL out the pointers after freeing them (#21278) + + Reproducible: + Configure a server that uses the keyboard driver with an invalid ruleset, + e.g. (Option "XkbRules" "foobar"). Ensure that Option "AllowEmptyInput" is + "off" in the ServerFlags or ServerLayout section. Start the server. + After failing to init the keymap, the server will try to clean up after the + device, double-freeing some xkb structs that have not been reset properly. + + X.Org Bug 21278 + + Signed-off-by: Peter Hutterer + +commit ebfd6688d1927288155221e7a78fbca9f9293952 +Author: Søren Sandmann Pedersen +Date: Sun May 10 23:27:56 2009 -0400 + + Make compositing with transformed windows work again. + + The coordinate translation was broken in pretty much every way + imaginable. + +commit c2b668e6fff11c52f2a3f0dc3f4d1fc2c5bcf3e6 +Author: Jeremy Huddleston +Date: Tue May 5 20:27:53 2009 -0700 + + XQuartz: pbproxy standalone: added missing variable declarations. + (cherry picked from commit 2d947d17a3ef5fb8cef959bff0eb08b8bb5b6548) + +commit 9c5b761c8c86cca56f4b4af272f42ba89c9c3ba5 +Author: Peter Hutterer +Date: Sat May 9 16:19:13 2009 +1000 + + Revert "xkb: write the _XKB_RF_RULES_PROP to each device." + + This commit shouldn't have been pushed, we're still sorting out the API we + want to use. + + This reverts commit 876910a951053f0bd31e30098de3da15a1c1f5d6. + + Signed-off-by: Peter Hutterer + +commit 1e816065e5ec3b9394dc1fa5815457a664e15fd9 +Author: Alan Coopersmith +Date: Fri May 8 21:31:01 2009 -0700 + + Don't printf NULL pointers on HAL connection error + + Fixes Solaris bug 6801386 Xorg core dumps on startup if hald not running + http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=6801386 + + Signed-off-by: Alan Coopersmith + +commit b680bda34da130ce408783f04214771471e41e8d +Author: Alan Coopersmith +Date: Thu Apr 30 18:49:06 2009 -0700 + + Fix a couple off-by-one array boundary checks. + + Error: Write outside array bounds at Xext/geext.c:406 + in function 'GEWindowSetMask' [Symbolic analysis] + In array dereference of cli->nextSib[extension] with index 'extension' + Array size is 128 elements (of 4 bytes each), index <= 128 + + Error: Buffer overflow at dix/events.c:592 + in function 'SetMaskForEvent' [Symbolic analysis] + In array dereference of filters[deviceid] with index 'deviceid' + Array size is 20 elements (of 512 bytes each), index >= 0 and index <= 20 + + Error: Read buffer overflow at hw/xfree86/loader/loader.c:226 + in function 'LoaderOpen' [Symbolic analysis] + In array dereference of refCount[new_handle] with index 'new_handle' + Array size is 256 elements (of 4 bytes each), index >= 1 and index <= 256 + + These bugs were found using the Parfait source code analysis tool. + For more information see http://research.sun.com/projects/parfait + + Signed-off-by: Alan Coopersmith + Signed-off-by: Adam Jackson + Acked-by: Peter Hutterer + +commit 00bc043fa0398a1d14d46b87da2ff3031a9535dc +Author: Peter Hutterer +Date: Fri May 8 10:10:25 2009 +1000 + + dix: export subpixel precision in XI2 events for root/event coordinates. + + Signed-off-by: Peter Hutterer + +commit bae070914fc27db122e6131ae4838559c4a72f65 +Author: Peter Hutterer +Date: Thu May 7 16:52:31 2009 +1000 + + input: update to inputproto 1.9.99.8 XI2 defines. + + Signed-off-by: Peter Hutterer + +commit 83f32d3972b8bfb0a87069dfb3fcd64b6b7c6424 +Author: Peter Hutterer +Date: Thu May 7 10:05:29 2009 +1000 + + Xi: Add XI2 property requests. + +commit 9935bec6e860cba9a3cc5baadd372ddb89d72ef0 +Author: Peter Hutterer +Date: Thu May 7 09:43:56 2009 +1000 + + Xi: split some code out of the XI 1.5 property request processing. + + This is in preparation for the XI2 property requests that can re-use much of + this code. + +commit f6f1e417063d2d61d65731a7e3ebca3ec2bdaacc +Author: Peter Hutterer +Date: Wed May 6 22:25:28 2009 +1000 + + Xi: fix copy/paste error causing sizeof against wrong struct. + + This wrong check may cause BadLength to be returned to the client even if the + length is correct. + + Signed-off-by: Peter Hutterer + +commit fc0013d744a345199f013ba5b6ef0e44201d0e68 +Author: Peter Hutterer +Date: Wed May 6 16:55:32 2009 +1000 + + Xi: don't double-swap the XListDeviceProperties reply. + + Signed-off-by: Peter Hutterer + +commit 409012061b7d361a172b20455d4aaa5e47527c42 +Author: Peter Hutterer +Date: Wed May 6 16:42:46 2009 +1000 + + Xi: add missing break in XI event swapping function + + Signed-off-by: Peter Hutterer + +commit c3c64978c4a231a3a8c18211d0716df875c75efa +Author: Peter Hutterer +Date: Wed May 6 15:05:06 2009 +1000 + + Xi: fix a couple of wrong dixLookupDevice permission tags. + + Signed-off-by: Peter Hutterer + +commit 0d947aa8e87c5d92b702c60190c8bc5d32c9ba9c +Author: Peter Hutterer +Date: Wed May 6 12:07:07 2009 +1000 + + xfree86: fix xf86PostMotionEventP type checking + + We only put internal events into the queue now, so let's check for ET_Motion + rather than the MotionNotify. + + Signed-off-by: Peter Hutterer + +commit 9457ce448988dacea9a3a1c14b6955767118ff31 +Author: Simon Thum +Date: Thu Apr 30 13:01:17 2009 +0200 + + dix: remove superfluous includes from ptrveloc.c + + Signed-off-by: Peter Hutterer + +commit ee30e1b102d779403eeda97e007b81e85f006a37 +Author: Simon Thum +Date: Thu Apr 30 12:58:48 2009 +0200 + + dix: fix warning in pointer acceleration + + newer gcc's warn against how this cast is done (though it eludes me why), + and lrintf() is also faster especially on insane processors like the P4 + (http://www.mega-nerd.com/FPcast). + + Signed-off-by: Peter Hutterer + +commit 5cf70183812541b33a6e83c7e1e3bc6198730cbe +Author: Peter Hutterer +Date: Tue May 5 21:00:31 2009 +1000 + + xkb: remove _XkbAlloc, _XkbCalloc, _XkbRealloc and _XkbFree + + We all agree that wrapping is fun, but seriously. One of these days someone + will get hurt. + + Signed-off-by: Peter Hutterer + +commit 0e31d3906deaee5d9ada66e538b9e93574a4d610 +Author: Peter Hutterer +Date: Tue May 5 21:07:07 2009 +1000 + + xkb: remove some now-useless XFUNCPROTOBEGIN + + We bring them back if we start rewriting the server in C++, promise. + + Signed-off-by: Peter Hutterer + +commit 8b583ca2b21155359c6255f406c96599b277c762 +Author: Peter Hutterer +Date: Wed May 6 22:25:28 2009 +1000 + + Xi: fix copy/paste error causing sizeof against wrong struct. + + This wrong check may cause BadLength to be returned to the client even if the + length is correct. + + Signed-off-by: Peter Hutterer + +commit 7ab5e9b97c300bba793a23fa13506b0c77c50ddf +Author: Peter Hutterer +Date: Wed May 6 12:07:07 2009 +1000 + + xfree86: fix xf86PostMotionEventP type checking + + We only put internal events into the queue now, so let's check for ET_Motion + rather than the MotionNotify. + + Signed-off-by: Peter Hutterer + +commit 50cc8adafca4ba3838d468278d6eb8a4692d2488 +Author: Peter Hutterer +Date: Wed May 6 16:55:32 2009 +1000 + + Xi: don't double-swap the XListDeviceProperties reply. + + Signed-off-by: Peter Hutterer + +commit 3fc6fcfb267888d65a22ed7c3eda175a2530a3d5 +Author: Simon Thum +Date: Wed May 6 10:39:16 2009 +0200 + + dix: add 'none' pointer acceleration profile with number -1 + + This is a shorthand for disabling acceleration, while retaining the + possiblity to use constant deceleration. If constant deceleration is + also unused, it will optimize motion processing. + + Other possiblities to deactivate acceleration were quite hidden, + and didn't always work as expected. E.g. xset m 1 1 would retain + adaptive deceleration, while xset m 1 0 would not (in the default + profile). + + Also removes the 'reserved' profile; it was unused and it's trivial + to add new ones anyway. + + Signed-off-by: Peter Hutterer + +commit 9d1597cbefea6a5e7959f2099c46f1a284def7ad +Author: Simon Thum +Date: Thu Apr 30 13:01:17 2009 +0200 + + dix: remove superfluous includes from ptrveloc.c + + Signed-off-by: Peter Hutterer + +commit 35fce4e5c7fa34f98b3e4010c6cb09ce38a9205c +Author: Simon Thum +Date: Thu Apr 30 12:58:48 2009 +0200 + + dix: fix warning in pointer acceleration + + newer gcc's warn against how this cast is done (though it eludes me why), + and lrintf() is also faster especially on insane processors like the P4 + (http://www.mega-nerd.com/FPcast). + + Signed-off-by: Peter Hutterer + +commit d220d6907d1d5138d1528c48b739e77f65616225 +Author: Peter Hutterer +Date: Mon May 4 17:38:02 2009 +1000 + + Xi: add GrabButton and GrabKeysym code. + + We don't do keycode grabs in XI2, they're pointless. + +commit 1b593ced171d02f1d00034f0f733060706bb6d41 +Author: Peter Hutterer +Date: Tue May 5 16:57:46 2009 +1000 + + Xi: access the prev_state to get group/modifier state for key events. + + Key events may change the modifier state, so we need to get the prev_state for + those (i.e. without the changes by the event already applied). + + Signed-off-by: Peter Hutterer + +commit 35a4b8e7f4526a92d44cb16a783f21030cd1f6df +Author: Peter Hutterer +Date: Tue May 5 16:51:59 2009 +1000 + + xkb: remove oldState from XkbHandleActions. + + I really don't know what the purpose of this variable is or was, aside from + potentially clobbering up our key state since there's a path where it may be + used uninitialised. + + Also, this means that xkbi->prev_state is now accessible from the DIX with + meaningful data. + + Signed-off-by: Peter Hutterer + +commit d523fbe428e4513d85402caa83e9349256b680cc +Author: Peter Hutterer +Date: Mon May 4 17:00:57 2009 +1000 + + test: add test for xi2 struct sizes. + +commit 1b1b20d6e3e696e4437a9ef56128cde70a485f66 +Author: Peter Hutterer +Date: Tue May 5 13:47:55 2009 +1000 + + Change glib require for tests to auto. + + Signed-off-by: Peter Hutterer + +commit 749cae82204f347248c815b1068ffcae9825fe70 +Author: Peter Hutterer +Date: Thu Apr 30 14:58:00 2009 +1000 + + mi: don't crash if we're trying to update the pointer for a keyboard. + + If a device doesn't have a pointer, just return. + + Signed-off-by: Peter Hutterer + +commit 7ecedb0f2ee5b53513205d1f0aa381451893b9c5 +Author: Peter Hutterer +Date: Fri May 1 14:35:12 2009 +1000 + + include: up the number of max. input devices to 40. + + With the Xtest virtual slave devices we have 4 devices for each MD + pointer/keyboard pair, plus the AllDevices and AllMasterDevices reserved + deviceids. It's quite easy to hit the current limit. + + Signed-off-by: Peter Hutterer + +commit a38d33f6ac765bc5a3296b325e2c9b4a6a10ebf1 +Author: Peter Hutterer +Date: Fri May 1 15:24:02 2009 +1000 + + Xext: shut up compiler warnings in xtest.c + + Signed-off-by: Peter Hutterer + +commit 14d5471981151b27678de826daa078977217477a +Author: Peter Hutterer +Date: Fri May 1 15:22:37 2009 +1000 + + Xext: return BadDevice from XTest if we don't have keys/buttons/valuators. + + BadDevice is an XI error, but this cannot happen for core XTest fake input + anyway since the device will be the matching virtual XTest slave device. + + Signed-off-by: Peter Hutterer + +commit f1bed0959e803784af142a0f45177ee3b675d78a +Author: Peter Hutterer +Date: Fri May 1 15:15:04 2009 +1000 + + Xext: fix core Xtest button presses, don't call PickPointer. + + We already did the device selection before, so dev should be the XTest virtual + pointer. + + Signed-off-by: Peter Hutterer + +commit 94648bb797d94b025746c60679c584e5be2fae28 +Author: Jesse Barnes +Date: Mon May 4 15:38:22 2009 -0700 + + Don't prepare outputs & crtcs if set_mode_major is present + + A driver with this hook will take care of preparing the outputs & crtcs, + so calling the prepare functions will just cause unnecessary flicker. + + Fixes bug #21077 + +commit 7d85169c7ab1d05c21fdbb2877ffd79f344dbb6b +Author: Michel Dänzer +Date: Mon May 4 10:14:43 2009 +0200 + + Fix typo in ProcRenderCreateAnimCursor. + +commit 59358de4b68a01452bf68593beb83605087ebc8f +Author: Peter Hutterer +Date: Mon May 4 17:30:19 2009 +1000 + + input: ensure various ProcUngrabKey/Buttons have the right grabtype set. + +commit 834ea071b619506e0bef5bdbf0e3c59831cf6e99 +Author: Peter Hutterer +Date: Mon May 4 15:58:15 2009 +1000 + + dix: if grab types differ, two grabs cannot be the same. + +commit 63a3c727b35d1ffd8a850a5177bd6623cd78b179 +Author: Peter Hutterer +Date: Fri May 1 15:24:02 2009 +1000 + + Xext: shut up compiler warnings in xtest.c + + Signed-off-by: Peter Hutterer + +commit 226dd90597ef29998ac5db15c5457aad3fe9729e +Author: Peter Hutterer +Date: Fri May 1 15:22:37 2009 +1000 + + Xext: return BadDevice from XTest if we don't have keys/buttons/valuators. + + BadDevice is an XI error, but this cannot happen for core XTest fake input + anyway since the device will be the matching virtual XTest slave device. + + Signed-off-by: Peter Hutterer + +commit 0fdff0a47a57a0096974ee83e07fefbad03075dd +Author: Peter Hutterer +Date: Fri May 1 15:15:04 2009 +1000 + + Xext: fix core Xtest button presses, don't call PickPointer. + + We already did the device selection before, so dev should be the XTest virtual + pointer. + + Signed-off-by: Peter Hutterer + +commit f0124ed93cdcee8d22a6690e639a3f60b9a63c54 +Author: Peter Hutterer +Date: Fri May 1 14:35:12 2009 +1000 + + include: up the number of max. input devices to 40. + + With the Xtest virtual slave devices we have 4 devices for each MD + pointer/keyboard pair, plus the AllDevices and AllMasterDevices reserved + deviceids. It's quite easy to hit the current limit. + + Signed-off-by: Peter Hutterer + +commit 08cd846000286eb5e07cc8d8cb313aff2d6e7faf +Author: Peter Hutterer +Date: Fri May 1 11:06:06 2009 +1000 + + dix: only free the old cursor if the grab was successful. + + Testcase: start thunderbird and move a message around, crashes after two or + three moves. + + Signed-off-by: Peter Hutterer + +commit 557dbadf3be273255e8fdb12d9321f4e88bf2b65 +Author: Alan Coopersmith +Date: Sat May 2 00:26:18 2009 -0700 + + XkbSetNamedIndicator should ignore SD's without LED's + + When ProcXkbSetNamedIndicator is called on a core device, and we + walk the slaves to set the LED's on each of them, ignore any slaves + that do not have either a KbdFeedbackCtrl or LedCtrl structure. + + (This is much more critical in xserver-1.5-branch, where we walk + *all* devices, not just the slaves of the specified master, and + thus return failure when setting an LED on the Core Keyboard and + hit a xf86-input-mouse device with no LED's to set.) + + Signed-off-by: Alan Coopersmith + Acked-by: Peter Hutterer + +commit 0d9d3f3e361f769822caedccf4c2a58cc9930ecc +Author: Ian Romanick +Date: Mon Apr 27 15:11:10 2009 -0700 + + DRI2: Force allocation of real-front buffer for non-windows as well + + For redirected rendering we end up with pixmaps (which the app thinks are + windows) that are double buffered. + + Signed-off-by: Ian Romanick + Tested-by: Pierre Willenbrock + +commit 3d30789a05a730a03faa6058c73a5eda36ef3779 +Author: Peter Hutterer +Date: Thu Apr 30 14:58:00 2009 +1000 + + mi: don't crash if we're trying to update the pointer for a keyboard. + + If a device doesn't have a pointer, just return. + + Signed-off-by: Peter Hutterer + +commit 876910a951053f0bd31e30098de3da15a1c1f5d6 +Author: Peter Hutterer +Date: Thu Apr 30 14:46:45 2009 +1000 + + xkb: write the _XKB_RF_RULES_PROP to each device. + + We only have one root window and writing the rules used to the same property + for each device is quite pointless if you don't have the same RMLVO on all + devices. So let's be sensible and write the same property to the device too, + so at least we know which device got loaded with which RMLVO. + +commit 44044adc92e6bb6537f48c84727523d5cb23528c +Author: Peter Hutterer +Date: Thu Apr 30 11:39:50 2009 +1000 + + xfree86: print a message if NIDR fails due to AutoAddDevices off. + + Signed-off-by: Peter Hutterer + +commit 43dbbc30327f826068bc7cd2cdaf870ed61b4522 +Author: Peter Hutterer +Date: Tue Apr 28 21:01:36 2009 +1000 + + Xi: fix typo in ProcXIGrabDevice + + Signed-off-by: Peter Hutterer + +commit 6a618929a0c06ba0d6dac13d7e644cd9658d98ed +Author: Peter Hutterer +Date: Tue Apr 28 16:49:45 2009 +1000 + + input: reshuffle CreateGrab and friends to take a GrabParameters param. + + This is cleaning up work in preparation for XI2 passive grabs. + +commit e8e26f700c9c70d3f1bb53bdb71d1100f5c43a69 +Author: Peter Hutterer +Date: Sun Apr 26 21:43:25 2009 +1000 + + Xi: split some grab parameter checking out of GrabButton and GrabKey. + +commit 57aff88c7d0761e590806d07bee1c9410680c89f +Author: Eamon Walsh +Date: Wed Apr 29 01:04:37 2009 -0400 + + Fix most remaining deprecated resource lookups. + + Callsites updated to use dixLookupResourceBy{Type,Class}. + TODO: Audit access modes to make sure they reflect the usage. + +commit 1abe0ee3da5e1268c7315f841d31337ea6524cf0 +Author: Eamon Walsh +Date: Tue Apr 28 23:51:40 2009 -0400 + + kdrive: fix Xvfb build with separate libmain. + +commit 737b49199a05299486064e6e762cf2a2f6f95be6 +Author: Peter Hutterer +Date: Thu Apr 9 15:40:24 2009 +1000 + + xfree86: restore default off for DontZap + + Zapping is triggered by xkb these days, so note in the man page that it's the + Terminate_Server action. Since it's XKB, personal preferences towards or + against zapping should be achieved through xkb rulesets. + If Terminate_Server is not in the xkb actions, then we can't zap anyway and we + don't need a default of DontZap "on". + + This patch restores the old meaning of DontZap - disallow zapping altogether, + regardless of XKB's current keymap. + Ideally, this patch should be accompanied by b0f64bdab00db652e in + xkeyboard-config. + + Signed-off-by: Peter Hutterer + +commit 98a1fa994158feca8bc80a10296a7675465c0e56 +Author: Peter Hutterer +Date: Tue Apr 21 22:18:23 2009 +1000 + + test: add InternalEvent to core event conversion tests. + + Signed-off-by: Peter Hutterer + +commit 990e8362a76fadc173e890d9052a2e5274b6eac5 +Author: Peter Hutterer +Date: Thu Apr 16 11:41:58 2009 +1000 + + test: add a simple test to verify device axis intialization. + + Signed-off-by: Peter Hutterer + +commit 4eac0df060b897470fb01cfd3b70f500da757c74 +Author: Peter Hutterer +Date: Wed Apr 15 17:21:08 2009 +1000 + + Add a test-suite for in-server unit-testing. + + This patch adds a test/ directory that contains the setup for a unit-testing + suite designed for in-server unit-testing. All functions available to the X + server are available to the test binaries through static linking. + + This test suite uses the glib testing framework. + Do not use glib calls outside of the test/ directory. + + Signed-off-by: Peter Hutterer + +commit c2c515ead38d9a6c9eae0b83aa7a82208f177b7e +Author: Peter Hutterer +Date: Wed Apr 22 13:26:40 2009 +1000 + + dix: remove all but main() from main.c + + All other functions are pushed into where they seemed to fit. + main.c is now linked separately into libmain.a and linked in by the various + DDXs. + + Signed-off-by: Peter Hutterer + +commit dc2767d1c5db60385867c76ba2de507fe0cb8a90 +Author: Peter Hutterer +Date: Tue Apr 21 22:18:23 2009 +1000 + + test: add InternalEvent to core event conversion tests. + + Signed-off-by: Peter Hutterer + +commit 1127ca097cb75450bcccfc5f5d82e435de2fb5b7 +Author: Peter Hutterer +Date: Thu Apr 16 11:41:58 2009 +1000 + + test: add a simple test to verify device axis intialization. + + Signed-off-by: Peter Hutterer + +commit 4124c465a85713fe44843a16c5e2b13ece17e9d2 +Author: Peter Hutterer +Date: Wed Apr 15 17:21:08 2009 +1000 + + Add a test-suite for in-server unit-testing. + + This patch adds a test/ directory that contains the setup for a unit-testing + suite designed for in-server unit-testing. All functions available to the X + server are available to the test binaries through static linking. + + This test suite uses the glib testing framework. + Do not use glib calls outside of the test/ directory. + + Signed-off-by: Peter Hutterer + +commit 987579c930bda803427a28cb82773c389f5110d6 +Author: Peter Hutterer +Date: Wed Apr 22 13:26:40 2009 +1000 + + dix: remove all but main() from main.c + + All other functions are pushed into where they seemed to fit. + main.c is now linked separately into libmain.a and linked in by the various + DDXs. + + Signed-off-by: Peter Hutterer + +commit 2a2a7fa6856b2e1954e904577535118f20cd9133 +Author: Alan Coopersmith +Date: Mon Apr 27 20:23:53 2009 -0700 + + Solaris: use instead of stale copy in Xorg sources + + Signed-off-by: Alan Coopersmith + +commit 707af5f8c5fae3b6c4c5125a91917667bb0ab908 +Author: Simon Farnsworth +Date: Thu Apr 23 16:15:01 2009 +0100 + + Make the cursor completely optional + + For embedded use, it's convenient to be able to disable the cursor + completely, without having to audit and fix up all your third-party + code (e.g. Mozilla Firefox). + + Add -nocursor and -cursor server options to enable and disable the + cursor. The default is still -cursor, but embedded users can run the + server with -nocursor to hide the cursor regardless of what + application developers do. + + Signed-off-by: Peter Hutterer + +commit 40a8f2f4088aa353de733a11327b628ef13a52e6 +Author: Pierre Willenbrock +Date: Sun Apr 19 21:15:22 2009 +0200 + + Fix obvious copypasta + + Reviewed-by: Ian Romanick + +commit a8cf63fd18b0ddadbadf7b47d620159854920050 +Author: Aaron Plattner +Date: Thu Apr 23 14:57:04 2009 -0700 + + Export CompositeRegisterAlternateVisuals. + + Drivers need this if they add visuals that require implicit redirection. Add a + new SDK header, compositeext.h, to contain the prototype for this function. + +commit 48573e7ea13e279593249036e6d30f1d7507882f +Author: Peter Hutterer +Date: Sun Apr 26 21:32:37 2009 +1000 + + Xi: swap stuff->cursor in SProcXIGrabDevice + +commit ff6c7764c2909e4126403b7211faa6c58556b341 +Author: Ian Romanick +Date: Fri Apr 24 12:49:19 2009 -0700 + + DRI2: Implement protocol for DRI2GetBuffersWithFormat + + This change implements the protocol for DRI2GetBuffersWithFormat, but + the bulk of the differences are the changes to the extension / driver + interface to make this function work. The old CreateBuffers and + DeleteBuffers routines are replaced with CreateBuffer and DeleteBuffer + (both singular). + + This allows drivers to allocate buffers for a drawable one at a time. + As a result, 3D drivers can now allocate the (fake) front-buffer for a + window only when it is needed. Since 3D drivers only ask for the + front-buffer on demand, the real front-buffer is always created. This + allows CopyRegion impelemenations of SwapBuffers to continue working. + As with previous version of this code, if the client asks for the + front-buffer for a window, we instead give it the fake front-buffer. + + Signed-off-by: Ian Romanick + Reviewed-by: Kristian Høgsberg + +commit 28ddfc88d8d547941c7f4713db527a3c2f9ec35a +Author: Ian Romanick +Date: Fri Apr 24 12:09:21 2009 -0700 + + DRI2: Add interface for drivers to query DRI2 extension version + + Signed-off-by: Ian Romanick + +commit a1c98f7fa9aa0cd5075ef2054c5d2be4d1e740f0 +Author: Joel Bosveld +Date: Fri Apr 24 10:12:26 2009 -0700 + + Correct fcntl(F_SETOWN) error check in non-Solaris sigio code + + commit 64b7f96dca accidentally inverted the comparison, could + result in crashes on some systems later on. + + Signed-off-by: Alan Coopersmith + +commit 64b7f96dca23d2b33e0b8a8785d628d9a2f50e7b +Author: Alan Coopersmith +Date: Fri Apr 24 00:14:12 2009 -0700 + + Add SIGIO/SIGPOLL support for Solaris + + Signed-off-by: Alan Coopersmith + +commit 0c467b1f990f156a34d5ef117bd825df6b633114 +Author: Alan Coopersmith +Date: Fri Apr 24 00:05:28 2009 -0700 + + Add casts to generated entries in sdksyms.c to silence type mismatch warnings + + Signed-off-by: Alan Coopersmith + +commit 7d0f7518c2235a9dc783029971259ddaada2db20 +Author: Alan Coopersmith +Date: Thu Apr 23 08:42:38 2009 -0700 + + Fix byte swapping of XF86VidMode{Get,Set}GammaRamp + + Fixes OpenSolaris Bug 8315: + Xorg segfaults when screensaver fades in cross-endian xdmcp session + + + Signed-off-by: Alan Coopersmith + +commit 057fc9a4f89282e440b5e11458f1dd8655879dd8 +Merge: 95628b7 932d6bc +Author: Peter Hutterer +Date: Fri Apr 24 16:15:47 2009 +1000 + + Merge branch 'master' into xi2 + + Conflicts: + Xi/chdevhier.c + include/input.h + +commit 932d6bcbb68194c5bdfeb336f700dc8b31529223 +Author: Peter Hutterer +Date: Fri Apr 24 15:28:45 2009 +1000 + + mi: remove superfluous check. + + mieqProcessInputEvents doesn't process events from MDs anymore, so we don't + need to check for pDev->isMaster. + + Signed-off-by: Peter Hutterer + +commit fab563bf8f6b63906ce9d5a568c467425843265b +Author: Benjamin Close +Date: Fri Mar 27 16:44:15 2009 +1030 + + input: propagate XTst events through virtual slave devices. + + A XTest virtual slave device pair (kbd/ptr) exists for every master + device pair. This is so XTest events are correctly propogated via slave + devices up to Master devices and the classes are correctly changed along + the way. We add the XTest slave device pair to the Virtual Core pointer + and provide a simple way of creating the devices. + + A XTest Slave Device is identified by the XTstDevicePrivateKey property + being set in the devices devProperties + + XI events are still propagated through the matching device, in the hope the + client knows what it is doing. + + Signed-off-by: Peter Hutterer + +commit ccd28ca2ad40aecf951ab058d89062828e3e5d8a +Author: Jeremy Huddleston +Date: Wed Apr 22 19:23:17 2009 -0700 + + XQuartz: Localization update + (cherry picked from commit 9435fc5e20e25ce66ec85bde033daff51f39f69c) + +commit a362d0f4fb79150f7b8e0eb3d1dc1cdee2f98d95 +Author: Jeremy Huddleston +Date: Sat Apr 11 19:06:45 2009 -0700 + + XQuartz: Make sure the bgMouseLocationUpdated state is consistent + (cherry picked from commit 8fa8a8e1db60c8ac60d10f1f15517471d2291f77) + +commit 8dc8812c226222fc64f8e489f9f123687a95d3d1 +Author: Dave Airlie +Date: Thu Apr 23 12:13:35 2009 +1000 + + exa: add missing exa.h header include + +commit 3ec6a121e110ba3cae964ee201d4d5bf818b6dff +Author: Dave Airlie +Date: Thu Apr 23 12:13:05 2009 +1000 + + exa: missed exa.h change + +commit 02ae85c4c93bbd0013c063ce0014b365a8ba061f +Author: Dave Airlie +Date: Thu Apr 23 12:04:17 2009 +1000 + + exa: add CreatePixmap2 hook for driver pixmaps. + + This adds a revised pixmap hook for driver pixmaps, which is + required to support tiling on various chips. + + Signed-off-by: Dave Airlie + +commit 9d684ba0bcab53841a17ebd521507b6df8bc4b45 +Author: Dave Airlie +Date: Thu Apr 16 10:28:36 2009 +1000 + + exa: avoid offscreen pixmap swapped out flag for driver pixmaps + +commit 0dfb97f15f591f85e079f5829c77d0c328d00464 +Author: Julien Cristau +Date: Wed Apr 22 20:01:49 2009 +0200 + + Bug#21324: Add quirk for Iiyama Vision Master 450 + + Reported-by: Jeremy Henty + Signed-off-by: Julien Cristau + +commit 44227ef1b77467c76147b9bf79bdd0e6305a522a +Author: Ian Romanick +Date: Mon Apr 20 18:20:52 2009 -0700 + + DRI2: Send the version the code actually supports + + This prevents building an older server with a new dri2proto.h from + resulting in a DRI2 extension module that lies about the version it + supports. + + Signed-off-by: Ian Romanick + +commit d1e916d29be8b470cbc8cadcf6e83991fdbc5a9f +Author: Ian Romanick +Date: Thu Apr 16 12:10:34 2009 -0700 + + DRI2: Add missing front-buffer flush callback. + + Signed-off-by: Ian Romanick + +commit 826a5bff0136b2b4d55a9e6e6bc3a7a64da9031e +Author: Benjamin Close +Date: Thu Apr 16 15:33:30 2009 +1000 + + dix: Change AllocMaster into AllocDevicePair, allow creation of SDs too. + + Allocating a slave device is essentially the same as allocating a master device. + Hence we rename AllocMaster to AllocDevicePair and provided the ability to + indicate if a master or slave device pair is required. + + Signed-off-by: Peter Hutterer + +commit e2e5932bda3f473629d4be6f3ca4dcab18993eb6 +Author: Peter Hutterer +Date: Thu Apr 16 11:06:52 2009 +1000 + + dix: don't allow more than MAX_VALUATORS on one device. + + Some keyboards (?) advertise more than MAX_VALUATORS axes. Parts of the + internal event delivery relies on not having more than MAX_VALUATOR axes, so + let's cap it down. + If there's real devices that require more than the current 36, I'm sure we can + bump this up. + + Signed-off-by: Peter Hutterer + +commit 063833f3a6d9f8f657e3de309c8d6d5c3d606513 +Author: Julien Cristau +Date: Wed Dec 24 14:23:37 2008 +0100 + + Add XI 1.5 event and requests to protocol.txt + +commit 4f86ee61a4abf7a29e565d095aa08abd0ca9dc66 +Author: Julien Cristau +Date: Wed Dec 24 14:17:27 2008 +0100 + + Add RandR 1.3 requests to protocol.txt + +commit 6559f02ef89cb0ee365110d6a928ebd29632ad53 +Author: Colin Harrison +Date: Fri Apr 17 15:14:45 2009 +0100 + + xkb: set bell_func in InitKeyboardDeviceStruct. + + Signed-off-by: Peter Hutterer + +commit 95628b797e2b326ae15406df4f736e25f5d2f12f +Author: Peter Hutterer +Date: Mon Apr 20 14:04:54 2009 +1000 + + Xi: use the XI2 defines for AsyncPair, SyncDevice, etc, not the XI ones. + + And don't allow some values on SDs either. + +commit 335c63fcd6a53c1d3dffd6e00c0acd400c61e3d3 +Author: Werner LEMBERG +Date: Fri Apr 17 15:22:57 2009 +0200 + + Add newline to some LogMessage strings. + + Signed-off-by: Peter Hutterer + +commit d5ad14c8ed4d8360e1df8cd0bacf6a7c9c31df91 +Merge: 54716fd 0e0642e +Author: Peter Hutterer +Date: Sun Apr 19 22:28:22 2009 +1000 + + Merge branch 'master' into xi2 + +commit 54716fd3dbc251db9d251d1d0435942efaa63259 +Author: Peter Hutterer +Date: Sun Apr 19 22:12:11 2009 +1000 + + Convert to using int32_t fixed point values on the wire. + + Signed-off-by: Peter Hutterer + +commit 129ac9a9145323e3f126590b491e718f976f80ce +Author: Peter Hutterer +Date: Sun Apr 19 21:31:56 2009 +1000 + + dix: set root_x/y for device events. + + Signed-off-by: Peter Hutterer + +commit eeb1e4cd254c047539933c3b886037e67ef30762 +Author: Peter Hutterer +Date: Wed Apr 15 22:17:15 2009 +1000 + + Xi: add support for XIAllowEvents. + + Signed-off-by: Peter Hutterer + +commit 74d0fc3aee05bc4f505274d81dd44c0d2b4368ff +Author: Peter Hutterer +Date: Thu Apr 16 11:06:52 2009 +1000 + + dix: don't allow more than MAX_VALUATORS on one device. + + Some keyboards (?) advertise more than MAX_VALUATORS axes. Parts of the + internal event delivery relies on not having more than MAX_VALUATOR axes, so + let's cap it down. + If there's real devices that require more than the current 36, I'm sure we can + bump this up. + + Signed-off-by: Peter Hutterer + +commit edb70caf21fa5da3c336c47c62218a3f05c635b8 +Author: Peter Hutterer +Date: Wed Apr 15 22:16:49 2009 +1000 + + dix: remove un-used parameter "core" from AllowSome + + Signed-off-by: Peter Hutterer + +commit bb5418d4901017c657031181d3839f58b387a2a3 +Author: Peter Hutterer +Date: Sun Apr 12 22:22:21 2009 +1000 + + Xi: Add support for XI2 active grabs and ungrabs. + + Signed-off-by: Peter Hutterer + +commit c11ef87931f920ba782ba4e9b47d9c31ad7c1cf7 +Author: Peter Hutterer +Date: Sun Apr 12 22:16:14 2009 +1000 + + Xi: take XI2 requests into account when checking opcodes. + + Signed-off-by: Peter Hutterer + +commit c94ea5bc055e4efc323e84b7a8266e8b8a4af48e +Author: Peter Hutterer +Date: Sun Apr 12 17:38:28 2009 +1000 + + input: use a GrabMask union in GrabDevice to allow for XI2 masks. + + Signed-off-by: Peter Hutterer + +commit 6bb4b5b93701535402f65ea828348ed7747c7dbf +Author: Peter Hutterer +Date: Sun Apr 12 16:20:10 2009 +1000 + + Xi: add XI2 grab protocol request handling. + + Signed-off-by: Peter Hutterer + +commit 09f9a86077f0058ce88ee9b3df5d1ab854eeca43 +Author: Peter Hutterer +Date: Sun Apr 12 16:19:45 2009 +1000 + + input: replace GrabRec's coreGrab field with grabtype. + + Don't allow grabs of different types to override each other. + + Signed-off-by: Peter Hutterer + +commit 7fbe1b7d63c94b344e8ff9d09c0bb162ca104909 +Author: Peter Hutterer +Date: Sat Apr 11 13:10:31 2009 +1000 + + Xi: remove the GetExtensionVersion hack. + + Squashing data into a request that's defined to be padding is probably not the + best idea. + + Signed-off-by: Peter Hutterer + +commit 0e0642ee9466d3268476d0084a83a9d93a4aa555 +Author: Peter Hutterer +Date: Thu Apr 16 16:17:07 2009 +1000 + + os: don't malloc memory in LogVMessageVerb. + + LogVWrite is limited to a buffer size of 1024, so we don't loose anything here + by truncating. This way we can use LogVMessageVerb (and xf86Msg and friends) + during signal handlers with the normal message types. + + Signed-off-by: Peter Hutterer + Acked-by: Alan Coopersmith + +commit 62d2fb68638e9f2aa3c1d72027619c4d38f5b812 +Author: Peter Hutterer +Date: Thu Apr 16 17:06:33 2009 +1000 + + xkb: Add XkbFreeRMLVOSet helper function. + + Signed-off-by: Peter Hutterer + Acked-by: Dan Nicholson + +commit 73e1055b1275470dcc2d53b15898c866d9721d3e +Author: Peter Hutterer +Date: Sat Apr 11 11:19:29 2009 +1000 + + Xi: fix a typo in a #ifdef + + Signed-off-by: Peter Hutterer + +commit cbbc7fed994448852961a987a8b0eadf718b24bf +Author: Simon Thum +Date: Wed Apr 1 10:53:10 2009 +0200 + + dix: correctly utilize tracker buffer and protect from timer overruns + + two small related fixes hard to split up + + Signed-off-by: Peter Hutterer + +commit ca4918ac2ff003142aefbd8ac7ec2bc398ccc6eb +Author: Peter Hutterer +Date: Thu Apr 9 19:31:04 2009 +1000 + + Xi: remove fail from SProcXISelectEvent. + + Leftover from rebasing once too often, I guess. + + Signed-off-by: Peter Hutterer + +commit 463676f016a105a790cbb386d20eb5b188c04652 +Author: Peter Hutterer +Date: Thu Apr 9 19:11:59 2009 +1000 + + Xi: remove un-used variable pXIClient + + Signed-off-by: Peter Hutterer + +commit 32b13ec4cd42842a133aa6428f49638b0448efc9 +Author: Peter Hutterer +Date: Thu Apr 9 19:11:27 2009 +1000 + + Xi: silence compiler warnings about "wrong" event types. + + Signed-off-by: Peter Hutterer + +commit f2e957c5342aefc8329e9fe282b7072f767fcd28 +Author: Peter Hutterer +Date: Thu Apr 9 19:07:36 2009 +1000 + + mi: add prototype for CopyGetMasterEvent. + + Signed-off-by: Peter Hutterer + +commit d80798ba2ae018086aaf230fcd8bc71d3bf06b11 +Author: Peter Hutterer +Date: Thu Apr 9 20:27:00 2009 +1000 + + mi: fix wrong (*EnqueueEvent) declaration in miPointerScreenFuncRec. + + Signed-off-by: Peter Hutterer + +commit 4dedb900f4fbd3bdc9538b1e82cab2044438a770 +Author: Peter Hutterer +Date: Thu Apr 9 08:42:06 2009 +1000 + + xfree86: shut up compiler warnings - typecast to InternalEvent + + Reported-by: Eric Anholt + Signed-off-by: Peter Hutterer + +commit 4fa7fcf610d2ab91c1158f8034bbc0531060a86f +Author: Peter Hutterer +Date: Thu Apr 9 08:40:20 2009 +1000 + + mi: fix compiler warning - explicitly typecast to InternalEvent. + + Signed-off-by: Peter Hutterer + +commit 8bf2f562678107c346e8664f18c10f315c2768f5 +Author: Peter Hutterer +Date: Thu Apr 9 08:14:39 2009 +1000 + + dix: fix dev/keybd variable mixup. + + Reported-by: Eric Anhold + Signed-off-by: Peter Hutterer + +commit 012cb1956a69f6cf2a878c5331e792fb12c73e9d +Author: Peter Hutterer +Date: Thu Apr 9 08:06:32 2009 +1000 + + xfree86: fix use of uninitialized variable in DGAProcessPointerEvent. + + Reported-by: Eric Anholt + Signed-off-by: Peter Hutterer + +commit 73b01a9aac5fde46ccd90c08a070e9d2496092b9 +Author: Simon Thum +Date: Wed Apr 8 14:35:01 2009 +0200 + + dix: fix pointer accelerations remainder handling + + This didn't really work as intended, but did amazingly well thanks + to roundf() hiding the defect. Cheers! + + Signed-off-by: Peter Hutterer + +commit 4318075140cc287871d3c3b9f777289ea4ffa23a +Author: Peter Hutterer +Date: Tue Apr 7 21:51:02 2009 +1000 + + dix: store subpixel precision and send it down the wire to the client. + + For the valuator data, not yet for root x/y and event x/y. + + Signed-off-by: Peter Hutterer + +commit 9212948461c5e053abb61fff73ced4a00f138544 +Author: Simon Thum +Date: Sat Mar 21 18:19:19 2009 +0100 + + dix: allow relative motion buffer to accumulate in a natural way + + Since with XI2 we pass sub-pixel motion, anything else is broken. + + Signed-off-by: Peter Hutterer + +commit ad76656f8869e2065f0c4e66cfbeef0b42c61769 +Author: Adam Jackson +Date: Fri Apr 17 17:46:58 2009 -0400 + + randr: Accept gamma set requests from XF86VidMode clients too + +commit d9bf52b4abd29a3c206cd1e765b680659ddac1c6 +Author: David Jander +Date: Fri Apr 17 01:34:18 2009 -0400 + + [kdrive] Fix rotation of pointer + + Rotation matrix for pointer coordinates was incomplete and pointers with + absolute coordinates did not work correctly in xserver (kdrive) when the + sceen was rotated other than by 0 degrees. + + Signed-off-by: David Jander + Signed-off-by: James Cloos + +commit 4559d2ace6ac55fe361f572ded0769cdd1f3b545 +Author: Eamon Walsh +Date: Thu Apr 16 22:48:11 2009 -0400 + + security: Grant untrusted windows remove access on all windows. + + This allows untrusted clients to destroy their own windows when they + have been reparented by a trusted window manager. + +commit 3481b32ab971c41cb972f6819ae049f3e9f7033b +Author: Eamon Walsh +Date: Thu Apr 16 22:39:54 2009 -0400 + + security: Fix a crash caused by wrong ordering of format arguments. + +commit 6045506be0cebca4ebbe943ae77f020aafa703d4 +Author: Eamon Walsh +Date: Thu Apr 16 22:33:12 2009 -0400 + + security: Revert behavior of extension access for compatibility. + + Previously, three extensions were defined as "trusted" by the extension: + BIG-REQUESTS, XC-MISC, and XPrint. No other extensions were permitted + to be used by untrusted clients. + + In commit 8b5d21cc1d1f4e9d20e5d5eca44cb1e60a419763 this was changed for + some reason. Return to the old, compatible behavior. + +commit 56a5955c8cd87137248edb2cbc65d384376d72ad +Author: Peter Hutterer +Date: Tue Apr 14 17:05:04 2009 +1000 + + xkb: strdup the values returned by XkbGetRulesDflts + + XkbGetRulesDftls may get a copy of what will later be freed when passed into + XkbSetRulesDftls. + + On the second run of XkbGet/SetRulesDflts: + XkbGetRulesDflts(rmlvo) + rmlvo->rules = current-rules + + XkbSetRulesDflts(rmlvo) + free(current-rules) + current-rules = strdup(rmlvo->rules) + + Leaving us with garbage in current-rules. + + This patch requires callers of XkbGetRulesDflts to free the associated memory. + + See also + http://lists.freedesktop.org/archives/xorg-devel/2009-February/000305.html + + Reported-by: Benjamin Close + Signed-off-by: Peter Hutterer + Acked-by: Benjamin Close + Signed-off-by: Daniel Stone + +commit b406886bbffadaa52864a99f2a0520999eadc15d +Author: Peter Hutterer +Date: Tue Apr 14 16:57:29 2009 +1000 + + input: allow NULL as XkbRMVLOSet in InitKeyboardDeviceStruct. + + Virtually all callers use + XkbGetRulesDefault(&rmlvo); + InitKeyboardDeviceStruct(..., rmlvo); + + Let's save them the trouble and accept NULL as a hint to take the + default RMLVO. + + Signed-off-by: Peter Hutterer + Acked-by: Benjamin Close + Signed-off-by: Daniel Stone + +commit 4e4e263bc073bf452f19c932b937c4881ae71f64 +Author: Peter Hutterer +Date: Wed Apr 15 22:16:49 2009 +1000 + + dix: remove un-used parameter "core" from AllowSome + + Signed-off-by: Peter Hutterer + +commit dd6e8a14ec1c8f4ed9c51ca2764261e6e48d13b3 +Author: Peter Hutterer +Date: Sat Apr 11 11:19:29 2009 +1000 + + Xi: fix a typo in a #ifdef + + Signed-off-by: Peter Hutterer + +commit de1e43181bd670877b994db221ad8a04b5d63324 +Author: Ian Romanick +Date: Wed Apr 15 11:13:48 2009 -0700 + + DRI2: Don't leave empty entries in private->buffers + + This should fix bug #21130. + + Signed-off-by: Ian Romanick + +commit b3e3154cce47add97f5561088036ce3b9e7dc937 +Author: Robert Noland +Date: Wed Apr 15 12:06:19 2009 -0500 + + One = is more than adequate here. Make is sh safe. + +commit 74d27c8b5bac7c8d2ed02ba86e09bf09924ce05c +Author: Samuel Thibault +Date: Wed Apr 15 15:44:17 2009 +0200 + + Fix build on hurd-i386 + + http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=523970 + +commit faf7dfa099f5b42a703313fbd1bf8afdad07a179 +Author: Dave Airlie +Date: Wed Apr 15 18:26:06 2009 +1000 + + randr12: looking up these bits if randr isn't initialised is bad. + + When xinerama is enabled we don't get randr protocol, but the + driver might still want randr internals + +commit efa31092d6703397121a0ada4f7205a8ecad3d3d +Author: Ander Conselvan de Oliveira +Date: Mon Apr 6 16:01:20 2009 -0300 + + xfree86: Remove device from inputInfo.devices if ActivateDevice failed. + + After the call to xf86ActivateDevice, the new device will be added to + inputInfo.devices. However, if the subsequent call to ActivateDevice + fails, the correponding InputInfoRec for the device is deleted but an + entry still remains in inputInfo.devices. This might lead to a server + crash later on (on InitAndStartDevices for instance) when the device + control proc would be called for an invalid device. + + Signed-off-by: Peter Hutterer + +commit d79bad0aa70403ead8ec87bac8463a6e2005802c +Author: Peter Hutterer +Date: Mon Apr 13 17:49:00 2009 +1000 + + xfree86: don't synthesise a mouse section if synaptics devices are found. + + Signed-off-by: Peter Hutterer + +commit 4a27618565f3151ab17f0ca9ecbde12fa7ba13c0 +Author: Kristian Høgsberg +Date: Tue Apr 14 16:21:04 2009 -0400 + + composite: Fix resource lookups + +commit e7785e8af3e34f9d1089d8499d16802984ab9823 +Author: Kristian Høgsberg +Date: Tue Apr 14 14:24:31 2009 -0400 + + xfixes: Fix a couple of resource lookups + +commit 0eb19f9437b7d8c19592e49eedb028771d300d80 +Author: Adam Jackson +Date: Tue Apr 14 10:54:25 2009 -0400 + + xdmcp: Don't crash on X -query with more than 255 IP addresses. (#20675) + + You could be more clever than this, but the wire protocol says this + really is an array of not more than 255 ARRAY8, so it's not just a + matter of changing the types. + +commit f0543ae4ec0fcb5d696e7b2983653bd779f1eddc +Author: Eric Anholt +Date: Wed Apr 8 14:17:41 2009 -0700 + + Fix unused var warning from pci cleanups. + +commit 181cc08c8908a119fc403f970dea8cc98d3e0b9b +Author: Eric Anholt +Date: Wed Apr 8 14:17:40 2009 -0700 + + Add shave so that we can see the steaming piles of warnings generated. + + The old style output can be reenabled for build system debugging using + "make V=1", or --disable-shave at configure time. + +commit 4474c200a102feda72f9572a96cb588009aa0147 +Author: Eric Anholt +Date: Wed Apr 8 14:17:39 2009 -0700 + + Move VENDOR_* defines from AC_SUBST to a header to avoid angering shave. + + This is more sane anyway, as it ensures a rebuild when changing them. + +commit e72c85547b405fbd0117dc1236e5ca5a2126063c +Author: Eric Anholt +Date: Wed Apr 8 14:17:38 2009 -0700 + + Move contributed m4 to a subdir so we can more easily update contributions. + +commit bef1cfb2395df47458159a0b7ae27b9db15025ef +Author: Peter Hutterer +Date: Tue Apr 14 16:54:42 2009 +1000 + + xnest: remove unused variable 'names'. + + Signed-off-by: Peter Hutterer + +commit 7b6400a1b8d2f228fcbedf17c30a7e3924e4dd2a +Author: Kristian Høgsberg +Date: Thu Apr 9 13:16:37 2009 -0400 + + glx: Fix drawable private leak on destroy + + When a drawable goes away, we don't destroy the GLX drawable in full, + since it may be current for a context. This means that when the drawable + is destroyed in full later, the backend doesn't get a chance to + destroy resources associated with the drawable (the DRI2Drawable). + + With this patch, we destroy the GLX drawable in full when it goes away + and then track down all contexts that reference it and NULL their + pointers. + +commit 140463a197fb93d0a4bfad924efc35b860e8cc54 +Author: Jeremy Huddleston +Date: Sat Apr 11 16:53:59 2009 -0700 + + XQuartz: Use correct values for ProximityIn and ProximityOut + + This was the other underlying cause of teh 100% CPU tablet issue. + (cherry picked from commit a9cecf34c23fbcd59b56b380c51d31a9fabc3eb7) + +commit 80a7bb2605f9b439d6221c0495a629a39177a018 +Author: Jeremy Huddleston +Date: Sat Apr 11 17:26:07 2009 -0700 + + XQuartz: Solve the tablet 100% CPU bug + + This happened because we put a byte in the fd to wake up dispatch, but we never actually enqueued anything in mieq because the num_events was 0. + (cherry picked from commit c21ca7558d2faf93c61f5feaafd7c878e9e21942) + +commit 6f8f7c78f1b722bc70a0ea8f6340116a1e09e858 +Author: Jeremy Huddleston +Date: Sat Apr 11 16:13:42 2009 -0700 + + XQuartz: Re-enable background window checking + + since that code was not the culprit for the wacom tablet, background 100% CPU bug + (cherry picked from commit fc1dc5d71b2a488a8a94d953dd8e67353161a590) + +commit 8522a759c9e78478bb399b91d3a0af2d23ea6766 +Author: Jeremy Huddleston +Date: Sat Apr 11 16:12:23 2009 -0700 + + XQuartz: Only set MotionNotify on activation if it is updated. + (cherry picked from commit ae8077a251ef27381a755d57ff974767bda16148) + +commit 5e55becddccc376ab7338789562ad6d6cd750de9 +Author: Jeremy Huddleston +Date: Sat Apr 11 14:23:41 2009 -0700 + + XQuartz: Make sure the Fn doesn't trigger unneccessary calls to DarwinUpdateModKeys() + (cherry picked from commit 70a18558c6b0a02b633fd8974f002cdf3cdc713e) + +commit bd1d9179094657865b0606ed0ac835a8b8df3be1 +Author: Jeremy Huddleston +Date: Sat Apr 11 13:53:38 2009 -0700 + + XQuartz: Re-enable Fn as an option for 3button mouse simulation. + + Patch from Martin Otte + (cherry picked from commit b5ec3be6b5449b5d575bc1472fdd1c9cb15cb8be) + +commit 1b5758bef0840c6614244e321790231b3c9477c9 +Author: Maarten Maathuis +Date: Thu Apr 9 15:45:57 2009 +0200 + + exa: implement UTS based upload through CopyArea + + - Some image viewers (eog, gqview) trigger the CopyArea path of Xext/shm.c + - I'm not aware of any code path that wouldn't like UTS and trigger this code. + - miDoCopy should handle src coordinate clipping. + - Overlapping blits are obviously not an issue (both would have to be offscreen or not). + +commit 567cf67959b30432ae30f4851ec17b3a375ab838 +Author: Ian Romanick +Date: Thu Apr 9 14:38:24 2009 -0700 + + DRI2: Synchronize the contents of the real and fake front-buffers + + Signed-off-by: Ian Romanick + +commit f1a995d1496d73741731e32f475097c44a8da972 +Author: Ian Romanick +Date: Thu Apr 9 14:31:01 2009 -0700 + + DRI2: Do not send the real front buffer of a window to the client + + Signed-off-by: Ian Romanick + +commit aa2928325fe51d94a636dde9c090e8f54a311a12 +Author: Ian Romanick +Date: Wed Apr 8 15:44:34 2009 -0700 + + DRI2: Add fake front-buffer to request list for windows + + If a front-buffer is requested for a window, add the fake front-buffer + to the list of requested buffers. + + Signed-off-by: Ian Romanick + +commit 808fd2c67f303cb721769375b11ce8b90ffc1909 +Author: Jeremy Huddleston +Date: Thu Apr 9 20:00:37 2009 -0700 + + XQuartz: xprSetWindowLevel updated to store the level requested by the WM + (cherry picked from commit c28c2ddc3a8f3c5b9beec396953bb3ac9ee4714b) + +commit dfb0d7aefbbdfc0db966e3a84d52f638135d9138 +Author: Jeremy Huddleston +Date: Thu Apr 9 18:51:22 2009 -0700 + + XQuartz: Update window levels when changing rootless state + (cherry picked from commit 1359ded5bfc14a80fb998b01a54ecacb96c4ff88) + +commit d5ef88d7543ed787093099ab18db766c446b47d1 +Author: Jeremy Huddleston +Date: Thu Apr 9 18:27:22 2009 -0700 + + XQuartz: Fix window levels for rooted mode to allow showing the menu bar. + (cherry picked from commit 80759a4186bf0335edc85aecea2faf11fe09f491) + +commit 15146b863759640e7a73fab2301fb28ef4dead84 +Author: Jeremy Huddleston +Date: Thu Apr 9 17:47:41 2009 -0700 + + XQuartz: Properly set the window level for the root window + (cherry picked from commit bdf9286d1cbfeaaf8eaf03d28091e91ee587ee25) + +commit 8d5dcfe2154f217bd8fde5509d78f3383add8725 +Author: Jeremy Huddleston +Date: Thu Apr 9 04:36:26 2009 -0700 + + XQuartz: Properly set the menu bar and hotkey state when changing rootless mode. + + Currently no code path exhibits the broken behavior since we only toggle into rootless if we don't have the root. + (cherry picked from commit 970f100ca3c5fc0662ae7658d49d118fbd9de943) + +commit 82d7cf5cdcbc5e451a87f8c2f64cd5d2e3627d54 +Author: Jeremy Huddleston +Date: Thu Apr 9 03:55:13 2009 -0700 + + XQuartz: In rooted mode, make sure we start in the hidden state. + (cherry picked from commit 5ecc497f71c2133f773f6c56ad76cb778862ddd6) + +commit e86f4e93020d56385418850a9eebae8076dcb9ac +Author: Jeremy Huddleston +Date: Mon Apr 6 21:34:14 2009 -0700 + + XQuartz: Send MotionNotify before button presses when X11 is in the background + (cherry picked from commit c80d0ec18ef5b842447d31360406d0b5b9424222) + +commit 5ccfad8df099e8ebc4bf2dd53c3db1460903b135 +Author: Jeremy Huddleston +Date: Mon Apr 6 19:21:46 2009 -0700 + + XQuartz: Revert most of the previous override redirect patch + + The changes actually caused all windows to move to the current space. Instead, we're going with a fix entirely within Xplugin that depends on quartz-wm being the window-manager for now. + (cherry picked from commit 997b6f3142c622541bb5bac98652abae75d1101d) + +commit ff7f019bbcbc52618cc478db7baed57aa5b7c3d3 +Author: Simon Thum +Date: Wed Apr 1 10:53:10 2009 +0200 + + dix: correctly utilize tracker buffer and protect from timer overruns + + two small related fixes hard to split up + + Signed-off-by: Peter Hutterer + +commit 06aebecb19dd9d90d73b742a09b6068b862f1d05 +Author: Simon Thum +Date: Wed Apr 8 14:35:01 2009 +0200 + + dix: fix pointer accelerations remainder handling + + This didn't really work as intended, but did amazingly well thanks + to roundf() hiding the defect. Cheers! + + Signed-off-by: Peter Hutterer + +commit 98f4179156391752e6688339487458ad7828abf4 +Author: Alan Coopersmith +Date: Thu Mar 26 23:04:24 2009 -0700 + + Use RTLD_DI_SETSIGNAL to catch runtime dynamic loader errors and clean up + + Based on fix for Sun bug 6813925: Xorg needs to catch ld.so.1 failure + so it can close down devices cleanly + + + Signed-off-by: Alan Coopersmith + +commit a0b6a363dca8ce0dc6f4eb79333e48496153cd67 +Author: Alan Coopersmith +Date: Mon Mar 9 13:22:57 2009 -0700 + + Lift fatal signal handlers from DDX'es up to a common DIX implementation + + Signed-off-by: Alan Coopersmith + +commit fcc19e673e3ef33d64916dd933853f8aa667c4d7 +Author: Peter Hutterer +Date: Thu Apr 9 20:27:00 2009 +1000 + + mi: fix wrong (*EnqueueEvent) declaration in miPointerScreenFuncRec. + + Signed-off-by: Peter Hutterer + +commit 4fee979d0632751d3d54d2115e84e9654edf0622 +Author: Peter Hutterer +Date: Thu Apr 9 19:11:27 2009 +1000 + + Xi: silence compiler warnings about "wrong" event types. + + Signed-off-by: Peter Hutterer + +commit 6b467bf879eeb77d167ef321e6dda97ca9d7010a +Author: Peter Hutterer +Date: Thu Apr 9 08:42:06 2009 +1000 + + xfree86: shut up compiler warnings - typecast to InternalEvent + + Reported-by: Eric Anholt + Signed-off-by: Peter Hutterer + +commit 8a2a184da78a3e9cbeae8290431f40d5ec7f3636 +Author: Peter Hutterer +Date: Thu Apr 9 08:06:32 2009 +1000 + + xfree86: fix use of uninitialized variable in DGAProcessPointerEvent. + + Reported-by: Eric Anholt + Signed-off-by: Peter Hutterer + +commit 011cee3103c146c8096b7098a27993f99e07a824 +Author: Peter Hutterer +Date: Thu Apr 9 08:14:39 2009 +1000 + + dix: fix dev/keybd variable mixup. + + Reported-by: Eric Anhold + Signed-off-by: Peter Hutterer + +commit 630a6e9d14ffbf036fa72f580c72c0172d7c20bd +Author: Peter Hutterer +Date: Thu Apr 9 08:40:20 2009 +1000 + + mi: fix compiler warning - explicitly typecast to InternalEvent. + + Signed-off-by: Peter Hutterer + +commit e3bb7dea06b9ab0e30d801bf6c3a59f94290aaed +Author: Peter Hutterer +Date: Thu Apr 9 19:07:36 2009 +1000 + + mi: add prototype for CopyGetMasterEvent. + + Signed-off-by: Peter Hutterer + +commit 346e71525fc545c6ca4ad79425722282d1544459 +Author: Michel Dänzer +Date: Thu Apr 9 09:36:41 2009 +0200 + + EXA: If the driver can't composite to an a8 mask, try an argb mask for glyphs. + + Signed-off-by: Michel Dänzer + +commit 682d7b55699cacbb2dbcd84a5e816bf6e2d2f02a +Author: Eamon Walsh +Date: Thu Apr 9 02:48:04 2009 -0400 + + xselinux: Don't BadAlloc in List* requests if there are no items to list. + +commit 3a0ee199dcec39596756a995996eac388acf6315 +Author: Eamon Walsh +Date: Thu Apr 9 02:26:24 2009 -0400 + + config: fix crash caused by strdup(NULL) + +commit 7b3982eb6518da33ab01c2fbf7ceb45b89f841df +Author: Michel Dänzer +Date: Thu Apr 9 08:21:09 2009 +0200 + + glx: Test the error value, not its address... + +commit 03aebed519986c4dd03e02b3b3d4af1f64595ca7 +Author: Ian Romanick +Date: Wed Apr 8 14:54:30 2009 -0700 + + Use a #define instead of a magic number + + The number of buffers is likely to change in the future, so having + this as a define is the right way to go. + + Signed-off-by: Ian Romanick + +commit 0d9f3ca7eabd4c514808114d30627f682c8bd030 +Author: Ian Romanick +Date: Wed Apr 8 14:53:46 2009 -0700 + + Allow GLX sources to build against Mesa 7.4 sources + + Signed-off-by: Ian Romanick + +commit e8b324102f6e21ae2b8292a6f50d016dd6254dd6 +Author: Eamon Walsh +Date: Wed Apr 8 15:10:16 2009 -0400 + + xselinux: Don't require incoming context strings to be null-terminated. + +commit df27b870a8db7a5153b18a556fe77efa590f9eee +Author: Kristian Høgsberg +Date: Tue Apr 7 16:28:08 2009 -0400 + + Convert remaining GLX LookupIDByType() calls + +commit 92562747a0fdbef1dbedf734cb55dd6a9e1d2994 +Author: Kristian Høgsberg +Date: Tue Apr 7 13:58:53 2009 -0400 + + Add validGlxDrawable() and use dixLookupResourceByType(). + + Fixes deprecation warnings, and fixes a couple of GLX error codes + for failing drawable lookups. + +commit f70cfc8f90091ef0f5ed0a5b2e023e7fd6369b36 +Author: Kristian Høgsberg +Date: Thu Apr 2 17:30:19 2009 -0400 + + Don't stomp on dixLookupDrawable() return value in DoCreateGLXPixmap(). + +commit 30d81ad72e870cc37754bd8c8aadf605450ec16e +Author: Kristian Høgsberg +Date: Thu Apr 2 17:24:12 2009 -0400 + + Make GLX context lookup use dixLookupResourceByType() + +commit 66539cc05d0b017b9feb4a038499907810140623 +Author: Alan Coopersmith +Date: Thu Mar 26 22:22:32 2009 -0700 + + Don't leak default font path when appending built-ins + + Signed-off-by: Alan Coopersmith + +commit 6c3b633299f12051fcf37fb8439f358de876cf03 +Merge: adf21db 0cfd481 +Author: Peter Hutterer +Date: Tue Apr 7 19:36:27 2009 +1000 + + Merge branch 'master' into xi2 + +commit 0cfd48130842685828e0662a27272a3a65facd23 +Author: Peter Hutterer +Date: Mon Mar 23 11:41:16 2009 +1000 + + dix: Dont change the keyboard mapping on non-keyboard devices. + + Signed-off-by: Peter Hutterer + +commit 69e73e5ce0e3971eceed71216d2d7e7d9cd57cc2 +Author: Adam Jackson +Date: Mon Apr 6 17:56:49 2009 -0400 + + config: Remove useless xf86conf{{m,c,re}alloc},free} macros + +commit 485946120a6522db9298111752de9be3d675e0b3 +Author: Adam Jackson +Date: Mon Apr 6 17:51:35 2009 -0400 + + config: s/xf86configStrdup/strdup/ + +commit 5f769b73441cf29c05a05c728ab81bdde1fc00b6 +Author: David Jander +Date: Mon Apr 6 13:01:08 2009 -0400 + + kdrive: Fix segfault in tslib support + +commit 843166b033dc0544b9f3cf3c91fc3ae3650bda14 +Author: Adam Jackson +Date: Mon Apr 6 11:05:17 2009 -0400 + + os: signal handlers return void. + +commit aa6fbc2a3679896181610aee7ce2844d4322a5ce +Author: Adam Jackson +Date: Mon Apr 6 10:58:47 2009 -0400 + + Remove some OS/2 leftovers. + +commit 2c03a63ff6855272d587b4972a67f64115a1ce4c +Author: Jeremy Huddleston +Date: Sun Apr 5 03:20:40 2009 -0700 + + XQuartz: Use updated Xplugin API to send overide-redirect windows to the current space when they're ordered in + + This fixes the annoying "using a menu moves me to another space" bug + (cherry picked from commit 9e6dab89e08a26d764ba1aeaeb804c3d25c667da) + +commit 67621f3b39c387b3f7bfc65781051aea51c0b83f +Author: Jeremy Huddleston +Date: Sun Apr 5 00:06:57 2009 -0700 + + XQuartz: Send a MotionNotify event for the mouse cursor when activating X11.app + + Otherwise if X11.app was activated with a mouse click, the location of the even is the last location of the cursor before X11 was deactivated + (cherry picked from commit c7457d7b31ddd2ddfd04dd6294bed4532664e1ab) + +commit e1338d27141125af61634597c5dea880692c4017 +Author: Julien Cristau +Date: Sat Apr 4 23:43:28 2009 +0200 + + xfree86: Remove unused DEFAULT_UNRESOLVED and DEFAULT_BEST_REFRESH macros + +commit e25d55ad546905c4921f0a26385bdb9f22038429 +Author: Adam Jackson +Date: Fri Apr 3 22:38:00 2009 -0400 + + Revert accidental Makefile change from previous commit + +commit 472811dc4ba741ab09db184216878700f4adf745 +Author: Adam Jackson +Date: Fri Apr 3 22:27:51 2009 -0400 + + DPMS: Re-export the various DPMS variables. + + The drivers might not need them but extmod does. Should move it to + builtin though. + +commit cbb165ab88cb0810268001e84d87671440baf837 +Author: Adam Jackson +Date: Fri Apr 3 18:34:45 2009 -0400 + + os: Remove the useless -x option + +commit 6574ab092fb80b7731e06ede653b6af92fa7684e +Author: Adam Jackson +Date: Fri Apr 3 18:03:43 2009 -0400 + + DPMS: Remove the defaultDPMS* variables + +commit c1cf36ee29e41901ecd382f0258110ebb4a0124e +Author: Adam Jackson +Date: Fri Apr 3 17:51:55 2009 -0400 + + DPMS: Simplify command line parsing + +commit 2d0a4fff7985f58c21023cc4757090ee67067593 +Author: Adam Jackson +Date: Fri Apr 3 13:44:37 2009 -0400 + + DPMS: Remove unused DPMSGet() + +commit d52fddefaeb43f15b677eefbea4a288a9948373b +Author: Adam Jackson +Date: Fri Apr 3 13:33:03 2009 -0400 + + DPMS: Align the default timeouts with the default screensaver timeout. + + On a typical LCD, a black screensaver is actually worse for power + consumption than a normal screen, because it takes more energy to turn + the crystals opaque. Also, the intermediate DPMS states are essentially + useless and most monitors alias them to the 'off' state, so we may as + well do the same. + + As a pleasant side effect, this brings the default DPMS timeouts in line + with the EnergyStar Program Requirements for Computers: + + http://www.energystar.gov/index.cfm?c=revisions.computer_spec + + which state that products must be "shipped with the display's Sleep mode + set to activate within 15 minutes of user inactivity". + +commit 38576dc3ae8ddec2baedd2f7e64abdfc56bedf93 +Author: Adam Jackson +Date: Fri Apr 3 11:55:41 2009 -0400 + + DPMS: Simplify some macro silliness. + +commit 1c01127a214731a21c4784249a429c74f17bd424 +Author: Adam Jackson +Date: Fri Apr 3 11:39:18 2009 -0400 + + DPMS: Unexport all the various setup variables. + + The drivers don't need to know any of this. + +commit a99b0ab918998c9a1b9f8692591b8cb59a12eb9e +Author: Adam Jackson +Date: Fri Apr 3 11:32:00 2009 -0400 + + DPMS: Fix cast abuse + +commit 4da347adb1fcc39e8f4fb2cadd98e887d7ac877a +Author: Adam Jackson +Date: Fri Apr 3 11:18:21 2009 -0400 + + DPMS: Code motion. + +commit 1731882341c191b4ffd78e0c1fd9297c636e0401 +Author: Havoc Pennington +Date: Fri Mar 27 12:30:37 2009 -0400 + + Set bg pixmap of composite overlay window to None (#20912) + + Otherwise it's impossible to get the COW without a white + flash on the screen, because it's on top, mapped immediately, + and unaffected by composite redirection. This makes + initial login ugly when it doesn't need to be. + +commit d2690375dfd994817f004cda133ca2a492c0b956 +Author: Adam Jackson +Date: Fri Apr 3 11:06:50 2009 -0400 + + kdrive: s/KdSaveString/strdup/g + + This isn't even funny anymore. + +commit 91b697efdefba125348dbcaf584ee51a7f8c9bf6 +Author: Kristian Høgsberg +Date: Wed Apr 1 17:42:33 2009 -0400 + + Support setTexBuffer2 in AIGLX. + + Fixes broken GLX_tfp, specifically, lets compositors ignore un-defined + alpha channel for pixmaps. + +commit 1c101d75d4855b2698e3fc8d2dd662f20585812f +Author: Alan Coopersmith +Date: Thu Mar 26 22:25:08 2009 -0700 + + Don't leak canonical module name and patterns if module is built-in + + Signed-off-by: Alan Coopersmith + Acked-by: Peter Hutterer + +commit 9180081ec32b6c655d193de4c89b59d222a72579 +Author: Jeremy Huddleston +Date: Wed Apr 1 13:45:02 2009 -0700 + + XQuartz: Still send mouse events while X11 is in the background if we have test extensions enabled + (cherry picked from commit f393fb686cfbedb11a1539d4eaf5c491af8b3a64) + +commit 029307e79a14cad5bde8f1f4a5d0a2f740271f06 +Author: Jeremy Huddleston +Date: Wed Apr 1 13:07:15 2009 -0700 + + XQuartz: Return BadRequest when SendPSN isn't implemented rather than success + (cherry picked from commit 49871e99179d5176a1ec516c8f449e2dc219faea) + +commit 0603fbc1d3b2db5c25faabbfb5e04914110b4d0e +Author: Jeremy Huddleston +Date: Wed Apr 1 12:46:51 2009 -0700 + + XQuartz: Pad xAppleDRINotifyEvent to 32bytes to match sizeof(xEvent) + (cherry picked from commit 276ee3bb087c772f57a987519aa4d94afeceadaa) + +commit b1dab580bdfb4acfe3feddeda6e760098ec4922a +Author: Tormod Volden +Date: Wed Apr 1 19:32:21 2009 +0200 + + xfree86: edid quirk for Philips LCD LP154W01-TLAJ + + This panel reports its vertical size in cm. + + X.Org bug#21000 + + Signed-off-by: Tormod Volden + Signed-off-by: Julien Cristau + +commit 7d94414cf4b16501ce4c4291846498f393f150a4 +Author: Adam Jackson +Date: Wed Apr 1 09:42:06 2009 -0400 + + input: Remove xf86ReloadInputDevs hack + + We have input hotplug now, no need to fake it. + +commit 3311ef6ec93df4325d198715e0fce73af19f0781 +Author: Jeremy Huddleston +Date: Tue Mar 31 18:11:19 2009 -0700 + + XQuartz: Don't report mouse events while X11 is not the foreground application + (cherry picked from commit 5d1dd35096334b59564d77caef7a97bd58845c5d) + +commit 89887fadc97f6fd5059886396a6d7cf61de1982f +Author: Adam Jackson +Date: Tue Mar 31 15:09:57 2009 -0400 + + APM: Don't carp if not available. + +commit 2e145989d10ef454581f8aa60358ed3782fd6442 +Author: Adam Jackson +Date: Mon Mar 30 17:22:28 2009 -0400 + + dix: Simplify InitClient() + +commit 252ec504817e05b185e4896a2d899e9c00b8aeef +Author: Adam Jackson +Date: Mon Mar 30 15:18:30 2009 -0400 + + Document which bits of ClientRec are currently unused + +commit 0b6e14d46feaf8c46fd3768ee5353b844391b79b +Author: Jeremy Huddleston +Date: Mon Mar 30 20:37:25 2009 -0700 + + XQuartz: Fix mouse tracking for quake, et. al. in wine + + Patch courtesy of Codeweavers + + Fix mouse movement tracking. For a non-window-related mouse-move event, + calculate the new position by adding the event's delta-x and delta-y values + to the previous mouse position. Do not rely on the current mouse position + because it may have been changed by a XWarpPointer call. + (cherry picked from commit 7a67935b05a475215b9bdbb959e4f7e15f32416f) + +commit c74220b853cd99d03e33cc62d32431ca3a879483 +Author: Adam Jackson +Date: Mon Mar 30 14:56:15 2009 -0400 + + APM: Use general handlers, not input handlers. + + Otherwise APM events get treated as input events, which messes up idle + time accounting and screensavers and such. Not, we hope, that anyone + is using APM anymore. + +commit 283a081572d8db787c77d09e5ba6bcadebf4f7fe +Author: Adam Jackson +Date: Fri Mar 27 15:56:15 2009 -0400 + + selinux: Only activate if policy says to be an object manager + +commit 8e7facfe3013abda12a0c39ad8b4d025618077a7 +Author: Adam Jackson +Date: Fri Mar 27 15:48:17 2009 -0400 + + config: Fall back to vesa for Intel Poulsbo. + +commit 17239c8e3163fef0159835d43c0dde20e78192a1 +Author: Adam Jackson +Date: Fri Mar 27 12:44:20 2009 -0400 + + misprite: Remove private header, fold into misprite.c + +commit bb89fc2771d2e54d7e0cd0371db68ddb754a64e5 +Author: Adam Jackson +Date: Fri Mar 27 12:36:42 2009 -0400 + + misprite: Fix software cursor on multiple ScreenRecs + +commit 2c000f4980d30015e6da95de3c77c0fdda0090e4 +Author: Adam Jackson +Date: Fri Mar 27 11:54:52 2009 -0400 + + misprite: Do window check first for Get{Image,Spans} too + +commit 8beced6c02610444105f3d0b8ee3a49e1072096d +Author: Adam Jackson +Date: Fri Mar 27 11:43:33 2009 -0400 + + misprite: RECT_IN_REGION -> miRectIn + +commit a232116b9ecd41ce55f8eeee9dfe2b67d76479d6 +Author: Adam Jackson +Date: Fri Mar 27 11:41:40 2009 -0400 + + misprite: Fix a typo + +commit 1f0eaf886d88f3935e43985d2c889b374f412e43 +Author: Adam Jackson +Date: Fri Mar 27 11:29:40 2009 -0400 + + misprite: Fix cast abuse + +commit e3c1096273446bb7eb993b179ceb989617d149cc +Author: Peter Hutterer +Date: Mon Mar 23 13:20:22 2009 +1000 + + dix: build_modmap_from_modkeymap needs to bounds-check its argument. + + Signed-off-by: Peter Hutterer + Signed-off-by: Daniel Stone + +commit 45baef0b77f6f0e5e6230eca045dfcbb1e48c647 +Author: Peter Hutterer +Date: Mon Mar 23 13:32:32 2009 +1000 + + dix: return BadValue as error in SetModifierMapping. + + Signed-off-by: Peter Hutterer + Signed-off-by: Daniel Stone + +commit 603db34337a61754e0c5f71525011d10eab78411 +Author: Peter Hutterer +Date: Wed Mar 25 15:51:43 2009 +1000 + + Xext: set POINTER_SCREEN flag in XTestFakeInput if necessary. (RH #490984) + + The POINTER_SCREEN flag must be set explicitly for XTest core events to avoid + out-of-range events when the lastSlave was an SD with an explicit axis range. + Device events sent through XTest don't need this flag, they are expected to be + in the valuator range of the device anyway. + + Red Hat Bug 490984 + + Signed-off-by: Peter Hutterer + +commit 4ab93f05664890ff6738ef1089bcd956ce3f06e9 +Author: Adam Jackson +Date: Mon Mar 23 15:19:47 2009 -0400 + + misprite: In SourceValidate, check that it's a Window first + + It almost never is. However, you have _lots_ of input devices now, and + walking them all on every Composite operation is not the cheapest thing + in the world. + +commit d698e62690e0a52fa537f1cd792cab7a5b1bafdf +Author: Maarten Maathuis +Date: Thu Mar 19 21:35:25 2009 +0100 + + xkb: plug a memory leak in XkbCopySrvLedInfo (#20756) + + X.Org Bug 20756 + + Signed-off-by: Peter Hutterer + +commit e26f79335bb7a9ad34fe9b23c283a214202f89ed +Author: Peter Hutterer +Date: Tue Mar 17 16:22:00 2009 +1000 + + xkb: put a few extra checks in against non-keyboards + +commit 844bb629fe786ecf536039df3f5f4b3b214c9d40 +Author: Peter Hutterer +Date: Thu Mar 19 15:44:34 2009 +1000 + + dix: remove a truly useless a = (foo) ? bar : bar; statement. + + Signed-off-by: Peter Hutterer + +commit 62183a06e5f72a18abc3d58c4bf5658a20ece217 +Author: Tomas Carnecky +Date: Fri Mar 20 16:19:33 2009 +0100 + + Remove two unused defines in C files + + These two defines were defined in C files but not used anywhere: + + dix/window.c #define DeviceEventMasks (KeyPressMask | [...] + os/connection.c #define MAXFD 500 + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit 4e0d7cc506b661b5e5cedf8be6e17aa4c8075aa8 +Author: Tomas Carnecky +Date: Fri Mar 20 16:19:34 2009 +0100 + + FID, whatever that was, isn't anymore + + No traces of FID in the xserver nor in the modules listed in + util/modular/xorg.modules + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit adf21dba7617542f08309415e315d4b2699c10e0 +Author: Peter Hutterer +Date: Thu Mar 19 20:40:43 2009 +1000 + + include: un-export a bunch of server-only functions. + + Signed-off-by: Peter Hutterer + +commit 111ef10375204a6981a5be5c045860854c4da4e6 +Author: Peter Hutterer +Date: Thu Mar 19 09:33:02 2009 +1000 + + dix: move ProcGrabPointer guts into GrabDevice. + + Yes, this means we have even more arguments to GrabDevice. But it beats having + a copy of most but not all of GrabDevice in ProcGrabPointer. + Also, reshuffle the order of parameters, the CARD* status is a return value + and should be last. + + Signed-off-by: Peter Hutterer + +commit dc153271b6458cacd63a4bc0208594011c6460b0 +Author: Peter Hutterer +Date: Wed Mar 18 15:13:00 2009 +1000 + + Xi: purge old device enter/leave masks. + +commit 15a969c0fcd4f19f3c560a7037c1a6a58bce51f0 +Author: Peter Hutterer +Date: Wed Mar 18 10:21:59 2009 +1000 + + dix: remove now obsolete mskidx parameter from DeliverEventsToWindow. + + mskidx would always be dev->id anyway, so if we're already passing in the + device, mskidx is superfluous. + +commit f5409aa026b2cb501170867d04c6e220f3fe0a44 +Author: Peter Hutterer +Date: Fri Mar 13 15:51:50 2009 +1000 + + mi: remove deprecated miPointerAbsoluteCursor + + Functions that don't specify the device are so last year. + +commit 46145a9312eaf5dcd0e8b6788ed8219e086f790e +Author: Peter Hutterer +Date: Fri Mar 13 15:12:55 2009 +1000 + + dix: remove coreMods field from GrabRec. + + Nobody uses it anyway, and it's taking up a whole bit! + +commit 40e2a61e37109cd9c2ab57954ee40dcf4a062e31 +Author: Peter Hutterer +Date: Thu Mar 12 16:53:56 2009 +1000 + + Add Get/SetDeviceFocus handling. + +commit d9b7343eac1157490442cc0fb3b5b6d7d6e9a705 +Author: Peter Hutterer +Date: Thu Mar 12 12:56:18 2009 +1000 + + dix: even if we don't get a XI1 event, continue processing. + +commit e11dc10f01603b181e224c90d1dd4abe940f2ce7 +Author: Peter Hutterer +Date: Thu Mar 12 10:21:00 2009 +1000 + + dix: Send HierarchyEvents when devices are added/removed/enabled/disabled. + +commit 8634e1bcbae9317fc22132f7c9bde1a4a881b9a7 +Author: Peter Hutterer +Date: Thu Mar 12 10:42:05 2009 +1000 + + dix: send presence events when floating and unfloating SDs + +commit c9483a53bea1f721d2902ac11467f67317e7de14 +Author: Peter Hutterer +Date: Wed Mar 11 16:20:22 2009 +1000 + + include: add a few prototypes to silence compiler warnings. + +commit a4b61cfc78d81de02a71fa6ad8a44a06616f6794 +Author: Peter Hutterer +Date: Wed Mar 11 16:08:24 2009 +1000 + + dix: un-static FixUpEventFromWindow and deal with focus events too. + +commit 4cc6a96d7171e567a9bd5a10f552bb953077aafb +Author: Peter Hutterer +Date: Tue Mar 10 16:08:14 2009 +1000 + + input: add support for RawDeviceEvents. + +commit a668d91e28d5a3042a8ce0d087474883b046869a +Author: Peter Hutterer +Date: Mon Mar 9 16:36:07 2009 +1000 + + dix: store the xi2mask on grabs and pass it around as needed. + + This enables passive and implicit passive grabs for XI2 events, except that we + don't have the protocol spec yet to request them. + +commit 32f338263ff7de1a2e76d570c98f5be979c18d4e +Author: Peter Hutterer +Date: Sun Mar 8 21:32:31 2009 +1000 + + Xi: Deliver XI2 HierarchyEvents when the hierarchy changes. + + Signed-off-by: Peter Hutterer + +commit 97e89a59572a4be6757510a317c142ec1d82e8f8 +Author: Peter Hutterer +Date: Fri Mar 6 23:22:00 2009 +1000 + + Allow XI2 event selection for AllDevices and AllMasterDevices. + +commit 808a158182b0a3b985385899252c190947b4dbbd +Author: Peter Hutterer +Date: Fri Mar 6 14:45:26 2009 +1000 + + dix: Enable XI2 delivery for events and focused events. + + No support for grabbed events yet. + +commit 87ff1159b497c66554a4036ae619f92d5631f00f +Author: Peter Hutterer +Date: Wed Mar 11 09:01:23 2009 +1000 + + dix: deliver device enter/leave events. + +commit 04ed0bcb2530866f7248b412974ecd15f0fbf6ac +Author: Peter Hutterer +Date: Mon Feb 23 17:07:55 2009 +1000 + + xkb: remove Device/Enter leave handling - XI2 enter/leave don't have compat state. + +commit 3f37923a727d5fc4b1880a53dac1145884ba7140 +Author: Peter Hutterer +Date: Mon Feb 23 16:12:25 2009 +1000 + + Xi: send XI2 focus events. + +commit daf7dd3bfecc9029bdd10c7a4ea5d2c875a89c41 +Author: Peter Hutterer +Date: Wed Mar 11 09:01:12 2009 +1000 + + dix: Add GetWindowXI2Mask helper + +commit 8b6a370058ad5a20e0a0e49ec9443daf03775de8 +Author: Peter Hutterer +Date: Mon Feb 23 15:58:07 2009 +1000 + + Add XI2 masks and XISelectEvent() request handling. + + XI2 event masks are simply stored in the OtherEventMasks as a separate field. + This replaces the XiSelectEvent code. + +commit 38bba0c1b75b84e8bbdfa7975cf701a9414a3afd +Author: Peter Hutterer +Date: Fri Feb 20 16:07:33 2009 +1000 + + Xi: Change ChangeMasterDeviceClasses to new XI2 events. + + Split ChangeMasterDeviceClasses into an extra XISendDeviceChangedEvent that + assembles the XI2 wire event for the DeviceChanged event. Re-use this when + detaching the last SD. + + Not quite perfect yet, we still copy the device classes from the slave now + rather than from the data we had when the event occured. But it's a start. + + (We can now unexport SizeDeviceInfo and CopySwapDevices, not needed anymore) + + Signed-off-by: Peter Hutterer + +commit 181e41511d5772962b8bc998a1940d2a9ba62d5e +Author: Peter Hutterer +Date: Mon Feb 23 16:14:51 2009 +1000 + + dix: fix up device enter/leave for XI_Enter/XI_Leave. + +commit 0befeb36c175b14eac8d5416851f1556af90dd86 +Author: Peter Hutterer +Date: Tue Feb 24 15:27:07 2009 +1000 + + dix: Add device info to DeviceChangedEvent, and fill in CCCE. + + We need to fill the info here, as the device may change until we get a chance + to process it. + + Signed-off-by: Peter Hutterer + +commit b2ba77bac4fc3edab39372dfdc529e745ae74f77 +Author: Peter Hutterer +Date: Thu Mar 5 14:23:47 2009 +1000 + + dix: add EventToXI2 and GetXI2Type. + + Signed-off-by: Peter Hutterer + +commit dfa2e8e966b699f9d292628119d5a9cbd2dbdf5a +Author: Peter Hutterer +Date: Mon Feb 23 15:12:32 2009 +1000 + + Xi: add XIQueryDevice request handling. + +commit 560c58b53e4e60ebd26e0c66dd00399c438bd619 +Author: Peter Hutterer +Date: Fri Feb 20 15:17:52 2009 +1000 + + Xi: name-space and fix XI2 requests + + All XI2 requests have XI prefixes. + Requests affected: + ChangeDeviceCursor + QueryDevicePointer + WarpDevicePointer + SetClientPointer + GetClientPointer + XIChangeDeviceHierarchy. + +commit 903449615b1a1188955c0287a552aec8ac5b3fbb +Author: Peter Hutterer +Date: Fri Feb 20 15:03:35 2009 +1000 + + Xi: add XIQueryVersion request handling. + +commit 77d51b94bd264be5a8625a2bdd98f2bfb2aa33d3 +Author: Peter Hutterer +Date: Sun Mar 8 11:47:26 2009 +1000 + + dix: FixUpEventForWindow needs to handle XI2 events. + +commit 95ed2ab715fa881d36d508adf4ae5dadd92ae7b2 +Author: Peter Hutterer +Date: Thu Mar 12 11:46:07 2009 +1000 + + Add XI2_EVENT check. + +commit 92e0a324b5c738166ed4046cb00fe85ab8aea5a0 +Author: Peter Hutterer +Date: Thu Mar 12 14:15:02 2009 +1000 + + require inputproto 1.9.99.7 + +commit 906fe9547957917ac463d646935473ceca10a72b +Author: Peter Hutterer +Date: Fri Mar 20 09:36:09 2009 +1000 + + Xi: don't allow OpenDevice on any MD. + + This restores the original behaviour of XI 1. Any slave device may be opened, + but none of the MDs. + + Signed-off-by: Peter Hutterer + +commit c7eb27f3927ba9a36f9904938996f346cca33021 +Author: Peter Hutterer +Date: Wed Mar 4 18:45:09 2009 +1000 + + Xi: always only list VCP/VCK and the SD's in XListInputDevices. + + If a client wants other MDs, it should use XI2's XQueryDevice() instead. + + Signed-off-by: Peter Hutterer + +commit 801f14d4b742e506201b619d472087a7e82bd018 +Author: Peter Hutterer +Date: Tue Mar 3 17:04:08 2009 +1000 + + Xi: don't send attachment info down with ListInputDevices. + +commit 66dd8879b3527f30309fb747cad4a161d0709d91 +Author: Peter Hutterer +Date: Fri Feb 20 15:01:05 2009 +1000 + + Xi: purge ExtendedGrabDevice request handling. + +commit 43e48900fccd245e98d6d697e4cdbb6fa866a0dd +Author: Peter Hutterer +Date: Fri Mar 6 22:27:54 2009 +1000 + + dix: Force deviced ids of 2 and higher for actual devices. + + 0 is now reserved for the "AllDevices" virtual device. + 1 is now reserved for the "AllMasterDevices" virtual device. + + This also means that wherever we passed in (mskidx = 0), we now need to pass + in the deviceid. + +commit 70896f6127e2c12c13479a027fdd6e67853d1f8b +Author: Peter Hutterer +Date: Fri Feb 13 16:42:02 2009 +1000 + + dix: abstract event filters through GetEventFilters. + + Don't let everyone acces the filters[] array directly. This is necessary once + we start dealing with GenericEvents, where the filters are a bit more + complicated. + + Signed-off-by: Peter Hutterer + +commit 8a8763f1502c9ddd1961f96f2cc7c73fbc0b49b5 +Author: Peter Hutterer +Date: Thu Feb 12 16:07:15 2009 +1000 + + dix: restructure DeliverEventsToWindow, a little bit anyway. + + Desperate attempt of making the code easy to read and understand. A failed + desparate attempt, possibly. + +commit 891cfc802028c3e008ee6c2d3e2438a31f57f39f +Author: Peter Hutterer +Date: Thu Feb 12 10:12:45 2009 +1000 + + dix: clean up DeliverDeviceEvents. + + Move mask checking into external function and streamline the code. + + Signed-off-by: Peter Hutterer + +commit 6b6b660c3633af75a94302725137549df0dbe446 +Author: Peter Hutterer +Date: Tue Mar 10 10:15:45 2009 +1000 + + Xext: remove ev_fill from GEExtensions. + +commit 445daa62e720fe58fa2d6c3cdc26663671b2bc48 +Author: Peter Hutterer +Date: Thu Jan 15 17:16:22 2009 +1000 + + Xext: purge XGE event masks. + + The masks were originally designed to generically handle event masks for + extensions. Since all that is in-server anyway, it's much better writing + custom event masks for those extensions that need it and not providing a + unified mechanism. + XI2 needs more than the current implementation, which is already too complex + for most other extensions. good riddance. + + Signed-off-by: Peter Hutterer + +commit 6fe4c28bf4f3a9f7c341b9403213470e9ded5cb3 +Author: Peter Hutterer +Date: Mon Feb 16 16:34:05 2009 +1000 + + Add agressive event type checking. + + Best to FatalError if a wrong event comes in. At least that forces me to fix + it really quickly. + + Signed-off-by: Peter Hutterer + +commit 836864b65794dc0954a01245e418e714cead8125 +Author: Peter Hutterer +Date: Fri Mar 13 12:43:17 2009 +1000 + + xkb: don't overrun the map index when accessing symbols. + + Signed-off-by: Peter Hutterer + +commit 497a12533905b98f388775b6ba49adf21017cc75 +Author: Peter Hutterer +Date: Fri Mar 13 11:52:33 2009 +1000 + + dix: remove duplicate PickKeyboard() command. + + Signed-off-by: Peter Hutterer + +commit 7d4df0ee4414779ad5e519e7bd297c5dfec409a3 +Author: Peter Hutterer +Date: Thu Mar 19 20:24:06 2009 +1000 + + dix: fix device sync state when calling SyncBoth during AllowEvents. + + This did access the wrong device's sync state, potentially freezing or not + thawing the actual device that was supposed to be thawed. + + Signed-off-by: Peter Hutterer + +commit f1c7b95d83948160a0d5796ef6c16fc0d1bf5c5d +Author: Peter Hutterer +Date: Fri Mar 20 13:55:00 2009 +1000 + + dix: do percentage check before device check in ProcBell + + This is just for correctness. The server should return BadValue for anything + not in [-100, 100]. + + Signed-off-by: Peter Hutterer + +commit d60391d8ca9918d2089c23c0baef5c91177325f0 +Author: Peter Hutterer +Date: Fri Mar 20 14:01:30 2009 +1000 + + dix: fix uncredible fail in PostSyntheticMotion. + + Signed-off-by: Peter Hutterer + +commit 1a71862d333282e2d251ff0036866cec22bcce85 +Author: Simon Thum +Date: Sat Feb 28 22:17:47 2009 +0100 + + dix/xfree86: simplified velocity approximation algorithm + + Replace multi-stage filtering with simple linear velocity, + tracked several instances backwards. A heuristic ensures + only approximately linear motion is considered, so velocity + remains valid in any case. Numerical stability is much + better, and nothing changes to people who didn't tune the + advanced features of the previous algorithm. + + Signed-off-by: Peter Hutterer + +commit 5ae129baef85b47590c02e4cf61b23904d8f7aa9 +Author: Peter Hutterer +Date: Tue Mar 17 15:47:57 2009 +1000 + + Xi: check for existence of the button class before accessing it + + Signed-off-by: Peter Hutterer + +commit e8094d8f3f69f45850af031efd79a3928e584638 +Author: Peter Hutterer +Date: Thu Mar 19 10:09:04 2009 +1000 + + dix: ProcSendEvent shouldn't use inputInfo.keyboard directly. + + Signed-off-by: Peter Hutterer + +commit 12aeddf5ad41902a180f8108623f356642b3e911 +Author: Maarten Maathuis +Date: Thu Mar 19 19:07:59 2009 +0100 + + exa: Accept scratch pixmaps with offscreen memory as such. + +commit a635bb5357648779815ec6729d55f6b478493213 +Author: Maarten Maathuis +Date: Tue Mar 17 19:34:45 2009 +0100 + + exa: round of fb_pitch to the next byte + +commit ec328783593339e33772301a8c5fea35fd63c0a2 +Author: Maarten Maathuis +Date: Tue Mar 17 10:45:21 2009 +0100 + + exa: check if the pixmap is pinned for unsupported AUX indices. + +commit 3992dd38caf33b343affd8d732c94880d1099dcf +Author: Adam Jackson +Date: Mon Mar 16 13:24:48 2009 -0400 + + selinux: Add support for avc_acquire_netlink_fd() + + Requires libselinux 2.0.79 or newer. Without this, libselinux will + check for policy updates on the netlink socket on basically every policy + lookup. Statistically speaking, they never happen, and the check + translates to at least one more syscall on basically every operation. + + Instead, take control of the fd from the library, and check it in + WakeupHandler if it polls readable. + +commit be6dc9023b5fb3995a9ce56d607627e247918aef +Author: Peter Hutterer +Date: Fri Mar 13 16:29:36 2009 +1000 + + xfree86: remove a superfluous assignment. + + If dev is NULL, we don't have a VCP and that means we have other issues + anyway. + +commit bfab422dcdbb9c009b2a91fe0dba288fac9bc859 +Author: Peter Hutterer +Date: Fri Mar 13 16:27:49 2009 +1000 + + xfixes: useless (void) typecast removal + + Signed-off-by: Peter Hutterer + +commit 4eeaee1e5a05ea973cbeb653abf99f2a352edc70 +Author: Peter Hutterer +Date: Fri Mar 13 16:25:12 2009 +1000 + + xkb: xkbi has a pointer to the device - use this instead of inputInfo.pointer. + +commit 3f801ba62ac493d4aeb664a5ae638e3fb0c9a262 +Author: Peter Hutterer +Date: Fri Mar 13 16:23:50 2009 +1000 + + xkb: the VCP can post device events, don't stop xkb filtering on it. + +commit 23686e5680d565c36e0b6320ce62c38c7830d06b +Author: Peter Hutterer +Date: Fri Mar 13 16:20:56 2009 +1000 + + xkb: Fix a mis-use of inputInfo.keyboard. + +commit c97c6c3de337f247406eaffd5845a847630ee6c1 +Author: Peter Hutterer +Date: Fri Mar 13 16:17:20 2009 +1000 + + xkb: fix a couple of device checks when looping through all devices. + + Generally, we want to apply stuff to the device and to all attached slave + devices. + +commit d21a546e9604ce2d3c914900f6e6ff2df9f5ca8b +Author: Peter Hutterer +Date: Fri Mar 13 16:02:22 2009 +1000 + + Xi: VCP and VCK may register for ext. events, so delete them if necessary. + +commit 6ec49098f7ef259b43870139511006627b308bb6 +Author: Peter Hutterer +Date: Fri Mar 13 15:57:27 2009 +1000 + + render: replace a wrong inputInfo.pointer with pDev + + Signed-off-by: Peter Hutterer + +commit 0b2b330f6fbf8633ae682a17fbd95f6b5db2c112 +Author: Peter Hutterer +Date: Fri Mar 13 15:54:27 2009 +1000 + + mi: remove a bunch of useless inputInfo.pointer assignments. + + Signed-off-by: Peter Hutterer + +commit 31301861c1dd87a2f55c5c4300ac7c8f54154bf8 +Author: Peter Hutterer +Date: Fri Mar 13 15:49:22 2009 +1000 + + dix: remove wrong InputInfo.pointer usage - should be pDev instead. + +commit 639658b2eab480a5bfb943dff51e4bdfc1e16e3b +Author: Peter Hutterer +Date: Fri Mar 13 15:34:29 2009 +1000 + + dix: use GetPairedDevice in ProcQueryPointer rather than inputInfo.keyboard. + +commit f01ee198ff0cbd7ca418217a84248d1c6131a0c6 +Author: Peter Hutterer +Date: Fri Mar 13 15:22:23 2009 +1000 + + dix: don't use inputInfo.keyboard to get the focus window in ActivateKbdGrab + + I'm not sure if that's the right solution, but the other one is wronger. + +commit de415743f4c1878df2de79f84f5fe087a88d5241 +Author: Peter Hutterer +Date: Fri Mar 13 15:19:23 2009 +1000 + + dix: fix XACE checks in ProcWarpPointer + +commit 18430616cfd5d0bc5a3efe61fcc2982b1726b0de +Author: Peter Hutterer +Date: Fri Mar 13 14:59:18 2009 +1000 + + dix: use PickPointer() and PickKeyboard instead of inputInfo.pointer/keyboard. + + Affected calls: + ProcChangeKeyboardControl + ProcBell + ProcUngrabKey + ProcUngrabButton + +commit e31727158cda7729283233cfdc9bc5bcb59a52c1 +Author: Peter Hutterer +Date: Fri Mar 13 14:51:55 2009 +1000 + + dix: fix ProcChangePointerControl's wrong inputInfo.pointer usage. + + Signed-off-by: Peter Hutterer + +commit fc5e8dc4a24c2b5bed4e6b5d00670c4401ec92d9 +Author: Adam Jackson +Date: Fri Mar 13 14:00:37 2009 -0400 + + Unexport xf86SetPriority + +commit 89d7b88f32f9227fa72a043675367b94c4c9283c +Author: Maarten Maathuis +Date: Fri Mar 13 00:40:43 2009 +0100 + + exa: allow exaModifyPixmapHeader to set sys_ptr for EXA_HANDLES_PIXMAPS + + - exaModifyPixmapHeader would save sys_ptr if needed, but it would be NULL'ed afterwards. + - This is needed to support pixmaps that are not offscreen. + +commit 544cd9e7b50cd9905afc52404de1a5a2bcff91b5 +Author: Ander Conselvan de Oliveira +Date: Thu Mar 5 12:57:28 2009 -0500 + + Honour Option "DPMS" "off" on xorg.conf + + Signed-off-by: James Cloos + +commit 7abb72c5c711a26a27a82604bd2398bcb6df0a99 +Author: Simon Thum +Date: Wed Feb 25 10:55:36 2009 +0100 + + xfree86: remove obsolete code + + remove a few lines which redo part of the pointer acceleration + init. Properties is the way to go for them. + + Signed-off-by: Peter Hutterer + +commit 97cb3cc28e4a5694d17f156276efd84555a96d18 +Author: Peter Hutterer +Date: Thu Mar 12 10:17:16 2009 +1000 + + dix: Merge DevicePresence notify events generation into a single function. + + Signed-off-by: Peter Hutterer + +commit 7f1ba804a1c1aef725e20a04e5f6ef74478ecf44 +Author: Peter Hutterer +Date: Tue Mar 10 11:55:04 2009 +1000 + + dix: s/numEvents/num_events/ in GetKeyboardValuatorEvents + + GPE and GProxE use the same. + + Signed-off-by: Peter Hutterer + +commit 067f26f7ba4c8cf37af7ddf0eb204b7e1d06096d +Author: Peter Hutterer +Date: Fri Mar 6 19:42:27 2009 +1000 + + Xi: set the modifier + group state during POE. + + Signed-off-by: Peter Hutterer + +commit da724e8f1d1867c36c9550a153c58829939800fc +Author: Peter Hutterer +Date: Fri Mar 6 15:50:05 2009 +1000 + + Xi: Set the button state on the event during POE. + + Signed-off-by: Peter Hutterer + +commit b368dd1720c708aaec492b4a67fb7e4f87569c11 +Author: Peter Hutterer +Date: Thu Mar 5 15:22:41 2009 +1000 + + dix: remove a few pointless (void) casts of return values. + + Signed-off-by: Peter Hutterer + +commit cadf65a6e190a8952ad3cc216dc9ea55241de91a +Author: Matthias Hopf +Date: Tue Mar 10 15:49:49 2009 +0100 + + randr: Nuke broken set_origin shortcut + + Shortcut is impossible to implement this way, because we don't know for sure + whether the crtc of an output has changed or not. + +commit 6544490700051b3b5e88ac1890d71b35634c9100 +Author: Eamon Walsh +Date: Mon Mar 9 17:28:40 2009 -0400 + + Correct access mode in call to dixLookupWindow() within RRSelectInput. + + Reported by Alan Coopersmith. + +commit f8dd80d13bb5313a11b38b280f8ad3e22f0a6300 +Author: Keith Packard +Date: Wed Aug 6 15:26:24 2008 -0700 + + Replace dixLookupResource by dixLookupResourceBy{Type,Class} + + dixLookupResource attempted to automatically detect whether the caller + wanted a lookup by-type or by-class, unfortunately, it guessed wrong for + RT_NONE. Instead of trying to make the guess better, this patch just reverts + the unification and creates separate functions for each operation. + +commit 0d9a42dc0380d1583889b6b6521bd5a2451735d4 +Author: Peter Hutterer +Date: Mon Mar 9 08:59:01 2009 +1000 + + Xi: don't crash on a NULL property name, just return None. + + strcmp doesn't like NULL-pointers. + + Signed-off-by: Peter Hutterer + +commit 432a95032eb10f82fb59caa29e52d3f56dfb363c +Author: Peter Hutterer +Date: Sun Mar 8 21:34:02 2009 +1000 + + Remove useless (void) typecasts. + + Signed-off-by: Peter Hutterer + +commit f508446a2bcd887df9cf5db786b15fc3cbc435a0 +Author: Peter Åstrand +Date: Fri Mar 6 19:41:35 2009 -0800 + + Composite: Uninitialized borderClipX/borderClipY + + Valgrind warned me about two other uninitialized variables, which are not + padding. Valgrind output at the end. I'm a bit unsure of what these should + be initialized to, is zero fine? + +commit d1338a94805cc774fe0a5d00c2225a3ee9673a9f +Author: Winfried Grünewald +Date: Fri Mar 6 12:59:35 2009 -0500 + + [hw/xfree86] Fix StaticGray cmap. + + Fix this bug report: + + ,----< from http://bugzilla.freedesktop.org/show_bug.cgi?id=20504 > + | Using the Visual StaticGray (8 bit depth) is missing one gray level. + | The gray level of index zero and index one are the same and all + | other levels are shifted by one. The max level (255) cannot be used. + `---- + + Signed-off-by: James Cloos + +commit eba3bab71ff26d58bb0e49e4a9726fc21ff07258 +Author: Alan Coopersmith +Date: Thu Mar 5 15:09:28 2009 -0800 + + Remove #ifdef macII code left over from ancient A/UX 3.0 support + + Signed-off-by: Alan Coopersmith + +commit daa9727a02701cfac7b92f3e61d752be6fcc005e +Author: Jeremy Huddleston +Date: Wed Mar 4 11:33:00 2009 -0800 + + XQuartz: Implement SendPSN for AppleWM + (cherry picked from commit 779e9fdc2de9bfebc5f0a4c9d5f42e389fdd8593) + +commit ce6e1771be5f2c21af6f72a9705795df26210413 +Author: Maarten Maathuis +Date: Mon Mar 2 17:05:28 2009 +0100 + + exa: fix a serious issue in exaChangeWindowAttributes (and some more related things) + + - fbChangeWindowAttributes can create pixmaps (and access them) without use preparing access. + - Also handle the destroyed pixmaps by finishing them first. + - Switch to DEST indices again in exaCreatePixmapWithPrepare, because they are obviously being rendered to. + - Also avoid calling FinishAccess on pixmaps that are destroyed (and their memory potentially invalid). + +commit 7fb68e8b31d259325ce311ad67befc43a933f009 +Author: Maarten Maathuis +Date: Sat Feb 28 22:29:42 2009 +0100 + + exa: remove a few pExaPixmap checks. + + - And make some fatal for a debug build. + +commit bd2f35ef0d4b8c56eb205b8694b4fe5d2fa2c049 +Author: Maarten Maathuis +Date: Sat Feb 28 22:15:04 2009 +0100 + + exa: fix unwrapping of ModifyPixmapHeader upon CloseScreen. + + - Cleanup wrapping too. + +commit da8ea41a542787691ea1120e5c8c7dc3182cbea5 +Author: Maarten Maathuis +Date: Sat Feb 28 21:59:09 2009 +0100 + + exa: increase/rework safety checks in Prepare/FinishAccess. + +commit 3ea3d505e8128ab3e878edcef697fd5656b0c917 +Author: Maarten Maathuis +Date: Sun Mar 1 01:15:28 2009 +0100 + + exa: avoid a potential Prepare/FinishAccess inbalance. + +commit 10334cf7e64e2e633cd507f2aa216027e840c8e5 +Author: Maarten Maathuis +Date: Sat Feb 28 21:46:00 2009 +0100 + + exa: simplify exaPixmapIsOffscreen + + - This includes properly handling the framebuffer. + +commit 57bed76e218471b60d340b18c346d492f37798c2 +Author: Tomas Carnecky +Date: Wed Mar 4 10:29:34 2009 -0500 + + Remove dead EXTENSION_PROC_ARGS define + + Signed-off-by: Tomas Carnecky + +commit a6fb9f4ccfc69dc090b152e29142e3159f812166 +Author: Tomas Carnecky +Date: Wed Mar 4 10:27:30 2009 -0500 + + Remove dead XTest code from extmod + + Signed-off-by: Tomas Carnecky + +commit c4081fee0995dae691665a00ea9bfdb8b8043b8c +Author: Topi Kanerva +Date: Wed Mar 4 10:23:10 2009 -0500 + + Make DGA optional. + +commit c7ebb4bef152b47880abb6ea4819416e89e6522c +Author: Eamon Walsh +Date: Tue Mar 3 14:02:36 2009 -0500 + + Fix 2 const warnings. + +commit c7cf926d25523d5d0d8e5c83671c901660070d71 +Author: Eric Paris +Date: Tue Mar 3 13:15:39 2009 -0500 + + This patch changes all places in the X code to use _raw functions. The + X server should never see, translate, or deal with a munged context. + Display managers which show contexts to the user should take care of + translating these to human readable form. + +commit d84a2f855da9fd293e984084168c1e27f5a5115c +Author: Peter Hutterer +Date: Sun Mar 1 16:59:14 2009 +1000 + + xfree86: Add linebreak to two debug statements. + + Signed-off-by: Peter Hutterer + +commit e0ed9f16d0ab871f3209837f54055446503c4108 +Author: Jason Vas Dias +Date: Tue Mar 3 13:24:54 2009 +1000 + + Fix build with --enable-debug. + + [amended by Peter Hutterer] + Signed-off-by: Peter Hutterer + +commit 6093d3eb1d7d1e9cbacc2c2f4d376352f787dd53 +Author: Peter Hutterer +Date: Thu Feb 26 15:38:45 2009 +1000 + + dix: fix two compiler warnings (old-style function definition). + + Signed-off-by: Peter Hutterer + +commit b874a5c0aafc766891c9e567af48b891ea14be4f +Author: Peter Hutterer +Date: Thu Feb 26 14:40:22 2009 +1000 + + mi: only print the "EQ overflowing" error once. + + Signed-off-by: Peter Hutterer + +commit a706dd8771f87cf0a73578755552124d162baa5d +Author: Peter Hutterer +Date: Thu Feb 26 10:04:14 2009 +1000 + + include: add a range of button labels. + + Mostly the same buttons as defined by linux/input.h, with five exceptions: + "Button Unknown" for a button that cannot be labelled. + "Button Wheel Up", "Button Wheel Down" for buttons 4/5, traditionally the + wheel buttons. + "Button Horiz Wheel Up", "Button Horiz Wheel Down" for buttons 6/7, + traditionally the horiz. wheel buttons. + + Signed-off-by: Peter Hutterer + Acked-by: Daniel Stone + +commit 8a6ed44a8b2fc5f14729dc54fec17607ced03859 +Author: Adam Jackson +Date: Mon Mar 2 14:26:23 2009 -0500 + + randr: Fix thinko in xf86TargetPreferred + + The only-one-output case would only work right if that also happened to + be the zeroth output. Oops. + +commit 7bfad37f7706da5a76174c0da78dce4d6c6a585b +Author: Maarten Maathuis +Date: Fri Feb 27 22:26:52 2009 +0100 + + exa: fixup aux indices and ensure that the indices are used as they should be. + + - In a previous patch i forgot to add a FALSE somewhere it seems. + - Rename AUX indices so the driver (think of driver managed pixmaps) can do optimisations based upon them. + - Fix one abuse of DEST index now that we have the AUX indices (same reason as above). + +commit 8c6b4a827624aa70aea94b449fee6d5df8504105 +Author: Adam Jackson +Date: Fri Feb 27 12:42:43 2009 -0500 + + DDC: Redo extended device probe slightly. + + We'll now only mention the E-EDID segment register if the device is + actually E-EDID-capable. While we're here, check for DDC/CI and + standard EEPROM support too. + +commit b030f858f2f1ce1fd27a73ebf7f9ec5db541a668 +Author: Adam Jackson +Date: Sat Feb 21 19:56:20 2009 -0500 + + selinux: Don't bother relabeling resources that are being destroyed + + Makes window destroy about 40x faster in Xvfb. + +commit f028b14876dc536b575d4b6e1df7f37ee525acec +Author: Maarten Maathuis +Date: Fri Feb 27 13:06:28 2009 +0100 + + exa: whitespace + +commit ed00515ec54b294f304b8ae2857e3073f97860d2 +Author: Maarten Maathuis +Date: Thu Feb 26 22:18:03 2009 +0100 + + exa: fixup exaAssertNotDirty. + + - Do the right thing based on prepare access. + +commit 2e88b6004f09dbcb888abf05d4d5554231b2bd1a +Author: Maarten Maathuis +Date: Thu Feb 26 21:14:23 2009 +0100 + + exa: minor glyphs cleanup. + + - This should fix subtle coordinate bugs and make the code a bit cleaner to read. + +commit 265d20068af5434489752b6dba0bf0065b3cc3ec +Author: Michel Dänzer +Date: Fri Feb 27 16:41:39 2009 +0100 + + EXA: Fix check for whether the glyph we're evicting from the cache is in use. + + Since commit f07f18231a921d3ae9dd9b75881c9e58e9e2e235 ('EXA: Allow using + exaCompositeRects also when we can't use a mask in exaGlyphs.') we were + checking the wrong set of coordinates in the buffer where glyphs to be rendered + are accumulated when no mask is used in exaGlyphs. + + This fixes occasional glyph corruption which can be corrected with redraws, in + particular with Qt4. + + Thanks to Maarten Maathuis for asking the right question: 'where do we protect + against evicting glyphs that are still needed?' + + Signed-off-by: Michel Dänzer + +commit 9d5141f7bcfed36417873b47f850332506af83ac +Author: Michel Dänzer +Date: Fri Feb 27 16:37:28 2009 +0100 + + EXA: No longer use the driver UploadToScratch hook. + + See http://bugs.freedesktop.org/show_bug.cgi?id=20213 . + + Signed-off-by: Michel Dänzer + +commit 4cfb36f6ad2df01215028fec48d99239a0e4496b +Author: Michel Dänzer +Date: Fri Feb 27 16:37:28 2009 +0100 + + EXA: Handle separate alpha maps properly in Composite fallback, take two. + + Preserve the EXA ABI by introducing a new driver flag EXA_SUPPORTS_PREPARE_AUX. + If the driver doesn't set this flag, we have to assume any Prepare/FinishAccess + driver hooks can't handle the EXA_PREPARE_AUX* indices, so we move out such + pixmaps at PrepareAccess time. + + Fixes https://bugs.freedesktop.org/show_bug.cgi?id=18710 . + + Signed-off-by: Michel Dänzer + +commit 4bf707f01822abe99286909fd561da7e7a4211d6 +Author: Michel Dänzer +Date: Fri Feb 27 16:37:27 2009 +0100 + + Revert "Fix up Xephyr build for recent EXA changes." + + This reverts commit 4e8d98b61e1f763c187e7994c683b543cca1a33c. + +commit de022f8e634baf5a7acd186934b370647df58355 +Author: Michel Dänzer +Date: Fri Feb 27 16:37:27 2009 +0100 + + Revert "EXA: Handle separate alpha maps properly in Composite fallback." + + This reverts commit 170cf1270dff38d3cce7f5ba5b940d1c0d70eff5. + + Conflicts: + + exa/exa_render.c + +commit d0dd649035fc3698c5b436f9d9d248116aa106a3 +Author: Jon TURNEY +Date: Thu Feb 12 16:03:16 2009 +0000 + + Remove references to rgb.txt from files section of Xserver and Xorg man pages + + The references to this file in the server code were removed in commit dda10c9066a660b647384179f82e1da8e063264f + + Signed-off-by: Jon TURNEY + +commit b0ad9e1ced9619f37acf77764c395c57b86cf463 +Author: Jon TURNEY +Date: Thu Feb 12 16:02:17 2009 +0000 + + Remove long-gone '-co' option from Xserver man page + + The '-co' option was removed in commit dda10c9066a660b647384179f82e1da8e063264f + + Signed-off-by: Jon TURNEY + +commit 7907b37d96dc81f333351a6fedd6a9f4e31ae66a +Author: Jeremy Huddleston +Date: Thu Feb 26 21:22:48 2009 -0800 + + XQuartz: Don't erase the mode_switch+keysym if it's the same as shift+keysym + (cherry picked from commit 3e2427e6e9675dd655c8ad328b82e12191b064f2) + +commit f07f18231a921d3ae9dd9b75881c9e58e9e2e235 +Author: Michel Dänzer +Date: Thu Feb 26 10:35:44 2009 +0100 + + EXA: Allow using exaCompositeRects also when we can't use a mask in exaGlyphs. + + This should give the full benefits of the glyph cache even when we can't use a + mask. + + This also means we no longer need to scan the glyphs to see if they overlap, + we can just use a mask or not as the client asks. + + Signed-off-by: Michel Dänzer + +commit b26c136ee9bf7015c583136af53d0c9e9da67ace +Author: Michel Dänzer +Date: Thu Feb 26 10:35:44 2009 +0100 + + EXA: Stop tracking damage for pixmaps subject to ModifyPixmapHeader. + + Signed-off-by: Michel Dänzer + +commit 812e8d681a49b33d8748983843799ef692b8b148 +Author: Keith Packard +Date: Wed Feb 25 11:08:01 2009 -0800 + + Oops. 62fc98c had crtc offsets applied backwards (fix from server-1.6-branch) + + Crtc offsets are in screen space, not crtc space, and hence should be + applied relative to the screen->crtc transform, not the crtc->transform. + + This fix was found in the 'cherry pick' of 62fc98c on server-1.6-branch, + clearly some new definition of 'cherry pick' that I am unaware of. + + Signed-off-by: Keith Packard + +commit c090f5514d28e1602a6ebbe7c909e98a0e3374d7 +Author: Keith Packard +Date: Wed Feb 25 10:14:36 2009 -0800 + + Pre-clip panning coordinates to keep crtc within panning region + + There is a separate panning region check, but that doesn't work under + transformation, so just pre-clip the mouse coordinates when computing the + panning offsets. This leaves the case where panning constants are changing + unresolved. + + Signed-off-by: Keith Packard + +commit 77c7a64e8885696665556c9fbcb3cffb552e367a +Author: Keith Packard +Date: Wed Feb 25 10:12:23 2009 -0800 + + RandR rotations and reflections offset by one pixel + + The matrix computation for rotation and reflection resulted in dropping a + row or column of pixels as the offsets used in the matrix computations used + width and height rather than width-1 and height-1. + + Signed-off-by: Keith Packard + +commit 4e8d98b61e1f763c187e7994c683b543cca1a33c +Author: Michel Dänzer +Date: Wed Feb 25 10:26:04 2009 +0100 + + Fix up Xephyr build for recent EXA changes. + + Signed-off-by: Michel Dänzer + +commit d6976e1436c035b8e4a6d1e10971b083b1b4184b +Author: Jeremy Huddleston +Date: Tue Feb 24 23:12:57 2009 -0800 + + XQuartz: Re-enable support for capslock + (cherry picked from commit 4901b8147e593d26d7a31a9b73a201254b948916) + +commit 4b63f88b0ad5f6614fee5b7af4f97ce4581ab935 +Author: Jeremy Huddleston +Date: Tue Feb 24 18:54:16 2009 -0800 + + XQuartz: Comment out the background pointer interaction that seems to be causing CPU spinning on some configurations + (cherry picked from commit 1d479fa4f3bca9cbd81808a6c87582422fd2a257) + +commit 5672e2d66444be74125c71b81373b82f8e1b7dd2 +Author: Peter Hutterer +Date: Fri Feb 20 14:55:51 2009 +1000 + + Xi: remove AllExtensionVersions. + + Keeping an AllExtensionVersions array to save all versions of + all extension is rather pointless if only one extension uses it. + Rename to XIVersion, reduce to a single struct. + + Signed-off-by: Peter Hutterer + +commit 59bc615ed2e60c595887dcc11fe04935b49bcaff +Author: Peter Hutterer +Date: Tue Feb 24 15:26:20 2009 +1000 + + dix: set the valuator mode in set_valuators + + We only have per-device mode for now anyway. + + Signed-off-by: Peter Hutterer + +commit cffa3c499a0495e132557bd6b08a8c0b858cdf3b +Author: Peter Hutterer +Date: Tue Feb 24 15:25:49 2009 +1000 + + dix: remove unused variable 'tmp' + + Signed-off-by: Peter Hutterer + +commit 8d6dd83a38020080808705a156d32ec2a135633c +Author: Peter Hutterer +Date: Fri Feb 20 13:13:58 2009 +1000 + + include: fix indentation for lastSlave/master. + + Yeah! + + Signed-off-by: Peter Hutterer + +commit b3412adca8b2ba3fff348bceb4f01b9e968c0afc +Author: Peter Hutterer +Date: Sun Feb 22 20:53:46 2009 +1000 + + Xi: don't need to set the XKB settings for new core devices. + + This is done by the XKB code these days anyway, so we might as well ignore it + and keep using the stanard stuff. + + Signed-off-by: Peter Hutterer + +commit a180c2367974abe2ccdc88bb5c6f138c7df4d2e9 +Author: Peter Hutterer +Date: Mon Feb 23 16:38:41 2009 +1000 + + dix: remove ConvertBackToXI. + + This was a temporary solution for the internal-events rework and is not needed + anymore. + + Signed-off-by: Peter Hutterer + +commit 26cf709e712573548990d165e71137f28271a8bb +Author: Peter Hutterer +Date: Mon Feb 23 22:39:08 2009 +1000 + + Doxygenify events.h and eventconvert.c + + Should have done that before pushing, but oh well. + + Signed-off-by: Peter Hutterer + +commit 170cf1270dff38d3cce7f5ba5b940d1c0d70eff5 +Author: Michel Dänzer +Date: Tue Feb 24 09:22:09 2009 +0100 + + EXA: Handle separate alpha maps properly in Composite fallback. + + Fixes https://bugs.freedesktop.org/show_bug.cgi?id=18710 . + + As this can't work without new EXA_PREPARE_AUX* indices, this requires a major + version bump, so we can also drop the UploadToScratch driver hook and + ExaOffscreenSwap*(). So this also fixes + http://bugs.freedesktop.org/show_bug.cgi?id=20213 . + + Moreover, introduce EXA_DRIVER_KNOWN_MAJOR to break compilation of drivers + which may not be able to handle EXA_PREPARE_AUX*, giving instructions how to + make them build again in the #error message. + + Signed-off-by: Michel Dänzer + +commit d3b355875ac46104a174966e1974e6af99e40fd6 +Author: Peter Hutterer +Date: Tue Feb 24 11:14:14 2009 +1000 + + dix: fix wrong condition for setting valuators on the event. + + Signed-off-by: Peter Hutterer + +commit ef320bdd5ec3419abba77041d3a4d96a3ff87563 +Author: Michel Dänzer +Date: Mon Feb 23 16:49:51 2009 +0100 + + DRI1: Make DRICreateDrawable return TRUE for pixmaps. + + GLX_EXT_texture_from_pixmap was broken since commit + a26c77ff432d2e85a2665fc36fca25143460c476 ('glx: fix retval checks when failures + occur for drawable creation.') + + Signed-off-by: Michel Dänzer + +commit 1df6716281579e2937743d840ab1079343c503ac +Author: Adam Tkac +Date: Mon Feb 23 09:28:26 2009 +1000 + + xfree86: don't try to UnInit virtual devices in DIDR. (#20087) + + DeleteInputDeviceRequest function doesn't handle "virtual" devices well. + TightVNC libvnc.so module to X (which makes bare Xorg VNC capable) uses such + kind of devices. + + Bare Xvnc (it is something like Xvfb) simply uses AddInputDevice & + RegisterDevice functions. Xvnc uses DeleteInputDeviceRequest from Xi/stubs.c + so everything works fine (now I see that DeleteInputDeviceRequest in + Xi/stubs.c should call RemoveDevice function, shouldn't it? :) ) + + Situation is quite different when you use libvnc.so module. It uses same + schema as Xvnc, so it simply calls AddInputDevice & RegisterDevice. Thus + device is created correctly. When server is terminated it calls + DeleteInputDeviceRequest (now from hw/xfree86/common/xf86Xinput.c) for each + device. Here is the difference - Xvnc calls DeleteInputDeviceRequest from + Xi/stubs.c as I wrote above. Thus Xorg gets sigsegv because "VNC" devices + don't have real input driver. + + X.Org Bug 20087 + + [This isn't really a fix (libVNC should behave correctly) but not crashing the + server sounds like an improvement.] + Signed-off-by: Peter Hutterer + +commit a50c40be1f28467bcef7dc71fab54ebdddacffdf +Author: Peter Hutterer +Date: Sat Feb 21 08:09:40 2009 +1000 + + dix: don't apply button mappings to SDs, or the device's MD. (#20122) + + Only ever change the button map on the device we actually care about, not the + attached SDs, not the current MD of the device. + + X.Org Bug 20122 + + Signed-off-by: Peter Hutterer + Acked-by: Daniel Stone + +commit 36583a49965c0bb40a84284939b1539b3cb9fc9c +Author: Peter Hutterer +Date: Tue Feb 10 12:45:49 2009 +1000 + + mi: split EQ popping and event processing into two functions. + + mieqProcessInputEvents() - pop an event off the EQ and pass it to + mieqProcessDeviceEvent() - process the event according to the MD/SD hierarchy. + + This way, we can use mieqPDE() from Xtest, xkb, and others to post an event. + + Signed-off-by: Peter Hutterer + +commit 47f136ed6fd80310f715a2555501d1b271dd084c +Author: Peter Hutterer +Date: Fri Feb 6 12:08:43 2009 +1000 + + mi: change custom handlers to internal events + + This should re-enable DGA, but XQuartz needs to be changed to internal events + too now. + + Signed-off-by: Peter Hutterer + +commit bdc262701a37a0c12ead810d63fa99a26cbb82ec +Author: Peter Hutterer +Date: Mon Feb 2 22:39:37 2009 +1000 + + dix: remove un-used getValuatorEvents and countValuatorEvents from getevents.c + + Signed-off-by: Peter Hutterer + +commit 763848d3ab84b4e33a0254120c44b5a1877c819d +Author: Peter Hutterer +Date: Mon Feb 2 17:20:17 2009 +1000 + + Input: change processing API to InternalEvents. + + Don't pass xEvent* and count through to processing, pass a single + InternalEvent. + + Custom handlers are disabled for the time being. And for extra fun, + XKB's pointer motion emulation is disabled. But stick an error in there so + that we get reminded should we forget about it. + + Signed-off-by: Peter Hutterer + +commit 0b4066c116e07918a13dc6b4159df7ac9eb92b4b +Author: Peter Hutterer +Date: Mon Feb 2 17:18:16 2009 +1000 + + xkb: _XkbFilterRedirectKey needs to pass InternalEvents down. + + Signed-off-by: Peter Hutterer + +commit 085d50360863ccc8280cd3eccea2bcb4f3dd9a14 +Author: Peter Hutterer +Date: Mon Feb 2 16:56:52 2009 +1000 + + dix: Fix PostSyntheticMotion to use a DeviceEvent for posting. + + Signed-off-by: Peter Hutterer + +commit 155986a93dc78d6aa060ca3038d5fafa3d8753c2 +Author: Peter Hutterer +Date: Mon Feb 2 16:06:15 2009 +1000 + + dix: CheckPassiveGrabsOnWindow moved to internal events. + + Signed-off-by: Peter Hutterer + +commit 1c38abd4b136301008ba77d4e68aea34508f4c22 +Author: Peter Hutterer +Date: Mon Feb 2 14:54:34 2009 +1000 + + dix: switch DeliverGrabbedEvent to internal events. + + Signed-off-by: Peter Hutterer + +commit fb858774b80eb75c2f8e81fe893bbbdd37065fbd +Author: Peter Hutterer +Date: Mon Feb 2 14:44:13 2009 +1000 + + dix: switch DeliverDeviceEvents to internal events. + + Signed-off-by: Peter Hutterer + +commit 3aa9404fcd161e94a80b057a77ef47afe428a56d +Author: Peter Hutterer +Date: Mon Feb 2 14:00:41 2009 +1000 + + dix: switch DeliverFocusedEvent to internal events. + + And because of xfree() macro hilarity, rename "pointer" to "ptr". Oh, how we + laughed. + + Signed-off-by: Peter Hutterer + +commit 8c873e7f514844d1056f2b20e653f1dd75f4c327 +Author: Peter Hutterer +Date: Sun Feb 1 20:41:16 2009 +1000 + + dix: Switch DeliverGrabbedEvents to use internal events. + + Signed-off-by: Peter Hutterer + +commit a3718536989fa7d3358e0b2d859c25fde0a2d93e +Author: Peter Hutterer +Date: Sun Feb 1 20:18:33 2009 +1000 + + dix: Deliver{Grabbed|Focused|Device}Events API changed to InternalEvents. + + With the API change, we can now purge the XI conversion from POE. + + Note: this commit breaks DGA even more. + + Signed-off-by: Peter Hutterer + +commit 8f94ec6f788565474931cc7d5e3d4672f0f31670 +Author: Peter Hutterer +Date: Sun Feb 1 09:58:15 2009 +1000 + + dix: convert passive grabs to use internal events. + + deviceGrab.sync.event is now an internal event, and CheckDeviceGrabs and + friends is changed over. + + Note that this currently breaks some frozen grabs. See towards the end of + ComputeFreezes(). + + Signed-off-by: Peter Hutterer + +commit daa3245c479b19d445a070b5b76ee005915b5335 +Author: Peter Hutterer +Date: Sat Jan 31 08:46:43 2009 +1000 + + dix: fix EnqueueEvent to work with internal events. + + Note that we're only partially switched to internal events. The event in the + devices' event queue (dev->deviceGrab.sync.event) is still an XI event. The + events in syncEvents are InternalEvents only now. + This also implies fixing CheckVirtualMotion to work with internal events. + + Signed-off-by: Peter Hutterer + +commit 75595ba4aa9c3823ffe3b3388ce088929824816f +Author: Peter Hutterer +Date: Fri Jan 30 15:14:46 2009 +1000 + + Xi: make ProcessOtherEvents more InternalEvent aware. + + Get rid of the deviceValuator processing and a few other things, but still + drop back into XI before checking device grabs or doing anything else. + + NoticeEventTime now needs to take InternalEvents, and while we're at it, + change NoticeTime from a macro to a function. + + Signed-off-by: Peter Hutterer + +commit 3a02e538dbdb3cd482e01baeaf2aba2ddb7731df +Author: Peter Hutterer +Date: Fri Jan 30 14:59:22 2009 +1000 + + dix: update CheckMotion to deal with DeviceEvents. + + Signed-off-by: Peter Hutterer + +commit 8829d966a6bacb05d322b60531c59366b58f4514 +Author: Peter Hutterer +Date: Fri Jan 30 13:33:55 2009 +1000 + + Xi: support InternalEvents in UpdateDeviceState, parts of POE and EnqueueEvent + + Note that this breaks DGA. Life is tough. + + EnqueueEvent is a somewhat half-baked solution, we immediately drop back into + XI and store them. But it should in theory work. + + Signed-off-by: Peter Hutterer + + Don't let the dcce be random data. + +commit 007e93c869325cafcc29d975b356dbb8e7cd2ac1 +Author: Peter Hutterer +Date: Fri Jan 30 14:35:22 2009 +1000 + + xkb: Switch the xkb event processing path over to InternalEvents. + + Before dropping down into the DIX, convert back into XI events. This is a + temporary solution only, until the DIX is capable of handling InternalEvents + anyway. + + Signed-off-by: Peter Hutterer + +commit 64ea6078105f73d1b727619fc123920bc7e4a06c +Author: Peter Hutterer +Date: Mon Feb 2 14:15:52 2009 +1000 + + dix: change eventconvert to always return an array of xEvents + + Just alloc the memory on demand rather than doing things with EventListPtrs + etc. + + Signed-off-by: Peter Hutterer + +commit 4026c63e4eb16481bafc1a41ad67cd2556728d40 +Author: Peter Hutterer +Date: Fri Jan 30 14:34:02 2009 +1000 + + mi: switch the EQ to contain InternalEvents only. + + This gets rid of the nevents parameter, InternalEvents are always a single + item per event. Also remove the special DeviceValuator handling in both + enqueueing and dequeueing. + + Custom callback handlers are now broken until fixed. + + For bisectability, we copy the InternalEvent back into the XI required during + POE and friends. Consider this a temporary solution. + + Note: Because of misc linker bonghits, Xvfb won't link in this revision. + + Signed-off-by: Peter Hutterer + +commit 5a827593f9517fd5593751dd8bd90c611de06c5d +Author: Peter Hutterer +Date: Thu Jan 29 15:26:57 2009 +1000 + + dix: switch event generation to InternalEvents. + + GPE, GKVE, GProxE generate InternalEvents now. + DeviceClassesChangedEvents generates an InternalEvent now, but incomplete! We + need to tack on the information about the new SD in the ClassesChanged events. + + Note: To make the progress bisectable, we drop back into XI events at the end of the + Get*Events functions. So the rest of the server still uses XI events. + + Signed-off-by: Peter Hutterer + +commit 00b03683d0e5cda40fa23b9fe6a83d7227f86f5d +Author: Peter Hutterer +Date: Fri Jan 30 14:31:08 2009 +1000 + + dix: Add temporary conversion function ConvertBackToXI. + + Until the InternalEvents are used throughout the server, we can use this one + to drop us back into XI la-la land where every event is the wire format. + + Signed-off-by: Peter Hutterer + +commit 269d4d9f2e86fde8c6a28ef0293fb0e44b577886 +Author: Peter Hutterer +Date: Sun Feb 1 09:57:27 2009 +1000 + + dix: add GetCoreType and GetXIType. + + Convert from an InternalEvent type to the matching core/XI type. Currently + only for a few events, those we actually need in the server. + + Signed-off-by: Peter Hutterer + +commit 656491921e17b2371057041f4551ad6165067551 +Author: Peter Hutterer +Date: Wed Jan 28 15:27:38 2009 +1000 + + dix: add InternalEvent -> core/xi event conversion routines. + + Signed-off-by: Peter Hutterer + +commit e7867d12541ef9683d5d7fc766e918c13a742981 +Author: Peter Hutterer +Date: Fri Jan 23 15:37:23 2009 +1100 + + include: add XInternalEvent. + + This is the event we want to feed into the EQ and process on the way through. + Only applies for input events for now. + + Signed-off-by: Peter Hutterer + +commit 10dcf3ed9c543a6811340567e586ec0e476fcf61 +Author: Jon TURNEY +Date: Fri Feb 20 15:22:35 2009 +0000 + + Cygwin/X: Fix keyboard layout mapping for Latin American keyboards + + Fix keyboard layout mapping table entry for Windows keyboard layout 0x80a + (Latin American), layout should (now) be 'latam', not 'la' which is Laos + + Signed-off-by: Jon TURNEY + +commit 9531b96470a2fd812eb3fa0d1e10419331842496 +Author: Jeremy Huddleston +Date: Sun Feb 22 00:41:52 2009 -0800 + + XQuartz: DefineInitialRootWindow is gone + +commit c7c50157b5137a147e69abb70b43a28fac4a9ffa +Author: Jeremy Huddleston +Date: Sun Feb 22 00:29:48 2009 -0800 + + XQuartz: quartzKeyboard compile fixes for recent XKB Changes + + Seems we're less divorced from XKB than we'd like for now... hopefully not having any rules installed will still let us override things the way we want to... + +commit 61abf3189f3f97ec5228d45c035dbfd0c249f844 +Author: George Staplin +Date: Sat Feb 21 22:33:10 2009 -0700 + + XQuartz: GL: Change from xalloc to xcalloc to potentially avoid + uninitialized data in the __GLXAquaContext. + (cherry picked from commit d3120241f9f48d21f9a6ecfb848434a4a0270855) + +commit 5587f9b771f5b5427a81d4d657f403667d20f310 +Author: George Staplin +Date: Sat Feb 21 22:22:52 2009 -0700 + + XQuartz: GL: Fix a bug with an uninitialized GLX data structure. + + This was causing a crash randomly, due to random memory contents. + Use xcalloc to prevent this in the future, due to future changes or mistakes. + + Set the drawableType to include GLX_PIXMAP_BIT and GLX_PBUFFER_BIT. + The new libGL supports these. + + Set the max Pbuffer width/height, based on the results of a test program. + We may someday want to revisit this depending on what users need, so that + we create a CGLContextObj, make it current, and call glGetIntegerv to + gather the information at runtime. + (cherry picked from commit c7e338330943e0d03a99328c740540d03f018d20) + +commit ab61033700b5383a7a15370dd054eaa80e72e811 +Author: Alan Coopersmith +Date: Thu Feb 19 13:57:28 2009 -0800 + + Add Extensions section to xorg.conf man page + + Extensions section was added in X11R6.8.0 and documented in the release notes: + http://www.x.org/archive/X11R6.8.0/doc/RELNOTES2.html#3 + but never made it into the man page. + + Also fix a bonus typo. + + Signed-off-by: Alan Coopersmith + +commit 372977354c3a09ca77ca13ba0aec00a61046f60d +Author: George Staplin +Date: Wed Feb 18 21:43:19 2009 -0700 + + XQuartz: Fix the new ProcAppleDRIDestroyPixmap code REQUEST_SIZE_MATCH. + + It had a copy and paste mistake that I didn't notice. :/ + It was using the CreatePixmapReq. + + Also add a missing B16 to the end of the length for the DestroyPixmapReq struct. + + Now the AppleDRIDestroyPixmap request seem to work. + (cherry picked from commit 295fe25bd8fa2d141291a9d9b6ef7b75fcccb4dd) + +commit 225853d51d1fb610261ab0c295b1b5a96ce177d5 +Author: Dan Nicholson +Date: Thu Feb 19 06:45:05 2009 -0800 + + xkb: Use cached XKB keymap when rules haven't changed + + Rather than compiling a new keymap every time InitKeyboardDeviceStruct + is called, cache the previous keymap and reuse it if the rules have not + changed. + + Signed-off-by: Dan Nicholson + Acked-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 64e595d12e05c4df56b0230cc57924b9beb274d3 +Author: Peter Hutterer +Date: Thu Feb 19 09:18:53 2009 +1000 + + dix: check if keybd->key is valid before getting the keyboard state. + + For floating SDs, GetPairedDevice() returns the SD itself, hence keybd->key + may be NULL. + + Signed-off-by: Peter Hutterer + +commit e0a451eb7cc812f11db3614b730ba0f07d6ae18c +Author: Alan Coopersmith +Date: Tue Feb 17 19:37:22 2009 -0800 + + Obsolete InputDevices keyword in xorg.conf Files section + + Was only used to provide a list of input devices that XF86-Misc could use, + now that XF86-Misc is gone, was parsed and logged, then completely ignored. + + (Depends on previous patch that introduces OBSOLETE_TOKEN in parser to + make obsolete keywords like InputDevices & RgbPath be non-fatal errors.) + + Signed-off-by: Alan Coopersmith + Acked-by: Adam Jackson + +commit d2cf562bbad553d7f09b70202134f5b6ada0114e +Author: Alan Coopersmith +Date: Tue Feb 17 18:48:52 2009 -0800 + + Make RgbPath keyword in xorg.conf a non-fatal error + + Xorg shouldn't refuse to run just because the user has an xorg.conf that + had the previously-used RgbPath keyword in it. + + Signed-off-by: Alan Coopersmith + Acked-by: Peter Hutterer + +commit 344b1f2b25d627bdf7d802c5831b6a72d22ffe34 +Author: Jeremy Huddleston +Date: Tue Feb 17 23:25:14 2009 -0800 + + configure.ac: Support version strings like W.X.Y.Z-XXXXXX + (cherry picked from commit 43967514cd57ad836d7fb85c8c9e58ada07e0232) + +commit 67faf41b3f3e0cd66e35e3c176a9adb37ff1a759 +Author: Jeremy Huddleston +Date: Tue Feb 17 23:24:25 2009 -0800 + + XQuartz: Don't need GlxSetVisualConfig any more + + Fixes build failure resulting from 516f8e2cad1311a09764e2633644188d1e3c31bb + (cherry picked from commit 066b17028a35956a089815716e38571f305469c5) + +commit b349a764e98f0d8f221190157ffa0904b91beca5 +Author: Eric Anholt +Date: Tue Feb 17 09:59:17 2009 -0800 + + xinerama: Put the proto version in the code instead using proto headers. + + Proto headers updating resulting in the server advertising new versions is + broken. This should be applied to every extension. + + This fixes the build against slightly-older xineramaproto. + +commit 5394b7e66224d20888dd4020f5cb8ca930720fb4 +Author: Keith Packard +Date: Fri Feb 6 06:34:35 2009 +0100 + + Eliminate the shadow clear on transform change + + When the crtc transformation changes, the entire crtc must be repainted. + This was being done by clearing the shadow and then painting the rectangle + containing the screen image; the clear being required as the screen image + may not fill the crtc. When changing the transform rapidly, this leads to + flashing. Eliminate the clear by painting the entire crtc instead of just + the screen rectangle. + + Signed-off-by: Keith Packard + +commit 76f18b94bd2719a8199334742d021c4d0806187d +Author: Keith Packard +Date: Mon Feb 9 13:09:11 2009 +0100 + + Add XkbDir to Files config file section + + The XKB base directory was not configuable through the config file. + + Signed-off-by: Keith Packard + +commit 123093996507c4d3b6dc457240ce00f8ac42f410 +Author: Adam Jackson +Date: Tue Feb 17 12:59:16 2009 -0500 + + RANDR: Fail softly on GetPanning if the screen can't do it. + + Just return a zeroed-out reply in that case. This is unambiguous, and + distinguishes "you didn't name a CRTC" from "you named a CRTC that can't + do panning". + +commit b2ceea3635ec05dca9d4aa2f823b96ae9fce7fe8 +Author: Maarten Maathuis +Date: Tue Feb 17 19:40:59 2009 +0100 + + Revert "exa: the extent of the valid region is probably much larger than that of the pending damage." + + This reverts commit 97c1cbc70216366e92b9371de608ce94e60aa874. + + - Sorry for the thinko, pending damage is often not fragmentated. + - Should the dst region become fragmentated, you actually want to copy more to unfragmentate it. + +commit 3175646b10c602d17d5dd37bdace7c1c7ee92b3d +Author: Maarten Maathuis +Date: Tue Feb 17 19:06:35 2009 +0100 + + exa: exaPixmapDirty should use official damage functions. + + - Otherwise other users will not be notified of damage. + +commit 97c1cbc70216366e92b9371de608ce94e60aa874 +Author: Maarten Maathuis +Date: Tue Feb 17 19:01:44 2009 +0100 + + exa: the extent of the valid region is probably much larger than that of the pending damage. + + - Since the default has been changed from nothing to everything. + +commit 4039603413f9f46d7f725463a70b4a51838e0049 +Author: Ian Romanick +Date: Tue Feb 17 08:27:32 2009 -0800 + + glx: Inialize best_score before calculating visual scores + + This bug was pointed out by Peter Åstrand. + + Signed-off-by: Ian Romanick + +commit bac13e54b164ceff7b921a2d17c4a198b42aeccf +Author: George Staplin +Date: Mon Feb 16 17:33:19 2009 -0700 + + XQuartz: Add driWrap.h that I missed in the last commit. + + This should complete the GLXPixmap commits for now. + (cherry picked from commit 7e99b55651bf7f6275bb227d02f203fb885a67a2) + +commit 95052caa2c8a79981cdc7936888a8a47e09415e0 +Author: George Staplin +Date: Mon Feb 16 17:22:18 2009 -0700 + + XQuartz: Add support for GLXPixmaps to the AppleDRI. + + This involved wrapping some GCOps to get the proper behavior + when using X11 raster ops mixed with OpenGL (see driWrap.c). + + This extends the AppleDRI protocol with create and destroy pixmap + functions. + + The dri.c code has been extended quite a bit to enable this, and + to initialize the wrapping of CreateGC for GCOps. + + This has been tested with tests/glxpixmap and proven to work with + the new libGL. Existing applications seem to work fine too. Redraws + all appear to be correct. + + There may be some bugs lurking that I haven't found yet. I plan + to drive them out by extending the libGL test suite. + (cherry picked from commit 630518766b01022c49fe3a9e7d501808f71b06e2) + +commit 970725d23e83285d41e3f7cab678a825d9cc25cf +Author: Adam Jackson +Date: Mon Feb 16 18:44:00 2009 -0500 + + kdrive: Nuke dead AGP and VGA code. (#19921) + +commit a932744d983845f81791045cf0dfb6a359790158 +Author: Chris Ball +Date: Mon Feb 16 18:27:26 2009 -0500 + + Build fix, remove export of pciBusAddrToHostAddr + + a9d7d659.. (PCI: Remove pciBusAddrToHostAddr and associated nonsense) + removes pciBusAddrToHostAddr(), but not its prototype, resulting in: + + ./.libs/libxorg.a(sdksyms.o):(.data.rel+0xe64): undefined reference to + `pciBusAddrToHostAddr' + + Signed-off-by: Chris Ball + +commit ab331aa76f0bc6a85a519b2deaafa4ce8a070c53 +Author: Julien Cristau +Date: Mon Feb 16 21:56:03 2009 +0100 + + xfree86: ModeDebug is a boolean, not a string + + Signed-off-by: Julien Cristau + +commit f5bf1fdaf36163d5c2f1b9b51df96326ebbb0e9c +Author: Peter Hutterer +Date: Fri Feb 13 09:56:22 2009 +1000 + + xkb: Fix wrong colour reference in XKB geometry copying. #20081 + + base_color and label_color need to reference the color in the destination, not + in the source. + + X.Org Bug 20081 + + Signed-off-by: Peter Hutterer + Signed-off-by: Daniel Stone + +commit a9d7d659a05e4b6b25e6acd2ce59926911aac0f7 +Author: Adam Jackson +Date: Fri Feb 13 18:36:27 2009 -0500 + + PCI: Remove pciBusAddrToHostAddr and associated nonsense + + This was all a glorified no-op. We rely on pciaccess to create device + maps anyway, so we should have no reason to care about what the host + address is. + + Acked-by: Ian Romanick + Signed-off-by: Adam Jackson + +commit be8c3d4022faa7d169778ea10d1f11e8f90e1c0c +Author: Adam Jackson +Date: Fri Feb 13 17:54:33 2009 -0500 + + Linux: Remove dead USE_DEV_FB #ifdefs + +commit 51ae4d6bbfd95cac8a9cdefcd6f22fbe21a6cb9d +Author: Ian Romanick +Date: Mon Feb 16 12:08:28 2009 -0800 + + glx: Add comments around some extension string weirdness + +commit caa5310e163b040689383f6d25ccab5ef1b2e215 +Author: Maarten Maathuis +Date: Mon Feb 16 20:41:14 2009 +0100 + + exa: reintroduce src rect optimisation, with a slightly higher threshold. + + - I got some feedback on gtkperf line test regression. + - The increased threshold should ensure the destination optimisation is dominant. + +commit 46eeaf82e228df1c5971a24dc815566516e19be9 +Author: Maarten Maathuis +Date: Mon Feb 16 17:17:14 2009 +0100 + + exa: fix performance regression from 736b6fbd2c941b6276066cd1503523edebe7bf3d + + - The src optimisation is more aggressive and possibly harmful in light of the new initial state of pixmaps. + - There is now actually a performance improvement by almost always keeping the number of rects low. + +commit 6198373ff2a8e36113bb1dcaebe975530ff01b86 +Author: David Nusinow +Date: Sun Feb 15 18:36:01 2009 -0500 + + xorg.conf (5) refer to mousedrv (4). Debian #394058 + + Signed-off-by: Julien Cristau + +commit 2a8b8077d8f6001eb57deba60e1009fc99c28668 +Author: Alan Hourihane +Date: Mon Feb 16 11:39:34 2009 +0000 + + dri2: support glXWaitGL & glXWaitX by copying fake front to front and + vice-versa. + +commit ddb8d8945d1f44d16adc366b6612eef20ae813f7 +Author: Peter Åstrand +Date: Fri Feb 13 10:23:28 2009 +0100 + + xserver: Avoid sending uninitialized padding data over the network + + Signed-off-by: Peter Hutterer + +commit b735a4b4951b607e614682836f24d5fd86c1f7fb +Author: Peter Hutterer +Date: Thu Feb 12 14:29:23 2009 +1000 + + dix: Remove lastEventMask variable, substitute with a #define. + + Signed-off-by: Peter Hutterer + +commit 772e0f9159ca3ab8fb4e03ed6feee3cc93697724 +Author: Peter Hutterer +Date: Thu Feb 12 13:38:34 2009 +1000 + + dix: Don't set core events in SetMaskForEvent. + + Rather, modify the two callers to call separately for the two different. + events. Unexport SetMaskForEvent too. + And while we're at it, get rid of the MotionFilter macro, because it's one + half confusing and one half pointless. + + Signed-off-by: Peter Hutterer + +commit 58f3127919ba5fcb3bb467b6913c28ee1127c82b +Author: Peter Hutterer +Date: Thu Feb 12 08:37:49 2009 +1000 + + Xi: remove dynamic mask assignment for XI events. + + They end up being the same anyway on startup, so let's not have a dynamic mask + assignment mechanism and instead just hardcode them already. + Also unexport SelectForWindow and remove the valid_masks parameter. We can + check that before calling, since there's only one caller anyway. + + Signed-off-by: Peter Hutterer + +commit 7057a9a97b9316b1de954d05451e04223fdd9fd8 +Author: Peter Hutterer +Date: Thu Feb 12 22:14:48 2009 +1000 + + Remove two more define XKB leftovers. + + Signed-off-by: Peter Hutterer + +commit 736b6fbd2c941b6276066cd1503523edebe7bf3d +Author: Michel Dänzer +Date: Sun Feb 15 17:44:51 2009 +0100 + + EXA: Try to prevent the valid regions from growing too many rects. + + This helps contain region management overhead in some pathological cases, see + e.g. http://bugs.freedesktop.org/show_bug.cgi?id=16647 . + + Signed-off-by: Michel Dänzer + +commit 45e14f515337159e600a47ebf349692751552af8 +Author: Chris Ball +Date: Fri Feb 13 20:32:02 2009 -0500 + + Build fix for xf86EdidModes.c + + Commit 20ac314.. forgets to declare hsize/vsize/refresh. + +commit ae68be312619ad3bcb2d550a6267d38cce3284f5 +Author: Adam Jackson +Date: Fri Feb 13 17:25:32 2009 -0500 + + Input: Don't re-block SIGIO in xf86SigioReadInput() + + It's already blocked on the way in, because we asked for it to be + blocked at sigaction() time. + +commit c18f454e4a9cab03e5d55b6d761a4a735c943b6b +Author: Aaron Plattner +Date: Thu Feb 12 18:27:48 2009 -0800 + + Query the DDX in ProcRRGetScreenInfo. + + This is the RandR 1.1 version of GetScreenResources and needs to re-query the + DDX to see if the mode pool changed. + + Fixes Launchpad bug #325115. + Signed-off-by: Adam Jackson + (cherry picked from commit 660c2a7d4c6cb52fd0992e9d2008a24805dc8c02) + +commit 20ac3140ce16af460992b83aa8aeff4afe28c2ee +Author: Adam Jackson +Date: Mon Feb 9 13:36:05 2009 -0500 + + EDID: Hack for 1366x768 in standard timing descriptors + + All you get for standard timing descriptors is horizontal size in + multiples of 8 pixels (which means you can't say 1366) and height in + terms of aspect ratio (which means you can't say 768). You'd like to + just fuzzy-match this by walking the DMT list for sufficiently close + modes, but you can't because DMT is useless and only defines a 1360x768 + mode, because it's _also_ specified in terms of character cells despite + providing pixel exact timings. Neither can you use CVT or GTF to + generate the timings, because they _also_ believe that modes have to be + a multiple of 8 pixels. + + You'd also hope you could find a timing definition for this in CEA, but + you can't because CEA only defines transmission formats that actually + exist. So there's 480p, 720p, and 1080p, but no 768p. And why would + there be, after all, the encoded signal is never 768p so obviously no + one would ever make a display in that format. + + So instead, make a CVT mode since that's likely to be handled well by + just about everything, smash the horizontal active down by 2, and shift + the sync pulse by 1. Underscanning the hard way. + + Pass the suicide. + +commit bcafdfbed6f1e1f901c2459f60f94a0da506bd90 +Author: Adam Jackson +Date: Fri Feb 13 14:06:07 2009 -0500 + + RANDR: Validate entire mode list for interlace and doublescan + + Otherwise drivers have to refuse interlace twice: once in the output + config, and once in ->valid_mode() to catch output and config modes. + If you can't do interlaced modes, asking nicely for it in the config + isn't going to suddenly make it work. + +commit a26c77ff432d2e85a2665fc36fca25143460c476 +Author: Alan Hourihane +Date: Fri Feb 13 13:45:08 2009 +0000 + + glx: fix retval checks when failures occur for drawable creation. + +commit a38ca0063c91ef221f20bf0c4ac0dc84ce07d557 +Author: Benjamin Close +Date: Fri Feb 13 14:44:53 2009 +1030 + + xext: Use proto header rather than the Xext include file, this prevents userspace being pulled in causing issues + + Signed-off-by: Benjamin Close + Acked-by: Peter Hutterer + Acked-by: Daniel Stone + +commit d62c085f36892c327785e85571bd2c7c70513ac6 +Author: Benjamin Close +Date: Fri Feb 13 14:44:09 2009 +1030 + + dix: Remove includes which are not actually used directly + + Signed-off-by: Benjamin Close + Acked-by: Peter Hutterer + Acked-by: Daniel Stone + +commit 340f1576afcdaf883d185da356e5d6282aa65e19 +Author: Peter Hutterer +Date: Thu Feb 12 10:44:27 2009 +1000 + + dix: remove DefineInitialRootWindow() + + Obsolete. + + Signed-off-by: Peter Hutterer + +commit c178c3f814f7666f43cbf47d24c7543b289f9b1a +Author: Peter Hutterer +Date: Thu Feb 12 15:29:38 2009 +1000 + + dix: doxygen-ify enterleave.c + + Signed-off-by: Peter Hutterer + +commit cc696a2b7928ec497b6a0df3602fc70dec83a629 +Author: Peter Hutterer +Date: Thu Feb 12 15:05:57 2009 +1000 + + Doxygen-ify bits of events.c + +commit 8364bf7374aa5f6d991700f0c02921dc6c638c9f +Author: Peter Hutterer +Date: Wed Feb 11 16:06:36 2009 +1000 + + Document the event masks. + + Signed-off-by: Peter Hutterer + +commit ed9d58c3c25ee1b3dedbc4c116823c263ccf164d +Author: Simon Thum +Date: Sun Feb 8 17:21:09 2009 +0100 + + dix: refactor pointer acceleration + + The algorithm is split in a 2D-specific and a general part. + This potentially allows to accelerate more than just screen motion. + A state machine is intoduced to make code more explicit and readable. + It also improves handling of 'phase 1' mickeys when axial correction + kicks in (corner case). + + Signed-off-by: Peter Hutterer + +commit ac470dfb4fadaa0b28b6f8b57f4f13a20842b897 +Author: Alan Coopersmith +Date: Thu Feb 12 18:34:22 2009 -0800 + + Check for and report errors writing xorg.conf.new from Xorg -configure + + Signed-off-by: Alan Coopersmith + +commit bd713794ceaa1b2890522554562103c0a2d50f04 +Author: Alan Coopersmith +Date: Thu Feb 12 18:23:59 2009 -0800 + + Correct error message if specified config file is not found + + By making the "Unable to open config file" header a warning, it was + not appearing with the filename when a config file was specified and + not found. Now we make it an error message again, but only issue + the error if a filename was specified - if none was specified, then + we don't even issue a warning, just the "Using autoconfig" info message. + + Signed-off-by: Alan Coopersmith + +commit d229ba7068f0adf1bc8b30b7ccf2fce2c7e8b97b +Author: George Staplin +Date: Fri Feb 6 12:55:09 2009 -0700 + + XQuartz: xpr: The dri.c code for pixmaps was wrong in several ways. They weren't + being exported correctly by Xplugin. + + This should fix a bug with the surface for a window, when an export fails. + Before the export could fail and leave behind an invalid (freed) pointer in the dix privates. + + I have an idea of how to fix the GLXPixmaps now without using CGLSetOffScreen. + + This work is a step towards that. The Xplugin will need a small patch to fix an + issue that this change brought forth. + (cherry picked from commit 58c4116c47543b5e30c2232e7bee8efc0b9be176) + +commit b17d6bed97eddd0773d0845f93f842444b7de0ae +Author: Jeremy Huddleston +Date: Thu Feb 12 15:33:47 2009 -0800 + + XQuartz: Fix copyright info in Info.plist to be prettier when viewed in Finder + (cherry picked from commit 8b607c9c507a68a6abda63831d227d0f8947028c) + +commit d1ad43a8bc3a93a9cddcb4b0be1d722993bf5dd6 +Author: Ian Romanick +Date: Thu Feb 12 14:05:10 2009 -0800 + + Enable pbuffers + + If a driver does not actually support pbuffer rendering, it can just + not enable any pbuffer fbconfigs. + +commit 0065896e1ae6d419a42b318b66d6e4d5f17a40ca +Author: Colin Harrison +Date: Sun Nov 2 20:46:37 2008 +0000 + + Xming: Update keyboard layouts mapping table + + An update to the keyboard layout mapping table which updates some no longer valid keyboard layout names + (e.g. de_CH -> ch -variant de) and adds a few more reported layouts + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit d7c27bfe553235e55eb48799129ff9f8e7e0d4f6 +Author: Colin Harrison +Date: Tue Feb 3 15:57:29 2009 +0000 + + Xming: Check clipboard selection ownership after taking it + + When you take ownership of a selection you should always check you got + it (X11 ICCCM gospel). + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit cb6315072ff23740bdc5fc2f95dd960601df5664 +Author: Colin Harrison +Date: Tue Feb 3 15:54:31 2009 +0000 + + Xming: Fix a memory leak in clipboard integration code + + Free the memory allocated for ReturnData in FlushXEvents(). + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit d54d24dd5f25188b951b82b1c95eb9d36b5be970 +Author: Colin Harrison +Date: Tue Feb 3 15:53:43 2009 +0000 + + Xming: Fix prototype for winProcessXEventsTimeout() in clipboard integration code + + Function 'winProcessXEventsTimeout()' is declared 'static Bool' and + defined 'static int' (correct) in 'winclipboardwndproc.c'. This has been + wrong since dot, but luckily was of no significance + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit d8bae80305057acfb8e82341f68090eac0dccb90 +Author: Colin Harrison +Date: Tue Feb 3 15:52:57 2009 +0000 + + Xming: Improve clipboard integration startup in XDMCP mode + + Improve clipboard integration startup in XDMCP mode + Stop the "Clipboard not yet started, aborting" message from spamming the log. + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit 7cb1622ce44754d9dcd6f405fb8403bd79c3e7e4 +Author: Colin Harrison +Date: Tue Feb 3 15:51:02 2009 +0000 + + Xming: Select events correctly in clipboard integration code + + The clipboard should only select PropertyChange events and do this on + the correct window, the messaging window for the clipboard not the root + window! + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit b997b3849d902894dd5b38673b73a23a47edf502 +Author: Colin Harrison +Date: Tue Feb 3 15:48:04 2009 +0000 + + Xming: Add Colin Harrison to copyright and authors list for clipboard integration files + + Add Colin Harrison to copyright and authors list + Adjust license text for multiple authors + + Signed-off-by: Jon TURNEY + +commit 5ff289834d7445cc6f60dbb1cf18423e7236311a +Author: Jon TURNEY +Date: Mon Jan 19 17:42:54 2009 +0000 + + Cygwin/X: Place prototype for winSelectIcons() in a header file + + Fixes warning as prototype is now in scope for the definition + + Signed-off-by: Jon TURNEY + +commit ab2c98b86f1c3968581dfdad8b67432082feec78 +Author: Jon TURNEY +Date: Mon Jan 19 17:36:58 2009 +0000 + + Cygwin/X: Place prototypes for winCheckKeyPressed() and winFixShiftKeys() in a header file + + Fixes warnings as they are now in scope at the definition + + Signed-off-by: Jon TURNEY + +commit cc7a4c7c6e051fceacb32997740d83fc0655414e +Author: Jon TURNEY +Date: Mon Jan 19 17:29:28 2009 +0000 + + cygwin/X: Remove an unused variable to fix a warning + + Signed-off-by: Jon TURNEY + +commit 7c2fb098b206c0fb97c18cce8f88387cd0c0e82a +Author: Jon TURNEY +Date: Mon Jan 19 17:22:59 2009 +0000 + + Cygwin/X: Fix several prototypes to return HICON + + Fix prototypes of winOverrideIcon(), winTaskbarIcon() and winOverrideDefaultIcon() to return HICON + Also use HICON type in WINPREFS stucture + Remove various casts these changes make unnecessary + + Signed-off-by: Jon TURNEY + +commit 19b3e44ce51cc2f74ba388cb2cd3a26467848e9a +Author: Jon TURNEY +Date: Mon Jan 19 17:03:23 2009 +0000 + + Cygwin/X: Fix -Wold-style-definition warnings + + Fix old-style definition warnings caused by definitions with empty + parameter lists "()", which should be "(void)" to indicate the function + takes no parameters, rather than unspecified parameters + + Signed-off-by: Jon TURNEY + +commit 734b23e5982e171031077a2d5d6b5dc2a12e1a70 +Author: Maarten Maathuis +Date: Wed Feb 11 14:31:17 2009 +0100 + + fb: add fbDoCopy and fbCopyRegion compatibility wrappers. + +commit 619c4d60eb3a8a51a8300ccf8ab5b98a688e6e12 +Author: Eric Anholt +Date: Sun Feb 8 12:37:50 2009 +0100 + + glx: Replace broken GLX visual setup with a fixed "all" mode. + + With trying to match depths so that you didn't end up with a depth 24 + fbconfig for the 32-bit composite visual, I broke the alpha bits on the depth + 24 X visual, which angered other applications. But in fixing that, the + pickFBconfigs code for "minimal" also could end up breaking GLX visuals if + the same FBconfig was chosen for more than one X visual. + We have no reason to not expose as many visuals as possible, but the old + "all" mode didn't match any existing X visuals to GLX visuals, so normal + GL apps didn't work at all. + + Instead, replace it with a simple combination of the two modes: Create GLX + visuals by picking unique FBconfigs with as many features as possible for + each X visual in order. Then, for all remaining FBconfigs that are + appropriate for display, add a corresponding X and GLX visual. + + This gets all applications (even ones that aren't smart enough to do FBconfigs) + get all the options to get the visual configuration they want. The only + potential downside is that the composite ARGB visual is unique and gets a + nearly full-featured GLX visual (except that the root visual might have taken + the tastiest FBconfig), which means that a dumb compositing manager could + waste resources. Write compositing managers using FBconfigs instead, please. + +commit 7bbf05d9aa75b536a5321abf51f37852b8f109ed +Author: Shunichi Fuji +Date: Tue Feb 10 14:44:18 2009 +0900 + + dix: build fix if PANORAMIX is not defined. + + Signed-off-by: Shunichi Fuji + Signed-off-by: Peter Hutterer + +commit 4844bff58f296b2851be4e6b955c3a68d02437a9 +Author: Peter Hutterer +Date: Wed Feb 4 11:50:18 2009 +1000 + + config: if we can't connect to HAL, listen for a startup notification. + + If HAL isn't available when we try to connect, the registered NameOwnerChanged + signal handler waits until HAL is available. Once we connected to HAL, we + unregister the signal handler again. + This allows HAL to be started in parallel or after the server has started. + + Signed-off-by: Peter Hutterer + +commit 5b2bbffc12cd7d29e4200943f8736bd6336ae4aa +Author: Peter Hutterer +Date: Fri Jan 30 16:46:27 2009 +1000 + + dix: deduplicate SyntheticMotion. + + Signed-off-by: Peter Hutterer + +commit 34820f9c1304891bebb4bff6a492288aec4fc0d1 +Author: Peter Hutterer +Date: Fri Jan 30 16:37:19 2009 +1000 + + dix: remove XineramaChangeToCursor. + + It's nearly identical to ChangeToCursor now anyway. + + Signed-off-by: Peter Hutterer + +commit 942eae6868b8b0f343b6aa921ddf77e8bb70798a +Author: Peter Hutterer +Date: Fri Jan 30 16:25:22 2009 +1000 + + dix: Remove XineramaCheckPhysLimits. + + Signed-off-by: Peter Hutterer + +commit bf044f6d10ad9f2baca91a2f04e82dc1af2657cd +Author: Peter Hutterer +Date: Fri Jan 30 16:14:54 2009 +1000 + + dix: reduce redunancy in XineramaConfineCursorToWindow. + + Signed-off-by: Peter Hutterer + +commit cdce5699f085e717ee1a2ed8390746dfcb9ddadf +Author: Peter Hutterer +Date: Fri Jan 30 16:09:23 2009 +1000 + + dix: Get rid of XineramaCheckVirtualMotion. + + Yes, this is an ugly piece mess of #ifdefs, but it beats having two nearly + identical functions. + + Signed-off-by: Peter Hutterer + +commit 1cb5afdf1ef53e72e0eef28ff6197bf5e86ef758 +Author: Maarten Maathuis +Date: Sun Feb 8 18:59:17 2009 +0100 + + dix: always NULL pGC->tile.pixmap in Create{Scratch}GC. + +commit 6fd8fe9cb993f9b733cf7d878bc324acd09aa48d +Author: Maarten Maathuis +Date: Sun Feb 8 18:43:15 2009 +0100 + + exa: fix exaValidateGC. + + - use DEST in the createPixmap wrapper, because stipple already takes MASK (in case someone uses swappers). + - Anticipate some of the less common situations when fbValidateDrawable will access tile related pixmaps. + +commit 30daba6903378dd47f386c0012a0667fd2368f3e +Author: Maarten Maathuis +Date: Sun Feb 8 02:08:13 2009 +0100 + + exa: All fallbacks should have a GC, remove some code. + + - I did some testing with full fallbacks forced by the driver. + - I ran rendercheck, expedite and the (full) x11perf test suite. + - Thanks to ajax for pointing out this should be unneeded. + +commit d591c7d1b87c61ee8add2fce689e35cf097eed4f +Author: Maarten Maathuis +Date: Sat Feb 7 20:47:54 2009 +0100 + + mi: kill a few warnings + +commit 00226d0b589595cdd45c75e7e28237334a8883b1 +Author: Maarten Maathuis +Date: Fri Feb 6 20:56:49 2009 +0100 + + mi: clip exposures to pGC->clientClip. + +commit ffa72793e25f19a388a1369708f472921f721284 +Author: Maarten Maathuis +Date: Fri Feb 6 20:47:01 2009 +0100 + + exa: Calling exaMarkSync after UTS is the drivers responsibility. + +commit da086901c6579e41b28b1418fa12ebb82744fc87 +Author: Matt Turner +Date: Fri Feb 6 18:31:12 2009 -0500 + + Remove unused barrier macros on alpha. + +commit 68665d78e72c4c564073d66a0f59ac8d3f5fd756 +Author: Maarten Maathuis +Date: Thu Feb 5 17:07:31 2009 +0100 + + exa: don't use fbCopyNtoN + +commit 2e76958d304a3c4080d62f32449724eeb9b95d93 +Author: Maarten Maathuis +Date: Wed Feb 4 19:48:03 2009 +0100 + + fb: move some code to mi + +commit f06bde3d4bf9a0f245dfa6c65cd98f8e0c7dfdfc +Author: Maarten Maathuis +Date: Sat Jan 31 19:57:07 2009 +0100 + + exa: create ExaCheckGetImage + +commit 258fc4b106dfd7b583999f97caf4d4a7a11569b6 +Author: Maarten Maathuis +Date: Sat Jan 31 19:40:48 2009 +0100 + + exa: wrap the remainder of exa_unaccel.c + +commit 6fabf249494bf38b69e3c9c6c9718ee8b12fc6e1 +Author: Maarten Maathuis +Date: Sat Jan 31 19:25:20 2009 +0100 + + exa: use proper wrapping in exa.c + +commit 015c99a4ad572f5a1b68054efb070e1ffee1fed0 +Author: Maarten Maathuis +Date: Sat Jan 31 18:53:52 2009 +0100 + + exa: properly wrap GC functions + +commit 5e6a06fe691db63ccf2dc7161a1f6c71c09741b7 +Author: Maarten Maathuis +Date: Sat Jan 31 16:30:31 2009 +0100 + + exa: add GC private + +commit 5cc67ae94c066dcac78072ad8a819c3b602d8bab +Author: Maarten Maathuis +Date: Fri Jan 30 19:32:33 2009 +0100 + + exa: kill of exaImageGlyphBlt + + - It serves no obvious purpose, yet it directly accesses many fb + symbols. + +commit 6483834200fab90164c27c72d4710014edeb4479 +Author: Maarten Maathuis +Date: Fri Jan 30 18:40:46 2009 +0100 + + exa: Remove one of the many calls directly into the fb layer. + +commit f112d6bf59964f317ab236221f7bce7fe2603ab8 +Author: Michel Dänzer +Date: Fri Feb 6 18:30:43 2009 +0100 + + EXA: Try harder to keep current pixmap copy up to date in exaMigrateToward*. + + This is probably required, but apparently not sufficient, for making migration + heuristics other than "always" work correctly again. Not that I really care + about them... + +commit 3948b523893d3d44b6a088340c4252e969613769 +Author: Michel Dänzer +Date: Fri Feb 6 11:37:53 2009 +0100 + + EXA: Guard empty pending region warning by DEBUG_MIGRATE. + + It isn't very useful yet while the damage layer calls us for empty operations, + mostly confuses users. + +commit 9a1d07ecb74b7c3267a6910af66ada917a525110 +Author: Tomas Carnecky +Date: Wed Feb 4 20:20:24 2009 +0100 + + Fix "warning: cast from pointer to integer of different size" + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit 0e15697b53c9448ce9911aa6499b2ea0bda92af6 +Author: Peter Hutterer +Date: Wed Feb 4 10:11:13 2009 +1000 + + dix: die if we can't activate or init the VCP/VCK. + + If we have a busted xkb setup, the XKB initialization on the core devices + fails and leaves us with dev->key->xkbInfo == NULL. This in turn causes + segfaults lateron. + + Return BadValue when the XKB configuration for a master device failed, and if + that happens for the VCP/VCK, die semi-gracefully. + The VCP init can only fail on OOM. + + Reported by Aaron Plattner. + + Signed-off-by: Peter Hutterer + Acked-by: Daniel Stone + Acked-by: Dan Nicholson + Signed-off-by: Aaron Plattner + +commit db836715198a445dc3b32c3df0610f700930483f +Author: Adam Jackson +Date: Wed Feb 4 23:27:32 2009 -0500 + + Revert "Bus: Remove xf86SetAccessFuncs() and related machinery" + + This reverts commit 2fe79990e31ac684637ca3c47198ee0cead10abe. + + Reminder: nvidia uses this, because other parts of RAC are broken in + worse ways, and it needs to disable that. + +commit 2fe79990e31ac684637ca3c47198ee0cead10abe +Author: Adam Jackson +Date: Mon Jan 26 07:11:49 2009 -0500 + + Bus: Remove xf86SetAccessFuncs() and related machinery + + Only mga was using this, and even then only behind an ifdef that no one + ever built. Not a great idea in the first place. + +commit 9ff6180161267e71ae750498da0ec4889cd5c260 +Author: Adam Jackson +Date: Mon Jan 26 06:23:09 2009 -0500 + + Bus: Delete some obfuscatory macros + +commit 8eca37cb8934e79f02291460eb7b49a5d528f391 +Author: Adam Jackson +Date: Wed Feb 4 22:48:52 2009 -0500 + + Bus: Remove unused xf86ExtractTypeFromList + +commit b013111e98d50f5962e9b275eb28d969d5c9576e +Author: Adam Jackson +Date: Mon Jan 26 03:14:18 2009 -0500 + + Bus: make the resource list static. + +commit bde028dd8d856f8d47c39e3c6d3731c34c398c74 +Author: Adam Jackson +Date: Mon Jan 26 22:58:58 2009 -0500 + + loader: Remove icache flushes on alpha. + + If libdl doesn't already get this right for you, you have two problems. + +commit 910a1e88d23e0b7e4ca4679ab50a883a918f2590 +Author: Adam Jackson +Date: Mon Jan 26 07:54:07 2009 -0500 + + loader: Remove a vestige of archive support + +commit 366f23c6eb504fc23112f121769bcb719948474f +Author: Adam Jackson +Date: Mon Jan 26 07:53:05 2009 -0500 + + loader: Remove useless call to LoaderGetOS + + We know what OS you're running on anyway, it's printed in the uname + string at the top of the log. + +commit 05e415a5a3f376e28955be3bf976390e8631e25f +Author: Adam Jackson +Date: Mon Jan 26 07:52:07 2009 -0500 + + loader: Simplify loader magic + +commit bf1ca06a1693cbe5415a152936939ce3b088a022 +Author: Adam Jackson +Date: Mon Jan 26 07:51:35 2009 -0500 + + loader: Make a comment slightly less stupid + +commit 499908aeca9a31692c787070e23bfaa92d1b32a4 +Author: Adam Jackson +Date: Mon Jan 26 07:50:00 2009 -0500 + + loader: code motion. + +commit 28b6b1519c4c526662302fac4ed9b7f540bafd4a +Author: Adam Jackson +Date: Mon Jan 26 07:47:46 2009 -0500 + + loader: Simplify handle allocation and refcounting. + +commit 0bad0552bfd4d10529ab51d618af8d08c5aeea71 +Author: Adam Jackson +Date: Wed Feb 4 17:20:24 2009 -0500 + + XKB: Remove -kb and +kb from -help text and man page. + +commit 7968823cbc02615e1080c0d3f34dcebe14ea1771 +Author: Keith Packard +Date: Tue Feb 3 20:18:37 2009 -0800 + + Handle the combination of panning and crtc transforms + + This patch gets the shadow scanout buffer repainted on panning area changes. + It does not, however, track the mouse correctly. + + Signed-off-by: Keith Packard + +commit 763df9eec79a867978efc5138d8635a46c1a0d17 +Author: Keith Packard +Date: Tue Feb 3 20:15:41 2009 -0800 + + Damage re-used shadow scanout buffer using new transforms. + + When the shadow scanout buffer can be re-used, the underlying framebuffer + area must be damaged so that the scanout will be repainted. This patch + delays the addition of that damaged area until after the transform in the + crtc has been updated, otherwise the old transform would have been used and + the wrong area repainted. + + Signed-off-by: Keith Packard + +commit 62fc98cb88e4e8b636f343453fc1168a87c58972 +Author: Keith Packard +Date: Tue Feb 3 20:12:42 2009 -0800 + + Handle matrix computation overflow in RRTransformCompute + + If the computation of the composite fixed-point transform for RandR + overflows at any point, take the resulting floating point transform and + scale that back to fit in a fixed point matrix. This ensures that a matrix + will always be available, although perhaps at reduced precision. Someday we + should add floating point matrices to Render. + + Signed-off-by: Keith Packard + +commit 7a3ddef3bc27982f3558183c74fc8e365cf2e7fd +Author: Tomas Carnecky +Date: Wed Feb 4 00:07:02 2009 +0100 + + fbdevHWGetRec() doesn't return the private data + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit e04d9ab999978995b06f4b702293b96f69b0e8c3 +Author: Tomas Carnecky +Date: Wed Feb 4 00:08:21 2009 +0100 + + DBUS_API_SUBJECT_TO_CHANGE is already defined on the command line + + dbus-core.c:30:1: warning: "DBUS_API_SUBJECT_TO_CHANGE" redefined + : warning: this is the location of the previous definition + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit 9f60d127b5371f2853dcc53903c6ff8c7670ad1b +Author: Tomas Carnecky +Date: Wed Feb 4 00:07:40 2009 +0100 + + Fix "warning: braces around scalar initializer" + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit a88995dfb88719ae4a03483dc571dbbe83565285 +Author: Tomas Carnecky +Date: Wed Feb 4 00:03:58 2009 +0100 + + Make gcc happy: correct third argument of CopyGetMasterEvent() + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit dd3f4e829286e62f4c5c07ee93f9f576876acbf9 +Author: Tomas Carnecky +Date: Wed Feb 4 00:02:24 2009 +0100 + + Fix "warning: cast to pointer from integer of different size" + + Add parenthesis around the whole expression. + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit 731a907ac21a753f306b53706f689469cefeb953 +Author: Tomas Carnecky +Date: Tue Feb 3 23:58:25 2009 +0100 + + Fix "warning: XXX defined but not used" + + generic.c:80: warning: ‘read_legacy_video_BIOS’ defined but not used + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit bc57efffe654c493ceb62555cccb7926f4a646f1 +Author: Tomas Carnecky +Date: Tue Feb 3 23:54:53 2009 +0100 + + Fix "warning: unused variable XXX" + + events.c:4614: warning: unused variable ‘kbd’ + xkbUtils.c:361: warning: unused variable ‘maxKeysPerMod’ + xf86Events.c:409: warning: unused variable ‘ke’ + generic.c:131: warning: unused variable ‘cs’ + generic.c:130: warning: unused variable ‘size’ + xf86RandR12.c:591: warning: unused variable ‘crtc’ + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit 111fdef74dfc7e288cef675b171b673bf64fe056 +Author: Tomas Carnecky +Date: Tue Feb 3 21:18:53 2009 +0100 + + ISO C90: Move declaration to the beginning of the function + + devices.c: In function ‘DoChangeKeyboardControl’: + devices.c:1768: warning: ISO C90 forbids mixed declarations and code + + Signed-off-by: Tomas Carnecky + Signed-off-by: Peter Hutterer + +commit 9fe9b6e4ef669b192ee349e3290db5d2aeea273c +Author: Peter Hutterer +Date: Mon Feb 2 10:20:13 2009 +1000 + + mi: don't call UpdateSpriteForScreen if we have Xinerama enabled. #18668 + + In Xinerama all windows hang off the first root window. Crossing the screens + must not reset the spriteTrace, otherwise picking fails and events are sent to + the root window. + + X.Org Bug 18668 + + Signed-off-by: Peter Hutterer + +commit f80332a04d0714a74f16b2552cccf3457e6494af +Author: Peter Hutterer +Date: Mon Feb 2 14:28:16 2009 +1000 + + dix: for core events, pass the core filter into DeliverEventsToWindow + + Signed-off-by: Peter Hutterer + +commit 01879e583b0ce0d8fddc37e36d9b69fe4f1d1e72 +Author: Aaron Plattner +Date: Tue Feb 3 13:49:19 2009 -0800 + + Don't enable XAA offscreen pixmaps with Option "XaaOffscreenPixmaps" "no". + + xf86IsOptionSet is rarely the right function to use for boolean options because + it returns TRUE whenever the option is present, even if its value is "no", + "off", "0", etc. + +commit 5623c27700b7b23a8dbbd8c8f45e5d4fa0c667e3 +Author: Alan Coopersmith +Date: Mon Feb 2 19:25:14 2009 -0800 + + Constify atom name strings + + Changes MakeAtom to take a const char * and NameForAtom to return them, + since many callers pass pointers to constant strings stored in read-only + ELF sections. Updates in-tree callers as necessary to clear const + mismatch warnings introduced by this change. + + Signed-off-by: Alan Coopersmith + Acked-by: Peter Hutterer + +commit 6869efae74381e5305b2d6517bf286e3ef7fdcb7 +Author: Jeremy Huddleston +Date: Tue Feb 3 09:58:51 2009 -0800 + + XQuartz: Accept bundle version 2.1.6 to work with our startup method + (cherry picked from commit 7f2966557374cb5ea22bc7b73d5ca04891eb5e78) + +commit 43a1c91a3dac4f0c8cea5de1b9785e349e09978a +Author: Peter Hutterer +Date: Sun Feb 1 20:44:27 2009 +1000 + + Xext: fix typo in GEEventFill macro + + Signed-off-by: Peter Hutterer + +commit 412e09e1e37b74421abd4dd9606536d2785da440 +Author: Peter Hutterer +Date: Thu Jan 15 11:03:03 2009 +1000 + + dix: Use GenericEvent instead of LASTEvent to check for core events. + + Signed-off-by: Peter Hutterer + +commit 9f3c5d46f81887b2b413b3e6cc9df8282f06522c +Author: Peter Hutterer +Date: Mon Feb 2 12:55:39 2009 +1000 + + Xi: initialize the device before applying button map changes. + + This part got lost in bc909f71367a0. + Reported by Tomas Carnecky. + + Acked-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 6bf6a4c47e94780bf0eef47702f21a505c336556 +Author: Simon Thum +Date: Fri Jan 23 12:06:16 2009 +0100 + + Xi: create well-known atoms on demand, rather than preinit them + + Signed-off-by: Peter Hutterer + +commit 32adf8d34fb7db9b78859241f2670349b75c3f7e +Author: Peter Hutterer +Date: Wed Jan 28 20:52:45 2009 +1000 + + Xi: If the MD doesn't have a key/kbdfeed class, init the keyboard. + + This fixes crashes on mouse/keyboard combos where the mouse is both a keyboard + and a mouse. Upon copying the SD's key info into the VCP, the NULL xkbInfo of + the VCP would crash the server. + + Signed-off-by: Peter Hutterer + +commit caade11b8465a743b2ecc890a4d180277e9ca716 +Author: Peter Hutterer +Date: Wed Jan 28 20:44:47 2009 +1000 + + Xi: remove now obsolete oldXkbInfo. + + Signed-off-by: Peter Hutterer + +commit 5e0967f5fc2e0fcf2a55128eb03931cf6f5709e9 +Author: Peter Hutterer +Date: Fri Dec 12 10:49:53 2008 +1000 + + dix: add SetBit(arr, bit) and ClearBit(arr, bit) to include/inputstr.h + + Signed-off-by: Peter Hutterer + +commit faeb18eb7eb20f15528f9c8004d843146ebc2a50 +Author: Peter Hutterer +Date: Tue Jan 27 15:40:31 2009 +1000 + + Xext: rename shape's EventType to ShapeEventType to avoid name clashing. + + Signed-off-by: Peter Hutterer + +commit cfd3443fe81685e91a53063dee58e24a5684dc29 +Author: Peter Hutterer +Date: Wed Jan 28 13:34:10 2009 +1000 + + Xext: rename saver's EventType to SaverEventType. + + Avoid namespace clashing with the internal events. + + Signed-off-by: Peter Hutterer + +commit 8460425740c72e58bcb897557f033e2836251160 +Author: Peter Hutterer +Date: Wed Dec 10 12:31:59 2008 +0000 + + render: rename SetBit to RenderSetBit. + + Avoiding namespace collision with the SetBit macro soon to be used in the + input code. + + Signed-off-by: Peter Hutterer + +commit 940a7aeebc9b042b0208ee475ef162c485ac2274 +Author: Peter Hutterer +Date: Tue Jan 27 15:52:49 2009 +1000 + + include: remove now-unused sempahore macros. + + Obsolete with the new enter/leave model. + + Signed-off-by: Peter Hutterer + +commit f5c7f8a3ab5723b6c1c3cd21b90a87fd7f37cb7f +Author: Jeremy Huddleston +Date: Mon Feb 2 12:30:57 2009 -0800 + + XQuartz: GLX: Fix accidental exclusion of glBlitFramebufferEXT on Leopard + (cherry picked from commit 8620579483789ae9ac1b68dbefe32b70011386de) + +commit 5100d829a4d71ce4a9fbc2b81694a1fb90066ccf +Author: Eric Anholt +Date: Mon Feb 2 10:13:46 2009 -0800 + + glx: Don't match fbconfigs to visuals with mismatched channel masks. + + This fixes at least one known bug, where the depth 32 visual would end up + with a depth 24 fbconfig attached, angering compiz. + +commit 4475ae036c3d5a06a6ce88a359ce2a9a8ae9b4f1 +Author: Jon TURNEY +Date: Mon Jan 26 13:15:07 2009 +0000 + + Cygwin/X: Fix compilation for mandatory XKB + + Fix a stray '}' + Update to use RMLVO interface + Remove g_winInfo.xkb.disable, can never be set since noXkbExtension has been removed + Change to retrieve modifier key state using XkbStateFieldFromRec() from Colin Harrison + Update to use XKB defaults from xkb-config.h + + Signed-off-by: Jon TURNEY + +commit e2b4d3d4a1994347979508fbb483cd55c2232502 +Author: Jeremy Huddleston +Date: Sun Feb 1 15:08:02 2009 -0800 + + XQuartz: GLX: OpenGL.framework on Tiger doesn't have glBlitFramebufferEXT + (cherry picked from commit e01662cadcaa4052e3a5aa82cbaed00a2d0220e0) + +commit 66a6fbbfa3a1fd0d91fe61cf2ab299d0e3df46b9 +Author: Jim Huang +Date: Sun Feb 1 09:19:20 2009 -0800 + + kdrive: Move a close() to the right place so we don't close(-1) normally. + + Bug #18343 + +commit 0055fe66d5f73742cafab868ccdb7a6f36ea1dd5 +Author: Manuel Bouyer +Date: Sun Feb 1 09:14:19 2009 -0800 + + netbsd: Force the use of ev56 instructions for register access on ev56. + + This avoids 32-bit access which might affect other registers. The linux code + uses gcc flags to get this to happen, but this seems like more of a sure thing. + +commit 1375c1b07df114d864a0a06786bc75a3a4f65177 +Author: Manuel Bouyer +Date: Sun Feb 1 09:12:21 2009 -0800 + + netbsd: Fix alpha ev6 support. + + Bug #19721 + +commit ad8ffbe7cb50387e6a9cc4a24fea0d9c13b7bfc8 +Author: John Hein +Date: Fri Jan 30 21:30:35 2009 -0800 + + Actually require bigfontproto when it's enabled. Bug #13710. + +commit 29b3b88dc744f4919c6709747ddb7baac47486c5 +Author: Pierre Willenbrock +Date: Fri Jan 30 21:16:48 2009 -0800 + + Prevent double unref of glxdrawables + + Found by valgrind. Bug #18917. + +commit 60bcdd687040db76490851d4b459284ce37020e0 +Author: Alan Coopersmith +Date: Fri Jan 30 21:44:08 2009 -0800 + + x11-input.fdi: Add options needed to handle adding USB devices on Solaris + +commit 4de5705545792d77c9990bbec3e263a517757a2a +Author: Keith Packard +Date: Fri Jan 30 20:17:10 2009 -0800 + + RandR crtcs not fetched correctly when primary output is set. + + Primary outputs may not have a CRTC. + Loops fetching CRTCs respecting primary output were broken. + + Signed-off-by: Keith Packard + +commit f716e3f3445d443cbc6507d27f806e9ad387120a +Author: Eric Anholt +Date: Fri Jan 30 20:10:21 2009 -0800 + + modes: Protect xf86_crtc_supports_gamma() from non-RandR 1.2 drivers. + +commit 317f2b4a9fe4b606975711bc332166a82db5087d +Author: Eric Anholt +Date: Fri Jan 30 19:06:17 2009 -0800 + + randr: Avoid re-querying the configuration on everything but GetScreenResources. + + The new path should only re-query on the other requests when we haven't + gathered the information from the DDX yet (such as with a non-RandR 1.2 DDX). + + Bug #19037. + +commit c1f2be1f3fd0c80cb4c85b98140b98aa9311242b +Author: Keith Packard +Date: Fri Jan 30 19:14:36 2009 -0800 + + RandR SetOutputPrimary should work with or without a crtc connected + + The test was inverted from the protocol spec, and besides, the test is bogus + anyways. + + Signed-off-by: Keith Packard + +commit dd098501d800571e71d06ffc936635a6c9d44e05 +Author: Eric Anholt +Date: Fri Jan 30 16:17:58 2009 -0800 + + Move the apple fat binary hacks back to a header file, and make it apple-only. + +commit 6e49fdd2c839b6244c23ce95c3ae76a1a265b986 +Author: Keith Packard +Date: Fri Jan 30 15:43:04 2009 -0800 + + Make RandR CRTC info report panning area instead of just crtc area + + This makes the RandR info consistent with the Xinerama info. + + Signed-off-by: Keith Packard + +commit ea309e47457156b60aadbf113f04e5b6851029c8 +Author: Keith Packard +Date: Fri Jan 30 14:19:41 2009 -0800 + + Make crtc_notify wrap/unwrap code do nothing unless mode code is inuse + + Drivers not using the new hw/xfree86/modes code would crash in DRI due to + that code trying to monitor CRTC changes. + + Signed-off-by: Keith Packard + +commit fff00df94d7ebd18a8e24537ec96073717375a3f +Author: Keith Packard +Date: Fri Jan 30 14:16:12 2009 -0800 + + Ignore EDID-supplied monitor physical sizes for core screen size + + We report the EDID values in RandR, and we let people configure whatever + they like for the screen in xorg.conf. Reporting the EDID values in the core + means applications get inconsistent font sizes in the default configuration. + + Signed-off-by: Keith Packard + +commit 4e7bc2d7a9879297aff0924febeca9bd3b6c336d +Author: Paulo Cesar Pereira de Andrade +Date: Fri Jan 30 03:56:39 2009 -0200 + + Work around inclusion of + + The X Server build only needs the macros PANORAMIX_MAJOR_VERSION + and PANORAMIX_MINOR_VERSION from that header. + Addition of extra prototypes to + caused a X Server build failure. + +commit 15ac25627e7239629be59adedda5d8a9deb330ef +Author: Matt Turner +Date: Thu Jan 29 14:40:27 2009 -0500 + + Remove extra / from Loading module line + + In the X log, upon module load, it prints a line similar to the following. + + > (II) Loading /usr/lib64/xorg/modules/extensions//libdbe.so + + The attached patch removes the extra / before the module name. + + Code already exists in hw/xfree86/loader/loadmod.c:InitPathList to add a + trailing slash if needed, removing the one added by sprintf is harmless. + + Signed-off-by: James Cloos + +commit 3903e23153a0d47704050944fa83a7b31bc2eebc +Author: Maarten Maathuis +Date: Thu Jan 29 12:42:57 2009 +0100 + + XKB: ifdef XKB is dead, yet it was still present in a few places. + +commit 756a2c848305aa06bfe636314e28ad7f97c1168a +Author: Pierre Willenbrock +Date: Wed Jan 28 22:18:50 2009 +0100 + + Fix duplicate code, off-by one in space calculation, not initialized members + +commit b55db937fee5ddf34518a54f352c0ee9b8cc2c2c +Author: Søren Sandmann Pedersen +Date: Wed Jan 28 17:33:25 2009 -0500 + + [shadow] Store the closure in the buffer + +commit 82fc102568b3d6b0daeb6c5f5b3a1310a7f14fcd +Author: Peter Hutterer +Date: Wed Jan 28 07:51:32 2009 +1000 + + Disable Xnest by default. + + Xneest doesn't build right now and is in the process of being removed. So + let's not build it by default. + +commit f7585d020593345e7a61b1fe1a517c38da8f6d79 +Author: Paulo Cesar Pereira de Andrade +Date: Tue Jan 27 03:13:28 2009 -0200 + + Convert kdrive libraries to libtool convenience libraries. + + Also correct a link failure due to unresolved symbols. This + is arguably a libtool/ranlib/ld bug, that "may" be corrected + by linking all convenience libraries in a single one. But in + this case, it was preferred to just add a linker option to + Xfake_LDFLAGS, to force linkage of all libraries. + + This corrects #19725. + +commit b43a4e2d38adb9ff16204218f24df93cfd820d1b +Author: Magnus Kessler +Date: Tue Jan 27 09:30:37 2009 +1000 + + Xvfb: add missing include for new xkbsrv.h + + Commit 08363c5830bdea34012dcd954b45ccfdc79a3a7e added call to + XkbGetRulesDflts defined in xkbsrv.h + + Signed-off-by: Magnus Kessler + Signed-off-by: Peter Hutterer + +commit a7e6424d992151c504a75bb807fa34c57ad52cad +Author: Peter Hutterer +Date: Fri Jan 23 17:11:57 2009 +1100 + + dix: remove obsolete comment. Event lists should not be allocated by the DDX + +commit b137a0cdb1d517a807d081fbde8e8dc0539fb3ca +Author: George Staplin +Date: Mon Jan 26 03:02:38 2009 -0700 + + XQuartz: GL: indirect.c changes to fix the build with newer OpenGL frameworks. + (cherry picked from commit 143224405ba74929c702a95de52b56df140b0d1b) + +commit 34b896e50fa154ad890a458c60932964421e739f +Author: Jeremy Huddleston +Date: Fri Jan 16 19:26:07 2009 -0800 + + XQuartz: SnowLeopard: Help system book name changed in 10.6 + (cherry picked from commit b57cb05c69acbedb00a97234099ea104309aa2cb) + +commit c7d62d74bbd6e0067f511fd910b9e633576befb7 +Author: Paulo Cesar Pereira de Andrade +Date: Sun Jan 25 19:19:01 2009 -0200 + + Explicitly add libxorg_la_LIBADD to libxorg_la_DEPENDENCIES + + Otherwise, for example, when hacking config/*.c, it is required to + run make clean on that directory, to ensure the proper libconfig.a + will be linked in the generated Xorg binary. + +commit 0eec7348875035792097f606487a558422774e2c +Author: Magnus Kessler +Date: Fri Jan 23 12:11:00 2009 -0200 + + .gitignore: ignore sdksyms.dep + + Signed-off-by: Magnus Kessler + +commit 88efe9c76f85423d952ff38e0cfd8b690439902a +Author: Magnus Kessler +Date: Fri Jan 23 11:25:00 2009 -0200 + + Xvfb: Remove unused function GetLK201Mappings in InitInput.c + + Silence a gcc warning. After commit 08363c5830bdea34012dcd954b45ccfdc79a3a7e + GetLK201Mappings is no longer needed. + + Signed-off-by: Magnus Kessler + +commit 6ef46c40e62def4841a4cff4e0b443516a2ed782 +Author: Paulo Cesar Pereira de Andrade +Date: Sun Jan 25 19:01:10 2009 -0200 + + Update xnest keyboard code to match xephyr/kdrive. + + Xnest was not updated in the last batch of xkb changes. This + patch is basically cut&paste from hw/kdrive/src/kinput.c and + hw/kdrive/ephyr/ephyr.c, and appears to generate a Xnest as + functional as before the xkb changes. + +commit 759348f13ffa02fb1804238189ffce1c92d33e2a +Author: Adam Jackson +Date: Thu Jan 22 22:09:11 2009 -0500 + + Xephyr: Add some rudimentary Xnest command line compat + +commit 01ffaf6834fd6a3b1ddbd33d0f0b8d086be0737e +Author: Peter Hutterer +Date: Mon Jan 19 21:11:05 2009 +1100 + + Xi: define a range of axis labels. + + This is copied from linux/input.h, presumably that's the ones at least the + Linux kernel can give us for any device. + + Signed-off-by: Peter Hutterer + +commit 132b464d734b077038e19b21e46d3a6258f4b998 +Author: Adam Jackson +Date: Thu Jan 22 02:11:16 2009 -0500 + + Remove a bunch of useless casts. + + We've had void * for twenty years now people let's try to act like we + know how it works. + +commit 0fb4390526bb829ab17ff4635d41a3012f63c1b2 +Author: Adam Jackson +Date: Thu Jan 22 01:13:17 2009 -0500 + + Remove xorgcfg 'Options' list. + +commit b5ece46b76c9182a077141505cc0e788c0fca3ea +Author: Daniel Stone +Date: Sat Jan 17 20:07:40 2009 +0200 + + Input: Clarify valuator FatalError messages + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit ddb46c1421a98190313b79621c0f5ae43b97d3f6 +Author: Daniel Stone +Date: Sat Jan 17 20:03:10 2009 +0200 + + Input: Use previous state in valuator events + + Make valuator event state match other events by using the device state + from before processing the event, not after. Also, we already check the + number of valuators in UpdateDeviceState, so no need to do it again. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 18e51911f5748f9a27a18cd6bb421c899f5b048f +Author: Daniel Stone +Date: Sat Jan 17 13:44:47 2009 +0200 + + Input: Add postdown to ButtonClassRec + + This mirrors that in KeyClassRec: the state of the buttons as posted to + GetPointerEvents, rather than the state of the buttons as processed by + ProcessOtherEvent and friends. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit c05f3bc935bef4d98216d38c0e98bd57ced382df +Author: Daniel Stone +Date: Sat Jan 17 19:35:15 2009 +0200 + + Input: Don't allow DDX to generate repeat events + + We always use soft-repeat at the moment; XKB posts a release/press sequence, + which admittedly needs cleaning up, but that's for another day. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 6aef4e96affcc26b9415c6bc9c9bebb5af05a1a7 +Author: Daniel Stone +Date: Sat Jan 17 13:42:46 2009 +0200 + + XKB: Fix logic error + + Use logical or instead of bitwise or. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 0ec9b1069aeeaf66fd104c09f08202f7ebccd170 +Author: Daniel Stone +Date: Sat Jan 10 14:44:28 2009 +0200 + + XKB: Remove unused XkbProcessOtherEvent + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 7af53799ccefe27b996bdc6073075b7f12353d3a +Author: Daniel Stone +Date: Sun Jan 4 16:39:18 2009 +1100 + + Input: Remove unused CoreProcess{Keyboard,Pointer}Event + + Everything goes through XKB's Process{Keyboard,Pointer}Event on its way + through to ProcessOtherEvent now, so get rid of the old, useless functions. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 7c4c00649cae855e141ec9a3667bfe399e3156d0 +Author: Daniel Stone +Date: Thu Jan 1 05:25:52 2009 +1100 + + XKB: Remove unused DDX functions + + They were complete no-ops anyway. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 4fa3872dc2bcfd6d1fc88d0a94c7071683eea899 +Author: Daniel Stone +Date: Sat Oct 18 20:26:00 2008 +0100 + + Input: Remove core keysyms from KeyClassRec + + Instead of always keeping two copies of the keymap, only generate the + core keymap from the XKB keymap when we really need to, and use the XKB + keymap as the canonical keymap. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit bc909f71367a02297e725bb5769c2bcadab22395 +Author: Daniel Stone +Date: Tue Aug 12 00:35:56 2008 +0300 + + Input: Centralise pointer map changing + + Replace both core and Xi functions with one function that validates the + proposed map, and sends out both kinds of notification. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit f06a9d2e05e13466c115fc706966a90b1fb0518e +Author: Daniel Stone +Date: Sat Oct 18 20:59:30 2008 +0100 + + Input: Clean up keymap change notifications + + Keyboard map notifications are always generated from within XKB code, + which also takes care of copying the keysyms, etc. If you need to + mangle the keymap yourself, generate a new core keymap/modmap, and pass + it to XkbApplyMappingChange. + + SendMappingNotify is renamed to SendPointerMappingNotify (and ditto its + Device variants), which still only _sends_ the notifications, as opposed + to also doing the copying a la XkbApplyMappingChange. + + Also have the modmap change code traverse the device hierachy, rather + than just going off the core keyboard. + + Signed-off-by: Daniel Stone + +commit 1d1a0f67eee330a286fbdef17e967ce8ea201548 +Author: Daniel Stone +Date: Tue Oct 28 16:27:37 2008 +1100 + + Xi: Introduce XIShouldNotify + + XIShouldNotify just lets you know if you should send an event for a + keymap change (or similar) concerning a given device to a given client; + at the moment, this is only for devices which are sending events to that + client. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit b5242789edb726bb539e4d42fa35bfdaca0025c9 +Author: Daniel Stone +Date: Sat Aug 9 17:02:02 2008 +0300 + + XKB: Simplify keymap writing a bit + + We don't need no temporary variable. + + Signed-off-by: Daniel Stone + +commit feb757f384382c7782ceac55f99d54c7caadbd9d +Author: Daniel Stone +Date: Fri Apr 25 19:54:29 2008 +0300 + + XKB: Sanitise vmods for redirected keys + + Turn two unsigned chars into one unsigned int for both vmods and the + vmod mask. As a bonus, remove broken unused accessor macro for setting + the vmods. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit b5f49382fe48f0a762d9a15fb10a7d7e1183fc8d +Author: Daniel Stone +Date: Fri Apr 25 19:52:11 2008 +0300 + + XKB: Sanitise ctrls action + + Turn four unsigned chars into one unsigned long. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 1bd7fd195d85681e722161f8c636a28f11b40abb +Author: Daniel Stone +Date: Fri Apr 25 19:50:12 2008 +0300 + + XKB: Sanitise pointer actions + + Turn two unsigned chars into one int. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 61c508fa78aa08ea2666fde950fbafad95d65056 +Author: Daniel Stone +Date: Fri Apr 25 19:47:16 2008 +0300 + + XKB: Sanitise vmods in actions + + Turn vmods from two unsigned chars into one int. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 4f3078df2acf07364a17f738bc5c32f9a5962723 +Author: Daniel Stone +Date: Sat Aug 9 16:50:22 2008 +0300 + + XKB: Add a hell of a lot more FIXMEs + + For some reason, we insist on having daft internal representations that + make no sense, that always have to be converted to be used. We should + really sort this one out. + + Also, comment the hojillion members of XkbStateRec. + + Signed-off-by: Daniel Stone + +commit edeb033f29ae2d52e86c543ccbf22fb7ae1fdb7f +Author: Daniel Stone +Date: Wed Oct 15 18:07:52 2008 +0100 + + XKB: Explicitly decode action data + + Rather than requiring a one-to-one correspondence between XKM and struct + formats in action data, explicitly fill the action data, so we can break + API. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 699824a35791735b4b4378adf8648e3aaa599ace +Author: Daniel Stone +Date: Fri Jul 18 16:41:29 2008 +0300 + + XKB: Remove descriptions from maprules + + We don't use them, as they're not up to the task. We'll get a better + solution someday, promise. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 1ad80678d88525159e55bb814205fd4eaca060a3 +Author: Daniel Stone +Date: Fri Jul 18 16:29:35 2008 +0300 + + XKB: Remove support for setting combined keymaps + + We don't do full keymaps anymore. Deal. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 133e0bd6f1cc147a2ddbbe1519d10300e7dbe669 +Author: Daniel Stone +Date: Fri Jul 18 16:27:57 2008 +0300 + + XKB: Remove 'extra' functionality from rules parsing + + When we find something weird in the rules, don't stash it as an extra + freeform component, just state that the rules file is likely broken and + move on with our lives. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit cc5c6d628aa0dc87c2cc9d063972e09ad747a596 +Author: Daniel Stone +Date: Fri Apr 25 18:45:37 2008 +0300 + + XKB: Remove unsupported Xi operation flags + + We support every XKB operation on Xi devices, so always report that we + support everything, and that nothing is ever unsupported. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 32db27a7f867b503c2840ca7b815e96d10be9210 +Author: Daniel Stone +Date: Thu Apr 17 12:42:01 2008 -0700 + + Input: Remove modifierMap from core + + We already have modmap (in the exact same format!) in XKB, so just use + that all the time, instead of duplicating the information. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit f062e90a95f9b7ae5458ef2100615e8ace9b66a7 +Author: Daniel Stone +Date: Wed Apr 16 19:15:30 2008 +0300 + + Input: Remove modifierKeyMap + + Since modifierKeyMap is generated from modifierMap, just remove it, and + only generate it when we need to send the modifier map to the client. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit cf6a2fc2bd94b392cfea120444d5e032d26f1d37 +Author: Daniel Stone +Date: Sun Mar 30 17:25:39 2008 +0300 + + Input: Ignore modifiers in core input processing + + Modifiers get cleared by the XKB code when we drop down into core input + processing, so just delete the dead code path to simplify things a bit. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 6727ee94086d238f9f99379280d4e4827f388cce +Author: Daniel Stone +Date: Sat Aug 9 16:45:45 2008 +0300 + + Input: Remove state from KeyClassRec + + We already have state fully stored within XKB, so instead of duplicating it, + just generate the values to send to clients when required. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 08363c5830bdea34012dcd954b45ccfdc79a3a7e +Author: Daniel Stone +Date: Fri Dec 26 18:38:25 2008 +1100 + + Input: Overhaul keyboard initialisation process + + XkbInitKeyboardDeviceStruct is now the only valid keyboard + initialisation: all the details are hidden behind here. This now makes + it impossible to supply a core keymap at startup. + + If dev->key is valid, dev->key->xkbInfo->desc is also valid. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 40877c6680863bd6a8475f2bb0c54df55bcf0b0e +Author: Daniel Stone +Date: Tue Mar 18 10:05:57 2008 +0200 + + XKB: Make XKB mandatory + + No more #ifdef XKB, because you can't disable the build, and no more + noXkbExtension either. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit f9da417163b6b2d6234d2542c1f375e33db7159a +Author: Daniel Stone +Date: Fri Mar 28 19:26:42 2008 +0200 + + Xephyr: Disable state inheritance hack + + Prepare for the impending removal of the state field by disabling this hack + for a while: it's hell of nasty and I'm amazed it ever really worked. + + Basically, on focus out, it should do as current DDXes do and fake releases + for all keys (not just mangle the core state) that are currently down; + buttons too. When focus comes back in, we already have a KeymapNotify that + lets us know what's currently down, so we can use this to fake the + appropriate keypresses, and send it through the event routing layer. + + Signed-off-by: Daniel Stone + +commit 5c281446d2a8c38511d45baee6f0d25a640f12e1 +Author: Daniel Stone +Date: Fri Mar 28 11:56:34 2008 +0200 + + XKB: Remove lock actions disabling + + For some reason, XKB allows clients to set a global (!) flag that simply + turns lock keys into state no-ops. Ignore this flag. + + Signed-off-by: Daniel Stone + +commit 2f7cb7306709d5266688e05a066701d309323035 +Author: Daniel Stone +Date: Fri Dec 26 18:36:30 2008 +1100 + + XKB: Add XkbRMLVOSet + + XkbRMLVOSet is just a set of strings for rules, model, layout, variant + and options; use that in preference to XkbRF_VarDefsRec, which is a + hideously complicated monster that somehow managed to not include the + actual rules. + + While we're at it, clean up xkbrules.h so it doesn't require xkbstr.h. + + Signed-off-by: Daniel Stone + +commit 3208eff2d7b856a4ac0d650498ccb36e7e62e034 +Author: Alan Coopersmith +Date: Wed Jan 21 14:45:06 2009 -0800 + + Update xorg-server.h.in & Xnest to use new XKB default #defines + +commit cdf216aa17b776130a91fa6792c321620e810a18 +Author: Alan Coopersmith +Date: Tue Jan 20 16:50:43 2009 -0800 + + Fix XKB default configure options to use the right variables + +commit 0d3c3dad21857ef502fe0d0235620a1228248e13 +Author: Adam Jackson +Date: Tue Jan 20 23:46:48 2009 -0500 + + Don't include fontmod.h + +commit 4f004ab9a83e209943760e1faa39f06c8eeb27af +Author: Benjamin Close +Date: Thu Jan 22 01:36:31 2009 +1030 + + XNest: Fix the build caused by __XKBDEFRULES__ -> XKB_DFLT_RULES in 23862ede59a9ce11a06ec5151bde460fb836c603 + +commit e45a8a498d1ee8e40fc334805c38c05cb04b49d0 +Author: Adam Jackson +Date: Tue Jan 20 22:07:07 2009 -0500 + + Remove CreateUnclippedWinSize from window.h + + This is utterly wrong, but then, so is sdksyms.sh + +commit 20d2117eb82fb7ce91afd4f01fe1b9d4b1bf0459 +Author: Adam Jackson +Date: Tue Jan 20 00:17:42 2009 -0500 + + Move CreateUnclippedWinSize to mbuf + + It's the only user, so. + +commit e1611d8d55ade35bf8c2e1da7bc2dbf05b011301 +Author: Daniel Stone +Date: Thu Nov 27 13:34:05 2008 +1100 + + XKB: Be more verbose about XkbCopyKeymap failure in debug mode + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 5ee504cae5f17d16ee7599beaa4b1944a9a8aee6 +Author: Daniel Stone +Date: Thu Nov 27 13:31:18 2008 +1100 + + XKB: Trying to copy to the same keymap is not fatal + + Humour the user if they run XkbCopyKeymap(foo, foo). + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit d929ba6886c6f121b101c0bf0f7c63ef9d851a3e +Author: Daniel Stone +Date: Fri Dec 26 18:32:41 2008 +1100 + + XFree86: Input: Fail NIDR if we can't enable devices + + If we can't enable a device, bail out of NewInputDeviceRequest rather than + blithely continuing. Also, be more verbose when initialization failed. Also, + be more verbose when initialization failed. Also, be more verbose when + initialization failed. Also, be more verbose when initialization failed. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 534669b376a6a703fbc97269d279a5418cf60c98 +Author: Daniel Stone +Date: Sun Jan 4 16:24:44 2009 +1100 + + XKB: Remove unnecessary prototype + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 27ea1a7e4e7eca394e052791c64bc6be515e075e +Author: Daniel Stone +Date: Sun Jan 4 16:40:46 2009 +1100 + + XKB: Only Xi events are processed + + Core events aren't run through these functions, so don't bother testing + for them. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 2762cafc321166ac48f80146720709b1011e6894 +Author: Daniel Stone +Date: Thu Nov 27 13:35:42 2008 +1100 + + XkbCopyKeymap: inputInfo.keyboard is not a special case + + The device-walking code is still depressing, though. + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 07c3bb922b8c1ae8fe28198f6c451430b836f21e +Author: Daniel Stone +Date: Sun Jan 4 14:12:19 2009 +1100 + + XKB: Move XkbCopyKeymap definition to xkbsrv.h + + Signed-off-by: Daniel Stone + Signed-off-by: Peter Hutterer + +commit 4f35a34f3cd3df3c8af129936d1808dcc063d8a7 +Author: Peter Hutterer +Date: Thu Aug 14 17:26:02 2008 +0930 + + dix: Remove traces of CoreDevicePrivateKey + + Obsolete with MPX. + + Signed-off-by: Peter Hutterer + Signed-off-by: Daniel Stone + +commit 58a6b5b9eff25b1b9c6de239d6f5a952a31dfd15 +Author: Peter Hutterer +Date: Thu Aug 14 17:21:28 2008 +0930 + + Xi: don't care about CoreDevicePrivateKey when copying keys + + If we get here, we must copy. + + Signed-off-by: Peter Hutterer + Signed-off-by: Daniel Stone + +commit 23862ede59a9ce11a06ec5151bde460fb836c603 +Author: Daniel Stone +Date: Sat Jan 10 14:36:16 2009 +0200 + + XKB: Allow build-time configuration of XKB defaults + + Instead of hardcoding base/pc105/us, allow users to change the defaults at + ./configure time. Change the default model to be evdev on Linux. + + Signed-off-by: Daniel Stone + +commit 57a5eb2583bdf1069861fe2184de9739c0d0c7c1 +Author: Colin Harrison +Date: Sun Jan 4 17:40:54 2009 +0000 + + Xming: Update .rc file to tidy About and Exit dialogs + + Tidy up About and Exit dialogs, add keyboard accelerators + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit df3f701ef2d41ff4be4839f473ae71d33777bea3 +Author: Colin Harrison +Date: Sun Nov 2 20:17:23 2008 +0000 + + Xming: Ensure we dont try to print a null value when displaying xkb configuration + + Signed-off-by: Jon TURNEY + +commit c612f0df45d388c98108e1e0c80afdfae05e338e +Author: Colin Harrison +Date: Sun Nov 2 20:46:41 2008 +0000 + + Xming: Add styles keyword and attributes. + + Add handling for style keyword and atttributes in .XWinrc + Update man page to document these additions + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit 847a91ad2e8c9b0992c09f0ec38466885d6d25d4 +Author: Colin Harrison +Date: Sun Jan 4 19:57:56 2009 +0000 + + Xming: Notify X when the keyboard focus is lost to a pure Windows window in -multiwindow mode. + + Fix internal WM to notify X when the keyboard focus is lost to a pure Windows window in -multiwindow mode. + + Signed-off-by: Jon TURNEY + +commit 6c3e7e4d0568ecb7d10b814593bbbd817b7d2aab +Author: Colin Harrison +Date: Sun Jan 4 18:28:34 2009 +0000 + + Xming: Only allow WM_MOUSEWHEEL messages to act on the client area of a focused window. + + Fix internal WM so it only allows WM_MOUSEWHEEL messages to act on the client area of a focused window. + + Signed-off-by: Jon TURNEY + +commit d84e0e387764236c7e97509baf4606d96cfc19d4 +Author: Colin Harrison +Date: Sun Jan 4 18:20:44 2009 +0000 + + Xming: Correctly parent XA_WM_TRANSIENT_FOR windows in -multiwindow mode when a windows window is created. + + Fix internal WM to correctly parent XA_WM_TRANSIENT_FOR windows in -multiwindow mode when a windows window is created, + and to de-iconize parent windows when a child window acquires focus. + + XXX: Perhaps we should also shuffle parent(s) forward through Z-order when a child acquires focus? + + Signed-off-by: Jon TURNEY + +commit 8d82d806ee38d45ec7cdbf8b1f671ec1e4437567 +Author: Colin Harrison +Date: Sun Nov 2 20:46:38 2008 +0000 + + Xming: Correct the way display and screen number is reported in window titles and tooltips + + Correct the tooltip text for the toolbar X icon to be strictly correct, 'display-number:screen' should be ':display-number.screen'. + Also for the default window title. + Adjust the style of the Windows title in XDMCP mode from 'Xming - hostname' to 'hostname:display-number.screen'. + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit 656f92e247f2f257105daf1e0e6998e1b3d10eba +Author: Jon TURNEY +Date: Sun Nov 2 20:17:02 2008 +0000 + + Cygwin/X: should also use GetTickCount(), just like Xming + + Signed-off-by: Jon TURNEY + +commit 7f781e780ee2b261c5cb688a8e6c5dc1a53bdb6f +Author: Yaakov Selkowitz +Date: Wed Dec 17 19:13:30 2008 +0000 + + Cygwin/X: Cygwin doesn't have RTLD_LOCAL + + RTLD_LOCAL is not defined on Cygwin + + Signed-off-by: Jon TURNEY + +commit 5630ff809676ab6b4833757060e48e201e8768a5 +Author: Dan Nicholson +Date: Mon Dec 22 13:53:43 2008 -0800 + + Ignore generated dmx manpages + + Signed-off-by: Dan Nicholson + +commit 64f924fbf513b5f91b3f2bcf5767f6fe927d3f7a +Author: Dan Nicholson +Date: Mon Dec 22 13:43:30 2008 -0800 + + Kill off --with-mesa-source completely + + There were a few spots left in the source that were using the + --with-mesa-source defined headers or the now removed $(top_srcdir)/GL + directory. These aren't needed anymore as all the necessary source for + GLX is in $(top_srcdir)/glx. + + Signed-off-by: Dan Nicholson + +commit e1c8ee215797fa8c6a3a5421bb6f99b58064248b +Author: Dan Nicholson +Date: Sun Dec 14 14:18:58 2008 -0800 + + Don't reuse PKG_CHECK_MODULES identifiers + + Using GL for the PKG_CHECK_MODULES identifier multiple times means only + the first call will actually be used. Later calls will be skipped due to + GL_CFLAGS and GL_LIBS already being set. This changes DRI to using a + different identifier and DMX to just reusing GL_CFLAGS. + + Signed-off-by: Dan Nicholson + +commit ca46c01869768a8661a9d8a71493ed5f0760a8ab +Author: Jeremy Huddleston +Date: Fri Jan 16 13:58:20 2009 -0800 + + XQuartz: mieq: Wait for the server to finish initializing before letting other threads mieqEnqueue + + Avoid possible race condition whereby one thread might call mieqEnqueue before InitAndStartDevices finishes + (cherry picked from commit 94e417ac87a98cd5c6bf2d7c495d702748398931) + +commit b33905234025f005819c7e2acd653a3a0ecfeb82 +Author: Peter Hutterer +Date: Fri Jan 9 14:32:14 2009 +1000 + + xfree86: always force RAW mode under linux. + + The previous check for AEI on left us with the possibility that AEI is forced + off in the config, but devices are added through evdev nonetheless. A keyboard + added this way can CTRL+C the server. Even when we use kbd, we can set the + mode to RAW, so it's safer alround to to so. + + Signed-off-by: Peter Hutterer + +commit 646add4eb47c01c045ef64405510a08ecfb5bfd6 +Author: Peter Hutterer +Date: Fri Jan 16 16:04:07 2009 +1000 + + dix: remove XineramaCheckMotion, merge into CheckMotion. + + For 5 lines difference, we don't need to duplicate the whole function. really. + +commit e76f904abe530df9ddaa09510088c545298b0b91 +Author: Peter Hutterer +Date: Fri Jan 16 15:53:50 2009 +1000 + + dix: remove "notyet" conditional compilation. + + If it's "notyet" for 3 years, then "yet" won't arive anytime soon. + +commit 5c41b3f9c8b72f0115019237f812d74610b82c24 +Author: George Staplin +Date: Mon Jan 12 19:30:43 2009 -0700 + + XQuartz: xpr: Cleanup some of the code and possibly fix part of the GLX Pixmap problem. + + Split DRICreateSurface into 3 functions. + + Make CreateSurfaceForPixmap use xp_configure_surface. I suspect this is partly why + GLXPixmaps never worked. + + It will require some more work and thoughts for pbuffers, unless we fake those with + pixmaps and surfaces. + (cherry picked from commit 9cf264e67744262b9f45079e6cd752eb3e3b0e08) + +commit c5d52d4c37ebec92cb668b0653bf3de2bdfc5782 +Author: George Staplin +Date: Mon Jan 12 14:14:44 2009 -0700 + + XQuartz: GL: Make indirect.c build and work in the 1.6 branch. + + Reorder some header files and provide some types earlier on. + + Remove the static __GLXextensionInfo __glDDXExtensionInfo; that isn't + used in 1.4 or 1.5, and seems to have been removed from 1.6. Remove + the data structures associated with that too. + + Remove __glXAquaDrawableResize. The GLX structure doesn't use + it anymore, and the Apple version did nothing useful before. + + __glXAquaDrawableSwapBuffers(): base no longer contains a drawGlxc + member. Now provide the the Apple/Aqua context in the __GLXAquaDrawable + struct. Add the context member to the __GLXAquaDrawable struct. + + Remove the fallback configs. They aren't used in 1.4 or 1.5 either. + + Remove init_visuals(). It's not used in 1.4 or 1.5 either. + + In the drawable constructor initialize the state properly, including + the new context member. + + Remove glAquaResetExtension() -- it's not used anymore. + + This has been tested remotely and proven to work with glxgears, + fire, various texture programs I wrote, and various Mesa demos. + (cherry picked from commit d514152195452ae11ec7769c76489651620ad380) + +commit 1c504dff8a84804888f18ee9b184f0c4584185d7 +Author: Jeremy Huddleston +Date: Thu Jan 15 21:03:27 2009 -0800 + + XQuartz: Fix builddir != srcdir issues and undef _XSERVER64 where appropriate on fat binary compilation + +commit 717a961528ec69a6e630d536e15568670e0b398a +Author: Thomas Jaeger +Date: Sat Dec 20 16:17:02 2008 +0100 + + Don't release grabs unless all buttons are up + + Previously, only buttons <= 5 would count here, but the core protocol + allows for 255 buttons. + + http://lists.freedesktop.org/archives/xorg/2009-January/042092.html + + Signed-off-by: Peter Hutterer + +commit f7f85f696570541e2dd43462675de9e6ee46f545 +Author: Thomas Jaeger +Date: Mon Dec 22 00:55:09 2008 +0100 + + Count the number of logically down buttons in buttonsDown + + This fixes the following bug. Assuming your window manager grabs + Alt+Button1 to move windows, map Button3 to 0 via XSetPointerMapping, + then press the physical button 3 (this shouldn't have any effect), press + Alt and then button 1. The press event is delivered to the application + instead of firing the grab. + + Signed-off-by: Peter Hutterer + +commit 6f065a26db62aaf1d7aab9f9a3d0e6cd09d1c1dd +Author: Colin Harrison +Date: Thu Jan 15 21:42:26 2009 +0000 + + Xming: Reduce MOUSE_POLLING_INTERVAL + + Reduce MOUSE_POLLING_INTERVAL from 500ms to 50ms + + In -multiwindow mode, this determines how quickly X windows can track the mouse + when it is outside any X window... Move the cursor around an xeyes, and you will + see what this timing effects. + +commit c745db1674c3cb55249c9eb6e74939b74c42409c +Author: Jon TURNEY +Date: Mon Jan 5 13:52:45 2009 +0000 + + GLX: Avoid a crash when we have an uninitialized GL context + + If the GL dispatch table pointer points to glapi_noop_table, + (due to some kind of GL initialization failure), DoGetString(GL_VERSION) + (for example as invoked by glxinfo) will crash as it tries to + do atof() on the null pointer returned by the noop dispatch function + + Signed-off-by: Jon TURNEY + +commit e30032d0bbbf0bf71e1b53a4ac388f3bd6f68e53 +Author: Peter Hutterer +Date: Mon Dec 1 21:14:01 2008 +1000 + + Xext: clean up XGE macros. + + Signed-off-by: Peter Hutterer + +commit 1ae529bef50a8ccc4a8a264ed120aeae544b21e7 +Author: Peter Hutterer +Date: Wed Jan 14 14:56:10 2009 +1000 + + dix: replace manual check for event types with IsPointerEvent() + +commit 1d54479cb3c8b4f75b7564f8b5e1c5da940b20f4 +Author: Simon Thum +Date: Tue Nov 25 14:12:02 2008 +1000 + + xfree86: init properties when ptraccel configuration is found. + + Signed-off-by: Peter Hutterer + Signed-off-by: Simon Thum + +commit 19275ea8e9dd93d5e61906943706dfe49003dd9c +Author: Simon Thum +Date: Wed Nov 19 16:01:21 2008 +1000 + + dix: add property support for pointer acceleration. + + Note: properties don't need to be cleaned up, the DIX does it for us anyway. + Data that is stored in properties is cleaned up by the property system. + Handlers, etc. don't need to be unregistered while cleaning up, as they get + deleted when the device is removed anyway. + + Signed-off-by: Peter Hutterer + Signed-off-by: Simon Thum + +commit d645721170b1196e5064b397cfbffd1da8c79bb1 +Author: Peter Hutterer +Date: Thu Jan 8 11:53:30 2009 +1000 + + mi: ensure chained button mappings from SD -> MD (#19282) + + After copying the master event, flip the detail field to the mapped button of + the SD, not the physical button. This way if the SD has a mapping 1:3 and the + MD has a mapping of 3:4, a press on button 1 on the SD results in a core event + on button 4. + + X.Org Bug 19282 + + Signed-off-by: Peter Hutterer + +commit 639f289dcdbe00a516820f573c01a8339e120ed4 +Author: Michel Dänzer +Date: Tue Jan 13 13:05:32 2009 +0100 + + EXA: Declare glyph cache picture as component-alpha when necessary. + + Without this, rendering component-alpha glyphs may break without a mask. + + Fixes http://bugs.freedesktop.org/show_bug.cgi?id=19233 . + +commit f4e00b783d3f824058a5f140e6462142688f5116 +Author: Michel Dänzer +Date: Tue Jan 13 11:03:27 2009 +0100 + + Fix build with --enable-visibility --disable-xdmcp. + +commit d36adf52a2b2711d22b11105f7bd907d4493fb9b +Author: Peter Hutterer +Date: Mon Jan 12 15:29:36 2009 +1000 + + dix: fix WarpPointer calls for devices with custom valuator ranges (#19297) + + If the MD's lastSlave was a devices with custom axes ranges, then a + WarpPointer would position the cursor at the wrong location. A WarpPointer + request provides screen coordinates and these coordinates were scaled to the + device range before warping. + + This patch consists of two parts: + 1) in the WarpPointer handling, get the lastSlave and post the event through + this device. + 2) assume that WarpPointer coordinates are always in screen coordinates and + scale them to device coordinates in GPE before continuing. Note that this + breaks device-coordinate based XWarpDevicePointer calls (for which the spec + isn't nailed down yet anyway) until a better solution is found. + + X.Org Bug 19297 + + Signed-off-by: Peter Hutterer + +commit 488d45295105daf10ccd17ca93ae6a6f4d0104f1 +Author: Peter Hutterer +Date: Mon Jan 12 16:16:24 2009 +1000 + + dix: EnqueueEvent and PlayReleasedEvent need to handle DeviceMotionNotifies + + No MotionNotify events in the processing anymore, so let's have them treat DMN + instead. + + Reported by Thomas Jaeger. + +commit 1e470645983ad18fa38c0885d674bc9d1dc1e560 +Author: Julien Cristau +Date: Tue Jan 13 00:12:56 2009 +0100 + + require randrproto 1.2.99.3 + +commit a48c81dcdf569a3f634ac23e08d2491354de6a36 +Author: Peter Hutterer +Date: Fri Dec 5 16:24:57 2008 +1000 + + Xi: add XATOM_FLOAT to server-defined properties. + + This property is used to denote type float for input properties. Such + properties can be accessed easily through the XIPropToFloat() function. + + Code originally written by Simon Thum. + + Signed-off-by: Peter Hutterer + +commit 669f6810af9a89187d6149841925fe765f3988ff +Author: Peter Hutterer +Date: Wed Nov 19 15:50:57 2008 +1000 + + Xi: add XIPropToInt() auxiliary function. + + Converts an XIPropertyValuePtr to an integer, provided that type and format is + right. + + Code originally written by Simon Thum. + + Signed-off-by: Peter Hutterer + +commit 56efbc0986e782da45addb05ece9f456d41d7a90 +Author: Peter Hutterer +Date: Fri Jan 9 13:46:20 2009 +1000 + + dix: drop x/y back into last.valuators before updating the history (#19285) + + positionSprite needs to scale to screen coordinates and in the process of + doing so alters dev->last.valuators[0:1]. Drop the real coordinates back after + finishing and before updating the motion history. This way, we don't push the + screen coordinates into the motion history. + + X.Org Bug 19285 + +commit 67710ad1d172cc4d53e28e5cc7dab2efb57a9a85 +Author: Jeremy Huddleston +Date: Sun Jan 11 16:34:36 2009 -0800 + + XQuartz: Don't need explicit Activate/EnableDevice in InitInput + (cherry picked from commit b1d29784410b3b93037e5636f336ba608d8ad6e3) + +commit dde5c591be74139ec6f2a90a3fdc7904d7342554 +Author: Jeremy Huddleston +Date: Sun Jan 11 16:10:45 2009 -0800 + + mieq: Avoid possible race condition whereby one thread might call mieqEnqueue before InitAndStartDevices finishes + + This is more of a hack around the problem. This is something that will need to be addressed in a + more structured manner with the multi threaded input efforts. + (cherry picked from commit 3c596c061e75848cfa76dd9259c23a3f3a67444c) + +commit 0720816aea20971ee94b5df3faa9cb9f16e93f82 +Author: Jeremy Huddleston +Date: Sun Jan 11 14:49:39 2009 -0800 + + XQuartz: Only call DarwinUpdateModKeys when needed + + Previously, we were calling it on almost every itteration through sendX11Event + (cherry picked from commit 6461729647ff4441d80811e73f0c0d2f108f2700) + +commit c8bf582db8111f71b9250a602a9d5aa7bd828136 +Author: Jeremy Huddleston +Date: Thu Oct 23 11:41:22 2008 -0700 + + XQuartz: Comment explaining the 0x10 flag for [e data2] + (cherry picked from commit 3c695280641c1205b97d3bb9f1d5e15a19cfa45f) + +commit 5339c22a8ca8bce6cd2b51e97c904292e7c40e74 +Author: Jeremy Huddleston +Date: Sun Jan 11 01:56:45 2009 -0800 + + XQuartz: Add locking to make mieq thread safe on OSX + (cherry picked from commit 7a8d2266861e74176b5310b83652a9c10a170494) + +commit 1beff89810cd193e45d437846929b3938cee6d96 +Author: Jeremy Huddleston +Date: Sun Jan 11 01:55:26 2009 -0800 + + XQuartz: misc 1.6 updates (still --disable-glx) + rlAccel is not longer compatable, and it's not worth fixing + Don't override DeviceCursorInitialize with a noop + Don't do a SwitchCoreKeyboard (which wasn't even needed in the first place) + (cherry picked from commit c137f681680e1d04b1513a8be68aeda4d1c56fd5) + +commit 2d917cfe56af1e74205f80ddbaddfb5c6ce82f0a +Author: Jeremy Huddleston +Date: Sat Jan 10 01:39:08 2009 -0800 + + Apple: Don't use DRI2 + (cherry picked from commit a1d35cee5907a76977ee43c49cb55c8f411c9794) + +commit 7a8abb9c8554d7967d912dcd427804f65906dc5c +Author: Jeremy Huddleston +Date: Tue Jan 6 15:32:44 2009 -0800 + + XQuartz: Bundle version 2.4.0 + (cherry picked from commit 3a3ccf5354e4275abe9d01f41a92602df5d690dc) + +commit 63ff7d236f67e159943312e1a551fcf5bb3f463a +Author: Julien Cristau +Date: Sun Jan 11 08:45:24 2009 +0100 + + os: don't mix declarations and code + +commit f43a3c61ab1eab8aecdc0e4bdee9f3bae2cab234 +Author: Julien Cristau +Date: Sun Jan 11 08:44:53 2009 +0100 + + os: backtrace() returns int, not size_t + +commit 5321ae76cffc5fcfe4bebcfb10e08eecaf640503 +Author: Julien Cristau +Date: Sun Jan 11 08:44:29 2009 +0100 + + os: ANSI cleanups + +commit 47438a2161a92fec08081d2de28966cf8d31654a +Author: Julien Cristau +Date: Sun Jan 11 08:43:48 2009 +0100 + + randr: RRSetPrimaryOutput can be static + +commit 7f82114b1dc738f98b9f22f6e61d6d5191383b67 +Author: Julien Cristau +Date: Sun Jan 11 08:43:35 2009 +0100 + + xkb: ANSI cleanup + +commit 46f3c1b937002f9a860878b359dcf5756a82dcf2 +Author: Julien Cristau +Date: Sun Jan 11 08:42:57 2009 +0100 + + mi: don't mix declarations and code + +commit 5aa9483d3ee306a79a9235610ce6b2754d697585 +Author: Julien Cristau +Date: Sun Jan 11 08:42:28 2009 +0100 + + mi: ANSI cleanups + +commit 359ed32c9b013ed88e14c99aa7fd7a617a85e2c5 +Author: Julien Cristau +Date: Sun Jan 11 08:40:58 2009 +0100 + + xnest: only define ddxBeforeReset if needed + +commit 9651cd8d7ac26635b6fb1826806ac9f4e2bddb63 +Author: Julien Cristau +Date: Sun Jan 11 08:40:47 2009 +0100 + + xnest: ANSI cleanups + +commit 31439195c51f183f5a8989a8d4f95b3e6326eda0 +Author: Julien Cristau +Date: Sun Jan 11 08:37:06 2009 +0100 + + xfree86: linuxPci.c needs a prototype for xf86AccResFromOS + +commit a00dbd0015845e08ef2af9480cc3833819cf86f5 +Author: Julien Cristau +Date: Sun Jan 11 08:36:16 2009 +0100 + + xfree86: use %zu to format size_t arguments + +commit 17c8cb50990a9ad20bbd3f8ce14c149dac9d72ad +Author: Julien Cristau +Date: Sun Jan 11 08:34:56 2009 +0100 + + xfree86: don't mix declarations and code + +commit aec4c0caca41a06832536ba82ba5bcdccf4a0aea +Author: Julien Cristau +Date: Sun Jan 11 08:33:18 2009 +0100 + + xfree86: ANSI cleanups + +commit dba3b1944d04232f88bb310277639f2c1adcbe08 +Author: Julien Cristau +Date: Sun Jan 11 08:28:23 2009 +0100 + + Xvfb: ANSI cleanups + +commit a2a760e33c17e211650b1f53c05a89c013916cf3 +Author: Julien Cristau +Date: Sun Jan 11 08:25:39 2009 +0100 + + dix: don't mix declarations and code + +commit 466dddbb83aa37ae93fb32976d51ecaa947c3616 +Author: Julien Cristau +Date: Sun Jan 11 08:25:17 2009 +0100 + + dix: fix cast from pointer to integer + +commit 54332e4732577dc49e440d9c9f8dd7d791f03f8e +Author: Julien Cristau +Date: Sun Jan 11 08:20:06 2009 +0100 + + dix: ANSI cleanups + +commit ef9902652b8a1c716ea3851565d6e2a98bd30eb0 +Author: Julien Cristau +Date: Sun Jan 11 08:18:56 2009 +0100 + + config: ANSI cleanups + +commit e64473b14e091e4b75e6ac2bf9a35e9aaa907258 +Author: Julien Cristau +Date: Sun Jan 11 08:17:41 2009 +0100 + + Xi: fix missing declaration of XkbSetRulesDflts + +commit 6e20a4ae731b7bc72266e5809faa0ecfde2bdfe4 +Author: Julien Cristau +Date: Sun Jan 11 08:16:55 2009 +0100 + + xsync: make SyncAlarmCounterDestroyed static + +commit 292311b421e53ead53c7df67d735a3d6d3e5b515 +Author: Julien Cristau +Date: Sun Jan 11 08:16:12 2009 +0100 + + Xext: ANSI cleanups + +commit 7a05c8b1e70680ddd3b3e09ad448788f8d70a428 +Author: Julien Cristau +Date: Tue Dec 23 13:08:49 2008 +0100 + + xfree86/linux: fix log flood on acpid open error + + When we can't open the acpid socket, warn once, not once every second + +commit 0d8184de591c75794083c8a8ca3fb8459e9394bc +Author: George Staplin +Date: Fri Jan 9 16:08:26 2009 -0700 + + XQuartz: GL: Set the __GLXconfig renderType to GLX_RGBA_BIT. + (cherry picked from commit 3c14546f58f8a138fe67c9cacc3bd0b7fa90c29a) + +commit 9c3975d63e129ad7230cae585f08a962d45fb861 +Author: George Staplin +Date: Fri Jan 9 15:53:39 2009 -0700 + + XQuartz: GL: Make many more fbconfigs and visuals available for the 1.5 branch. + + Use a __GLXconfig linked list struct to store the configurations for the + fbconfigs and visuals in a pGlxScreen. + + Also, remove the __GLXvisualConfig/GlxSetVisualConfigs code that isn't used + anymore. There is more code we can remove later, but I want to do that in + separate commits. + (cherry picked from commit 94162b0f8a25267aca280d25e216cc6bde47da6e) + +commit a68373994f4957a73f4eab52c90ebcfb96f285f8 +Author: Jeremy Huddleston +Date: Wed Jan 7 18:27:21 2009 -0800 + + XQuartz: Update padding in appledristr. This shouldn't change how the struct ends up in memory, but should make it more obvious for human eyes + (cherry picked from commit f7b6c1907c326a1befd8b5c8a1d311d551e8670f) + +commit 5f3188228eb988bd8f08b62c84f98a8ff66ee283 +Author: Xiang, Haihao +Date: Tue Jan 6 10:03:19 2009 +0800 + + avoid a potential endless loop. + + Previously it is possible that creating rotation data, then cleaning + up and creating again so that pScreen->BlockHandler and + xf86_config->BlockHandler all point to xf86RotateBlockHandler. + See bug #19343. + +commit b2756a71a432f7cf7c870a48676c98625512558d +Author: Thomas Jaeger +Date: Fri Jan 9 02:02:38 2009 -0500 + + Xext: Send out correct events in ProcXTestFakeInput + + Signed-off-by: Peter Hutterer + +commit 9bf77569cc8f9b0cc25c99487672d375d7c32f99 +Author: Peter Hutterer +Date: Thu Jan 8 16:32:24 2009 +1000 + + Xi: call CheckMotion for floating SDs too. + + CheckMotion updates the sprite position so we need to call it for all devices + that have a sprite - including floating SDs (which have an invisible sprite). + +commit 83eb863e3cfcd62549b836f3195107d8908b36d0 +Author: Peter Hutterer +Date: Fri Jan 9 11:39:02 2009 +1000 + + Xext: don't accept DeviceValuator if the dev doesn't have valuators (in xtest) + + Reported by Chris Ball. + +commit a3f9e887d9ff5fdee6c35ccc0f7b324723a83a46 +Author: Peter Hutterer +Date: Fri Jan 9 11:40:18 2009 +1000 + + dix: don't accept Button 0 presses in GPE. + +commit 0c6987df3b9b3a37d201d740d8248c326449835e +Author: Adam Jackson +Date: Tue Jun 17 16:10:51 2008 -0400 + + XAA: Disable offscreen pixmaps by default. + + Say Option "XaaOffscreenPixmaps" to turn them back on. + + Apropos of bugs #13795 and #15098. But this still isn't correct. + +commit eb2d7b3d700952ba88c77deacf687b251300e660 +Author: Peter Hutterer +Date: Fri Dec 19 21:05:47 2008 +1000 + + dix: move focus handling into enterleave.c. + + This commit moves the focus handling from events.c into enterleave.c and + implements a model similar to the core enter/leave model. + For a full description of the model, see: + http://lists.freedesktop.org/archives/xorg/2008-December/041740.html + + This commit also gets rid of the focusinout array in the WindowRec, ditching + it in favour of a local array that keeps the current focus window for each + device. + + Signed-off-by: Peter Hutterer + +commit 673eb23aac578dcdc04e2a99d1fa5c2987eb58b8 +Author: Peter Hutterer +Date: Fri Dec 19 21:05:09 2008 +1000 + + dix: add a few auxiliary functions for the updated focus model. + + SetFocusIn and SetFocusOut, including the static array to keep all focus + windows. + + Signed-off-by: Peter Hutterer + +commit 38b28dcadd0990cb43f50db4300eebb8f044db96 +Author: Peter Hutterer +Date: Thu Dec 18 15:55:11 2008 +1000 + + dix: reduce FirstPointerChild complexity + + Instead of keeping a flag on each window for the devices that are in this + window, keep a local array that holds the current pointer window for each + device. Benefit: searching for the first descendant of a pointer is a simple + run through the array. + + Signed-off-by: Peter Hutterer + +commit 72ad4a85cc0ffe60a90011d65ef718d5852beae4 +Author: Peter Hutterer +Date: Thu Dec 18 15:43:18 2008 +1000 + + dix: remove now unused "exclude" parameter from FirstPointerChild + + Signed-off-by: Peter Hutterer + +commit 0aa4460c3b4f9bb17d5412d76fa8c4c501132429 +Author: Peter Hutterer +Date: Wed Dec 17 16:14:41 2008 +1000 + + dix: re-implement enter/leave model. + + The old model was implemented based on a misunderstanding of NotifyVirtual and + NotifyNonlinearVirtual events. It became complicated and was broken in some + places [1]. This patch wipes this model completely. + + A much simplified implementation is provided instead. Rather than a top-down + approach ("we have a tree of windows, which ones need to get which event") + this one uses a step-by-step approach. For each window W between A and B + determine the pointer window P as perceived by this window and determine the + event type based on this information. This is in-line with the model described + by Owen Taylor [2]. + + [1] http://lists.freedesktop.org/archives/xorg/2008-December/041559.html + [2] http://lists.freedesktop.org/archives/xorg/2008-August/037606.html + +commit f8ecc2f08fc375b14d60a74e2fdd7830bfdcd74f +Author: Christian Beier +Date: Tue Jan 6 11:31:03 2009 +1000 + + Xi: fix typo in WarpDevicePointer handling. + + Signed-off-by: Peter Hutterer + +commit 3d549438c29004d78032ecc50ab45ca0e3f49623 +Author: Thomas Jaeger +Date: Mon Jan 5 01:26:42 2009 -0500 + + Don't alter device button maps in DoSetPointerMapping + + Currently, if a device map differs from the core pointer map, then the + request may return MappingBusy, even though all the affected core + buttons are in the up state. + + Signed-off-by: Peter Hutterer + +commit b72c6b083baeadfd7b18b6025df054be502d6e28 +Author: Peter Hutterer +Date: Tue Jan 6 09:01:48 2009 +1000 + + dix: remove inputInfo.keyboard reference in QueryPointer handling. + + PickKeyboard() is to be used instead. + +commit 30d2cfcd3851870178d62e5067211aa36f87fbd2 +Author: Kim Woelders +Date: Mon Jan 5 19:35:18 2009 +0100 + + dix: Fix handling of do_not_propagate_mask window attribute. + + This was broken in 32aa252e988be8cbfd4f7e373fb7b7736ef1f5f2. + + Signed-off-by: Kim Woelders + Signed-off-by: Peter Hutterer + +commit 515ce3e4ba42605a1ee9979e8bb5acd3cf6470a3 +Author: Peter Hutterer +Date: Mon Jan 5 15:48:45 2009 +1000 + + xkb: fix typo - missing negation when checking button state. + + Introduced with a85f0d6b98237d8a196de624207acf1983a1859a. + + Reported by Thomas Jaeger. + + Signed-off-by: Peter Hutterer + +commit 9a313fb03153eb764aab3e9cb9988b6603541637 +Author: Paulo Cesar Pereira de Andrade +Date: Wed Jan 7 23:07:01 2009 -0200 + + Don't call free, use xfree macro. + + There are several other direct calls to free, check + % egrep '\ +Date: Wed Jan 7 19:37:03 2009 -0200 + + Default to use standard bitmap fonts, with builtins as fallback + + The builtin-fonts configure option was removed, as it at best should + have been a runtime option. Instead, now it always register all "font + path element" backends, and adds built-ins fonts at the end of the + default font path. + This should be a more reasonable solution, to "correct" the most + common Xorg FAQ (could not open default font 'fixed'), and also don't + break by default applications that use only the standard/historical + X Font rendering. + +commit 4f004231871c830e16c4525cfaab183fad0aee86 +Author: Jeremy Huddleston +Date: Wed Jan 7 10:55:08 2009 -0800 + + XQuartz: Fat binary buildfixes + + Let Xarch.h set X_BYTE_ORDER for us. + Determine _XSERVER64 at build time. + (cherry picked from commit f54736c2b0b537c8322ecb2db8e9cee5eea68165) + +commit fbf44d85bd59bfa629193d6494b16812a7343cab +Author: Jon TURNEY +Date: Mon Jan 5 23:18:11 2009 +0000 + + Cygwin/X: Tidy up an unused #define + + Signed-off-by: Jon TURNEY + +commit 1badeb78b0b974a381e563e3d4f4cf37debd4266 +Author: Rémi Cardona +Date: Wed Jan 7 09:31:17 2009 -0500 + + Include Xmd.h from edid.h + +commit fc14d52297f8f408693ed867ab649ea5b210b0b3 +Author: Jeremy Huddleston +Date: Sun Nov 16 15:35:17 2008 -0800 + + XQuartz: Use AvailabilityMacros.h instead of availability.h for Tiger + (cherry picked from commit 91f3c3435716195e6d606c8e2f6b0de52eb5972d) + +commit 4c326c4d342bede0de57e71c1431f56188f812ce +Author: Keith Packard +Date: Tue Jan 6 08:36:39 2009 -0800 + + Make cvt complain about invalid arguments more often. + + cvt doesn't correctly parse 'cvt -vr 1920 1080'. This patch doesn't fix + that, but it does at least generate an error message now. + + Signed-off-by: Keith Packard + +commit 1da4a41b0881b605c7cfa4c4bf637efa132c58fc +Author: Paulo Cesar Pereira de Andrade +Date: Tue Jan 6 05:01:32 2009 -0200 + + Add sdksyms.sh to EXTRA_DIST to correct make distcheck + + Thanks to for noticing the problem. + +commit 200230535fa80c04faaa562bbbd8272713e2763b +Author: Paulo Cesar Pereira de Andrade +Date: Mon Jan 5 16:49:57 2009 -0200 + + Update sdk headers to export new symbols. + + All symbols in installed sdk headers should be explicitly tagged + as exported symbols. Otherwise, to ensure it is not a mistake, one + could write it as something like: + extern /* NOEXPORT */ type name ...; + but the proper procedure really should be to use a non sdk header + (or a "noinst_" one). + This patch also removes prototypes to some functions that existed + only temporarily. + +commit af6152499654a7146250d4ad0df89f611345434e +Author: Jon TURNEY +Date: Wed Dec 17 14:15:34 2008 +0000 + + Cygwin/X: Add *.exe pattern to .gitignore + + Signed-off-by: Jon TURNEY + +commit 8428b92fcb478b041f778c0e8c66fdeb96c807f3 +Author: Colin Harrison +Date: Sun Jan 4 20:12:16 2009 +0000 + + Xming: Fix a couple of warnings + + Signed-off-by: Jon TURNEY + +commit f6f97419bd2db651384ecaec0b296840810463a4 +Author: Jon TURNEY +Date: Mon Jan 5 18:35:52 2009 +0000 + + Cygwin/X: Add -static back to linker flags + + Add -static back to linker flags to avoid screwing things up for + people who don't know that they need to avoid dynamically linking + with libXfont, as weak symbol handling requires static linking to + work correctly on Cygwin... + +commit 2ec0ac08addb938a3d5aa5b9888ce7c79a0daada +Author: Jon TURNEY +Date: Wed Dec 17 19:18:38 2008 +0000 + + Cygwin/X: Fix some remaining build issues + + Fixes for a few remaining build issues, omitted from commmit 6e85a8304b50c0bfd16a400a05d257f0168fb99b in error + + Provide VENDOR_RELEASE in XORG_VERSION_CURRENT for XWin + Remove -static from linker flags for XWin + + Signed-off-by: Jon TURNEY + +commit e351d10773742d1570e1111976b7c8b6adda5ae6 +Author: Daniel Stone +Date: Mon Jan 5 17:41:38 2009 +0200 + + RandR: Only export Xinerama symbols when building Xinerama support + + Otherwise compilation fails due to sdksyms. + + Signed-off-by: Daniel Stone + +commit 3387cbb7c13636eb62b22640d28df7011c2268f0 +Author: Daniel Stone +Date: Mon Jan 5 11:25:57 2009 +0200 + + XFree86: Xv: Add ClipNotify helper + + Add a ClipNotify helper that lets the driver know about changes in the + clipping of an Xv backing drawable. + + Signed-off-by: Daniel Stone + +commit 55c4ec0a1ef78481b4d82153c19c20fbf88f1624 +Author: Daniel Stone +Date: Mon Jan 5 11:24:24 2009 +0200 + + XFree86: fbdevhw: Add helper function to get fd + + Most devices need to do custom weird ioctls, so let them know what the + fd is. + + Signed-off-by: Daniel Stone + +commit 7da4e901e078e5597b8dbcb7ee39b949edf7ef03 +Author: Daniel Stone +Date: Mon Jan 5 11:23:39 2009 +0200 + + XFree86: Fix build with DGA disabled + + It's optional, so we might as well work when it's disabled. + + Signed-off-by: Daniel Stone + +commit 13ce7b98d48261b061da5768a6405bdbe9caa1fb +Author: Dan Nicholson +Date: Mon Dec 22 14:31:03 2008 -0800 + + dmx: Require newer libXi for XQueryInputVersion + + Signed-off-by: Dan Nicholson + Signed-off-by: Peter Hutterer + +commit 27011254c4de4e573a0851bf46892fb488db6522 +Author: Peter Hutterer +Date: Tue Dec 23 09:59:28 2008 +1000 + + xfree86: If an input device failed to activate, return immediately. + + Devices are only activated once - right after they've been added to the + server. If a device failes activation, it's dead. There's no reason to + continue. Return the error code from ActivateDevice() without setting up + sprite information or even sending a event to the client. + + Then - in the DDX - just remove the device again. + + Signed-off-by: Peter Hutterer + +commit e1a3a1a0d85c9971aea65c2228b5fd4dbf3bf57a +Author: Peter Hutterer +Date: Tue Dec 23 10:00:34 2008 +1000 + + xfree86: don't call CheckMotion if a device hasn't been enabled. #19176 + + X.Org Bug + +commit af820259a06a6e8dbd299fe22347aa83b0c985b0 +Author: Jeremy Huddleston +Date: Fri Jan 2 11:06:12 2009 -0800 + + XQuartz: Copyright Update... happy new year + (cherry picked from commit 9f1c26ae8d9ccf81cba0ab946326968a701f8c36) + +commit d8c9ba37abd0d63ea2c5394af434b46ec3b647fd +Author: Jeremy Huddleston +Date: Fri Jan 2 10:58:40 2009 -0800 + + XQuartz: pbproxy: Push dpy init and CFRunLoop hook setup into the pbproxy thread to avoid possible deadlock + (cherry picked from commit 49e59d32b88e4fad070f230b5efaa261b47f78db) + +commit 2f361186f93a791e446c4a9471bb0c3b9b818d98 +Author: Jeremy Huddleston +Date: Wed Dec 31 12:40:02 2008 -0800 + + XQuartz: Don't use NX_SECONDARYFNMASK, NX_NUMERICPADMASK, NX_HELPMASK + + We don't have keycodes for them, so don't try to use them + (cherry picked from commit cb912aca3a2834c4ad8e386c8a0d05c1bb31b0e7) + +commit a2abaa9fd3a5ad713c1f946e9d7f598825ad3a84 +Author: Jeremy Huddleston +Date: Wed Dec 31 11:57:49 2008 -0800 + + XQuartz: Make sure to reset the saved key state when deactivating X11.app + (cherry picked from commit 3eef78eb321f4f7dbca5a10c80666c621e28a1e0) + +commit 3986b683adc4f6f967cc5f643e973d89764234c7 +Author: Jeremy Huddleston +Date: Wed Dec 31 12:44:43 2008 -0800 + + XQuartz: Honor system key repeat rate + (cherry picked from commit 4303c9be39a86f5a21de108f72b90a989435905e) + +commit 1a69d56afda082298edc76271d2cb71af00526ce +Author: Jeremy Huddleston +Date: Mon Dec 29 19:54:12 2008 -0800 + + XQuartz: Better avoid stuck keys on context switches + (cherry picked from commit e9963f1a4f4f12f253eae9d4f01694b6cabe35ad) + +commit f464b5a9164b24f74adea198bcabe9affaf5552c +Author: Jeremy Huddleston +Date: Mon Dec 29 19:38:31 2008 -0800 + + XQuartz: Workaround OSX VNC server bug for modifier key state + + A better approach which ensures we have a L modifier key down if we are told neither are down and atleast one is down... =/ + (cherry picked from commit bc13dda345f716bf4de9bfe4e1d85969263b60c2) + +commit 3ead9aa2602cab9c883aa7bcbd2a8d82c5c88f5a +Author: Jeremy Huddleston +Date: Mon Dec 29 00:30:31 2008 -0800 + + XQuartz: Updated man page fullscreen_hotkeys fullscreen_menu + (cherry picked from commit 00f3a2e33c8e91da5f855821313a04ea97445656) + +commit 48dbaf173a82693fd72953983da9fc556cd1c6ed +Author: Daniel Stone +Date: Tue Dec 30 12:17:14 2008 +1100 + + XKB: Also copy keyboard feedback when copying the keymap + + When updating the XKB keymap, make sure the keyboard feedback is also + copied, to preserve autorepeat settings etc. + + Signed-off-by: Daniel Stone + +commit cf85a32b5f80694f3c2f434e91cf77edb33a7c17 +Author: Jeremy Huddleston +Date: Sun Dec 28 15:57:59 2008 -0800 + + XQuartz: pbproxy: We explicitly need libX11 for pbproxy + (cherry picked from commit 32824120da0749b8369c2592f851bb1030dea9f1) + +commit 997e8c321bb11067591785ba38103a3a5029301d +Author: Jeremy Huddleston +Date: Sun Dec 28 01:53:38 2008 -0800 + + XQuartz: Try harder to get the user's login environment + (cherry picked from commit 6bedaddd78aa04bd303df434a4c49bb87bd7deaa) + +commit ae8dc47fd910ec4b9edd64bf3b7ba33e212c5942 +Author: Jeremy Huddleston +Date: Wed Dec 24 00:39:06 2008 -0800 + + XQuartz: Reposition windows when we enter fullscreen to ensure our root window + (cherry picked from commit 7617d3659b5481ef85aecc1f936e7dd2662bdfbd) + +commit 895186e026881196ac9383aa165e123055d9cf09 +Author: Jeremy Huddleston +Date: Tue Dec 23 20:33:27 2008 -0800 + + XQuartz: Re-enable rlAccel + + It was incorrectly blamed for some crashes a few months ago, and it should be + safe to use until we get an EXA driver to replace it. + (cherry picked from commit 8121f30bd29591fc74fbc680fbbf20210271fa58) + +commit dc4498ec5cb7f25f710bdec8793746da2d2324ed +Author: Jeremy Huddleston +Date: Sun Dec 28 22:16:52 2008 -0800 + + Updated .gitignore + (cherry picked from commit 06964c322e4da81f3146022049a2de13f6b2892d) + +commit 3534a5e5d9c5af85149c799f324257f89507fa23 +Author: Maarten Maathuis +Date: Fri Dec 26 16:38:27 2008 +0100 + + exa: Allow drivers to set non-NULL devPrivate.ptr for !offscreen pixmaps. + +commit 027b440d4f9f0cdd46addff46fd2d5c44cd5c847 +Author: Maarten Maathuis +Date: Fri Dec 26 15:51:55 2008 +0100 + + exa: preparing as source and finishing access as mask is a bad idea + +commit 89bbd07e9ddd688df81405151f9c6284fa3367a8 +Author: Alan Coopersmith +Date: Tue Dec 23 21:40:21 2008 -0800 + + Remove obsolete XF86_DATE & xf86Date.h + +commit 8dc3a7d6817b7371bbfece1108ef856788256853 +Author: Alan Coopersmith +Date: Tue Dec 23 14:59:46 2008 -0800 + + Remove unused CLOG_DATE (leftover from pre-git ChangeLog) + +commit fe05a46f191ebd36251dda7a9703d90c25efec11 +Author: Alan Coopersmith +Date: Tue Dec 23 14:55:26 2008 -0800 + + Remove powerpc load/store functions from export list on other platforms + +commit a1b201bcb01e702f8f0a5889a3d49317503b9475 +Author: Jeremy Huddleston +Date: Tue Dec 23 12:39:38 2008 -0800 + + XQuartz: Use depth=24 instead of FatalError if we can't figure out our depth + (cherry picked from commit 80b65c5b78d125c4ad3620b87b121c9e666299c3) + +commit 86dc660588a615baefb1799d78a501c95a931d77 +Author: Paulo Cesar Pereira de Andrade +Date: Tue Dec 23 18:07:54 2008 -0200 + + Improve sdksyms.c automatic generation (Fix #19245). + + Since it is already parsing cpp output, create a dependency file + in the same process. This will cause sdksyms.c to be regenerated + whenever a sdk header is modified. + This also uses the gmake 'sinclude' directive (don't fail if + included file doesn't exist). This should not cause any problems + given that gmake only constructs are used in several other Makefiles. + +commit 477c32708718f4470ac2e786b2446d7a44e45b9c +Author: Peter Hutterer +Date: Fri Dec 19 08:56:35 2008 +1000 + + dix: don't disable uninitialized devices. + + If a device hasn't been initialized, it doesn't have a cursor yet. So don't + set the cursor to the NullCursor, and don't try to DisableDevice either. + + Signed-off-by: Peter Hutterer + +commit 25aac8b5793caa92116b766d5a9e12dbecdb0290 +Author: Peter Hutterer +Date: Tue Dec 2 16:16:58 2008 +1000 + + xkb: don't treat groups with different no of symbols as identical. + + Signed-off-by: Peter Hutterer + +commit f141c1b4bb9482f188783dd6f161d3c7960a3329 +Author: Peter Hutterer +Date: Tue Dec 2 14:54:50 2008 +1000 + + xkb: explicitly check for group replication in the core representation. + + Single-group keys may get replicated amongst all groups. Check explicitly for + this case and squash it down to one group. + + Signed-off-by: Peter Hutterer + +commit 70a977c021e107c4fabe46ec2f619be9fb55abeb +Author: Peter Hutterer +Date: Tue Dec 2 15:36:51 2008 +1000 + + xkb: don't replicate past the number of groups we have. + + Signed-off-by: Peter Hutterer + +commit a157575eeeb523cd43197c5caeb00cb3d56f9988 +Author: Peter Hutterer +Date: Tue Dec 2 13:31:58 2008 +1000 + + xkb: ensure enough symbols for core Group1 replication. + + A single-group key on a multi-group keyboard has to be replicated across all + three groups (see Section 12.4 of the XKB protocol spec). Ensure that there's + enough symbols available to actually do that. + + e.g. a key ABCD on a 3 group keyboard needs to be replicated as ABABCDCDABCD, + hence requiring space for 12 symbols, even if maxSymsPerKey is less than that. + + Signed-off-by: Peter Hutterer + +commit b5736d237a21d5e65d839c4d213dd3bda5a11e9d +Author: Maarten Maathuis +Date: Mon Dec 22 19:53:14 2008 +0100 + + randr/xfree86: Fix initial gamma computation. + + - The previous version overflowed sometimes. + +commit ba9dc353a60e9a6bba64e854b23271011bff7015 +Author: Jeremy Huddleston +Date: Sun Dec 21 21:03:59 2008 -0800 + + XQuartz: Don't use keycode 0 to determine !swallow since our most common key to swallow is actual keycode 0 (a) + (cherry picked from commit 33f43a7f03023bfbab25a957cb81fc25b4afa4ca) + +commit f0cdccd1e22198f4f5e1aa0d0617441b79ffbc44 +Author: Jeremy Huddleston +Date: Sun Dec 21 20:40:36 2008 -0800 + + XQuartz: Update our "screens" when we toggle rootless rather than when we toggle fullscreen + + This old behavior was used as a workaround for the menubar behavior in the older server, + but we handle it better now and need to update our screens when we toggle the rootless + state instead. + (cherry picked from commit 508aa95bc2cd3fdc3dff448ec090919bf807d153) + +commit 8f04a5237384fa97ff960b4607a79b688ea03fc1 +Author: Jeremy Huddleston +Date: Sun Dec 21 14:54:11 2008 -0800 + + XQuartz: Run applications via '/bin/sh -c ...' to support users who expect shell parsing + (cherry picked from commit 67455e716e3ecffd528930479192785958d37988) + +commit 029d255a654eca6401c374d145f63bbb923554b5 +Author: Jeremy Huddleston +Date: Sun Dec 21 14:43:12 2008 -0800 + + XQuartz: pbproxy: Release display notification lock when not needed to avoid deadlock + (cherry picked from commit 22f664ab95a0cae981e9cefad6f075d051583ca5) + +commit b1f166f2981595aea6f8e2a336d6d25436f8a07d +Author: Jeremy Huddleston +Date: Sun Dec 21 14:42:26 2008 -0800 + + XQuartz: update quoting in case X11.app is moved to a directory with a space. + (cherry picked from commit cc805dc799efa37c8dcefa3db04d87e9b835ffbd) + +commit 2f47bda645eb43c60b71315e8d265190acce481d Author: Jeremy Huddleston Date: Thu Dec 18 09:14:16 2008 -0800 - XQuartz: Changed X11.sh to allow use of a ~/.x11run as requested by users of alternate shells - (cherry picked from commit b62ed1f8eaf041a946bb591165bb18ee481dedbf) - (cherry picked from commit c3812aec973b7341a600e5d2d07d5a7f15abd609) + XQuartz: Changed X11.sh to allow use of a ~/.x11run as requested by users of alternate shells + (cherry picked from commit b62ed1f8eaf041a946bb591165bb18ee481dedbf) + +commit 83edcc08de4c815c5cbe815f521b3b728441cad6 +Author: Jeremy Huddleston +Date: Wed Dec 17 15:09:35 2008 -0800 + + XQuartz: Get rid of white rectangle bug + (cherry picked from commit 3269959033ed0c675a3a906666454df34086896a) + +commit 1335a300b9cd03208a583acfdb7f7153eaef4ea6 +Author: Jeremy Huddleston +Date: Wed Dec 17 14:43:02 2008 -0800 + + rootless: Make expose_1 static + (cherry picked from commit 60c8d2697036a125ca5381df8e2eaedabad4d242) + +commit 22b2811e6aa0bdbfa6a0682ac862063e52ec0c62 +Author: Jeremy Huddleston +Date: Wed Dec 17 13:54:53 2008 -0800 + + XQuartz: Name the startup shell script X11 for better compatability + (cherry picked from commit f84613d6fcbbb5f769ee34cff0900353a13693cf) + +commit 82fba7697fbbb6f66cc22424ba99755b99ff482e +Author: Maarten Maathuis +Date: Mon Dec 22 00:33:55 2008 +0100 + + xfree86/randr: Avoid a crash when xf86CrtcConfigPrivateIndex is -1. + +commit 1567b7243f4799808ab93fbd962df14ce3af2d49 +Author: Maarten Maathuis +Date: Sun Dec 21 00:08:56 2008 +0100 + + exa: A more correct fix. + +commit 2db7b66863ae6055c3ce13c88b36d620de8a4d75 +Author: Maarten Maathuis +Date: Fri Dec 19 23:12:37 2008 +0100 + + exa: a few cleanups + + - Some warnings silenced. + - Some whitespace cleanup. + +commit aedd2f566df585db7a1614f302cc8d3feda54275 +Author: Maarten Maathuis +Date: Fri Dec 19 19:10:23 2008 +0100 + + randr/xfree86: Fix a one off error in the panning calculations. + + - Example: mode 1280x1024, panned area 1281x1024 + panned_area.x2 = 1281 + mode.width = 1280 + If you substract 1280 from 1281, then that leaves you with one. + Which is the one pixel that you need to move to actually see the last pixel collumn. + Substracting 1 from this will consistently prevent you from seeing the right and bottom edge. + +commit 332d65ec7a6e94d75efe95d53742f137835274de +Author: Maarten Maathuis +Date: Fri Dec 19 18:59:27 2008 +0100 + + randr: Consider panned crtc's when calculating xinerama screen sizes. + + - This will allow window managers and applications to actually use the panned area. + +commit f2d1de6c2a3e6e50db6d9714c293eb26324fbd3d +Author: Adam Jackson +Date: Fri Dec 19 10:27:37 2008 -0500 + + xv: remove useless XVCALL macro + +commit 5a072c55350f4b73d911ea6a2aeddad844924834 +Author: Adam Jackson +Date: Fri Dec 19 09:51:52 2008 -0500 + + dmx: Fix calloc macro confusion. + +commit 8c488ac3b3990cd203baed7f2127b9bed8aab534 +Author: Alan Coopersmith +Date: Thu Dec 18 19:47:49 2008 -0800 + + Fix linking of Xorg with dtrace probes on Solaris + +commit 396433d0da721951e6e7abc02dc3b2f682495154 +Author: Alan Coopersmith +Date: Thu Dec 18 18:49:47 2008 -0800 + + Fix sdksyms.sh to work with Solaris/Sun compiler builds + + - Pass $(CPP) & $(AWK) settings from configure to sdksyms.sh + - Only reset sdk variable (tracks if header is part of sdk) if + a filename is included on the cpp # line, + since Sun compilers omit filename when it is unchanged from + previous line. + +commit aea6f19f25e13768b1d09fac4991d6a5e6c2cdac +Author: Peter Hutterer +Date: Tue Dec 16 17:10:03 2008 +1000 + + xfree86: don't restore the TTY mode if we didn't initialize it ourselves + + Restoring it unconditionally means we restore to whatever tty_mode has as + default value (i.e. 0). K_RAW happens to be 0x00, so we always restore to raw + mode if allowEmptyInput is off. + + Signed-off-by: Peter Hutterer + Signed-off-by: Adam Jackson + Signed-off-by: Julien Cristau + +commit 5d065a889074558634216eebd4bba35a60db9a09 +Author: Eamon Walsh +Date: Thu Dec 18 13:59:36 2008 -0500 + + xselinux: Use xace Xtrans wrappers instead of the now-inaccessible wrapees. + +commit f87e66486c3610c56888915b66ae5ab0af03da8b +Author: Eamon Walsh +Date: Thu Dec 18 13:58:35 2008 -0500 + + xace: Export wrappers around two Xtrans functions used by modules. + + Don't know a better way to do this, since Xtrans isn't a library that + can be linked into modules. + +commit 777408914dbf395e9969e3f34720286e6f740369 +Author: Eamon Walsh +Date: Thu Dec 18 12:35:57 2008 -0500 + + Add xace headers to the SDK when enabled and export the XaceHooks symbol + to modules. + +commit 9e4ef3cfe5ee178c266edd64a8edd767827aa0d7 +Author: Paulo Cesar Pereira de Andrade +Date: Thu Dec 18 15:00:25 2008 -0200 + + Use regex pattern understood by all known awk variants. + + Ubuntu uses mawk by default, but it doesn't understand posix character + classes (which are locale dependent, and this patch uses only valid C + identifiers). + Also make sure awk runs with LC_ALL=C to match the regex patterns. + +commit 3d3c234b434a3443a00c3be9f32c698bcced111c +Author: Maarten Maathuis +Date: Thu Dec 18 16:26:38 2008 +0100 + + randr: add some notes about the stuff that changed in driver ABI 2 + +commit 1b244feb4cf11d1cdba79fbead50035965535552 +Author: Maarten Maathuis +Date: Thu Dec 18 16:11:21 2008 +0100 + + randr: check for malloc failure + +commit 21a2df9f359584e634dc12d8602401d54defd7db +Author: Colin Harrison +Date: Sun Nov 2 20:46:44 2008 +0000 + + Xming,Cygwin/X: Fix crashes when using the DirectDraw '-refresh rate-in-Hz' option in -fullscreen (#11128) + + fd.o bugzilla #11128 + + Fix crashes when using the '-refresh rate-in-Hz' option in -fullscreen mode and using a DirectDraw engine. + On failure (typically 0x80004001 DDERR_UNSUPPORTED), use the default refresh rate for the driver, instead of the + entered value. + + Also fix some nearby error reporting + + Copyright (C) Colin Harrison 2005-2008 + http://www.straightrunning.com/XmingNotes/ + http://sourceforge.net/projects/xming/ + + Signed-off-by: Jon TURNEY + +commit b0ed1567c1782041cf3d14ec6dd4d622915c7d0c +Author: Colin Harrison +Date: Wed Dec 17 17:16:02 2008 +0000 + + Xming, Cygwin/X: Tidy up warnings, headers and unused variables (#11132) + + Tidy up various warnings, include headers and unused variables, etc. + + Signed-off-by: Jon TURNEY + +commit 15c4a6e088039e14b4f2387ff204cb5448916b39 +Author: Colin Harrison +Date: Wed Dec 17 16:44:07 2008 +0000 + + Xming, Cygwin/X: Fix crash in function winMultiWindowGetTransientFor() (#11147) + + The function winMultiWindowGetTransientFor() crashes, due to a bogus pointer + dereference, but this fact was previously hidden by the function not being called! + + Signed-off-by: Jon TURNEY + +commit 08a3d6928c87032998b6113d67f69b79b09eecee +Author: Peter Hutterer +Date: Tue Dec 16 14:19:12 2008 +1000 + + dix: Allocate the space for the DCCE when the last SD is removed. + + This still doesn't fix reset the MD's classes (a TODO that's been here for + ages), but at least we don't segfault anymore when detaching the last SD. + + Signed-off-by: Peter Hutterer + +commit 689be0cb4b259840208c17f974066a942196793a +Author: Peter Hutterer +Date: Tue Dec 16 14:19:38 2008 +1000 -commit 87df8e681988e546ed943b0f3e7df2851ada4712 -Author: Jeremy Huddleston -Date: Wed Dec 17 15:09:35 2008 -0800 + dix: set the correct length in CreateClassesChangedEvent. + + Fallout from aeff14d5f208d02f211b8b2a1a98999624e8c2cf. Yes, we don't malloc + anymore because we are inside a SIGIO and the memory is already there anyway. + But we still need to set the event length correctly, otherwise + mieqEnqueue/mieqProcessInputEvent don't know how much memory to copy. + + Signed-off-by: Peter Hutterer - XQuartz: Get rid of white rectangle bug - (cherry picked from commit 3269959033ed0c675a3a906666454df34086896a) - (cherry picked from commit fd6fb6a2771df152b57f9dfb159fa42a3b1d37cd) +commit 1001403301001fcc84ce6833938590956df40d2a +Author: Maarten Maathuis +Date: Thu Dec 18 00:19:06 2008 +0100 -commit 5ff6b1e7ed11a37b2878f2f9a37b2afce00baeb0 -Author: Jeremy Huddleston -Date: Wed Dec 17 13:54:53 2008 -0800 + Forgot to fix a comment. - XQuartz: Name the startup shell script X11 for better compatability - (cherry picked from commit f84613d6fcbbb5f769ee34cff0900353a13693cf) - (cherry picked from commit 102abeda37d6b62971a9952efa0453e38504ae0b) +commit 9fa15bef59881bdcf087889f16ab3c8d953da8f1 +Author: Maarten Maathuis +Date: Wed Dec 17 23:42:30 2008 +0100 + + randr: some improvements, fixes and crtc abi bump + + - Add active field to crtc. + - Set gamma (only) whenever a crtc becomes active. + - Check for xf86_config being NULL. + - Increase crtc abi to 3. + - A few other fixes. + +commit bf65523ab0b39774f07a7ae478ff3f5653fad469 +Author: Joe Krahn +Date: Sun Nov 2 20:17:03 2008 +0000 + + Cygwin/X: Fix for mis-aligned icon data creates bad background masks (#4491) + + fd.o bugzilla #4491 + + Transparent icon backgrounds appear with black stripes as mask data is incorrectly aligned. + Modified the icon DDB bitmap data alignment to 16 bits + + Signed-off-by: Jon TURNEY -commit ffdf0139051c51f94d6a7a9a5fc2ccc68959d14b +commit 91f73b79b7ae64e5b846d1efeb470bb61a913720 Author: Maarten Maathuis -Date: Fri Dec 19 19:10:23 2008 +0100 +Date: Wed Dec 17 16:56:26 2008 +0100 - randr/xfree86: Fix a one off error in the panning calculations. + randr: Improve per-crtc gamma support. - - Example: mode 1280x1024, panned area 1281x1024 - panned_area.x2 = 1281 - mode.width = 1280 - If you substract 1280 from 1281, then that leaves you with one. - Which is the one pixel that you need to move to actually see the last pixel collumn. - Substracting 1 from this will consistently prevent you from seeing the right and bottom edge. - (cherry picked from commit aedd2f566df585db7a1614f302cc8d3feda54275) + - The Gamma values from the monitor section are now used during initial config. + - The old colormap system is disabled when gamma set hook is available. + - Gamma values are now persistent for the lifetime of the xserver. + - This requires no driver changes and should be driver ABI compatible. + +commit 1556815d34cecb4b4b62d2a4ce813b1435a937ec +Author: Jon TURNEY +Date: Sun Nov 2 18:27:18 2008 +0000 + + Cygwin/X: Initialize native HWND atom when built !XWIN_MULTIWINDOWEXTWM + + Fix the native HWND atom identifier to be initialized correctly, when built without external WM support + + This is neccessary for native window naming to work correctly in multiwindow internal WM mode. + + Signed-off-by: Jon TURNEY + +commit 92082a0a00f05ecdc90814a8104828853c9c8f29 +Author: Jon TURNEY +Date: Sun Nov 2 18:27:15 2008 +0000 + + Cygwin/X: update to use standard DPMS stubs + + Bring Cygwin/X up to date with DPMS API changes by using the standard stubs + + Signed-off-by: Jon TURNEY + +commit 2850f26bb2999ef3fb2a10adb80fe358463aecb5 +Author: Jon TURNEY +Date: Sun Nov 2 18:27:14 2008 +0000 + + Cygwin/X: update for changes to shadow framebuffer + + Bring Cygwin/X up to date with the change to damage-based shadow framebuffer. + + Register the screen bitmap as the shadow framebuffer after screen resources + are created. + + Signed-off-by: Jon TURNEY + +commit 35d76028304f910edeaf642376df71b8134ec02c +Author: Jon TURNEY +Date: Sun Nov 2 18:27:13 2008 +0000 + + Cygwin/X: enqueue a pointer motion event on mouse movement + + Signed-off-by: Jon TURNEY + +commit 816e0d243df068bc06d328874b5655d4530fe16a +Author: Jon TURNEY +Date: Sun Nov 2 18:27:12 2008 +0000 + + Cygwin/X: update for changes in mieq API + + Bring Cygwin/X up to date with mieq event API changes + + Signed-off-by: Jon TURNEY + +commit bca7483c4e95ad450fb09048a3b85bb3056a0518 +Author: Jon TURNEY +Date: Sun Nov 2 18:27:11 2008 +0000 + + Cygwin/X: update for MPX device changes + + Bring Cygwin/X up to date with MPX device API changes + + Signed-off-by: Jon TURNEY + +commit 829a8a970dd17623ef3cb5022d34190744f74741 +Author: Jon TURNEY +Date: Sun Nov 2 18:27:10 2008 +0000 + + Cygwin/X: update for MPX cursor API changes + + Bring Cygwin/X up to date with cursor API changes + + Signed-off-by: Jon TURNEY + +commit d61e902aab92c262e6c8ee9cd70aec4493cf6cae +Author: Francis Giraldeau +Date: Wed Dec 17 09:49:39 2008 +0200 + + Don't log audit messages when -audit 0 specified + + make the auditTrailLevel check more general and don't log when not + using DTRACE and -audit 0 is set. + +commit 295a3fa721ee2f2226891de308c0c4b0ee26fe72 +Author: Paulo Cesar Pereira de Andrade +Date: Wed Dec 17 02:04:12 2008 -0200 + + Ensure symbols required by swrast_dri.so are visible. + +commit 6d8ea5104cf97dbf64612f58fc06f94f869ed5ec +Author: Colin Guthrie +Date: Tue Dec 16 17:44:27 2008 -0500 + + Fix compilation with -Werror=format-security + +commit 1bb5a919f4bf38ac96c73077021b0cdc82965f31 +Author: Adam Jackson +Date: Tue Dec 16 13:46:59 2008 -0500 + + config: Add vboxvideo and tga to the magic driver list + + Also sort the list while I'm here. + +commit 8511a964f81b3b06a526f0fca4232afb2152f405 +Author: Matthias Hopf +Date: Tue Dec 16 18:21:20 2008 +0100 + + randr: Update EDID_ATOM_NAME to reflect RandR 1.3 name change + +commit a058192d970f7417b06334aa666c17e9891795c6 +Author: Owen W. Taylor +Date: Wed Oct 22 16:12:53 2008 -0400 + + Xephyr: Fix crash with control-shift and GLX (#18185) + + When we are looking up the screen for an event, we need to take + into account the fact that the event may have been delivered to the + "peer window" that we create when implementing GLX. Since we only + ever create one such window per screen, just add a single peer_win + field to EphyrHostScreen. + +commit 6e85a8304b50c0bfd16a400a05d257f0168fb99b +Author: Jon TURNEY +Date: Sun Nov 2 18:27:09 2008 +0000 + + Cygwin/X: build machinery fixes + + The Cygwin/X build has been broken for a long time, probably since modular X + This patch fixes up configure/makefile and some general build issues (#18568) + + Signed-off-by: Jon TURNEY + +commit 7aa29b9d66c3cd0f8af4fafbe92efd0c0556d225 +Author: Robert Noland +Date: Tue Dec 16 00:07:42 2008 -0500 + + Support -sharevts on FreeBSD + + Bugzilla #11179 + + Submitted-by: Nathan Whitehorn + + Signed-off-by: Robert Noland + +commit 063eb6743cd0d98dd52d1a9559b804381ee5144d +Author: Keith Packard +Date: Mon Dec 15 19:53:45 2008 -0800 + + Patch brown-bag fix for bug 19017 (scrn->virtualX/virtualY 0 before PreInit) + + When a driver uses a crtc during device detection, the scrn has not yet been + configured and virtualX/virtualY are still zero. This caused the X server + to try and allocate a shadow frame buffer, which couldn't work. + + Detect this by checking for zero virtualX/virtualY values. + + Signed-off-by: Keith Packard + +commit fde2f961035609cfff8761ef1e0f23e6a9761be2 +Author: Keith Packard +Date: Mon Dec 15 11:39:24 2008 -0800 + + Use scrn->virtualX/virtualY in xf86CrtcFitsScreen. Fix bug 19017. + + pScreen->width/height are not initialized when doing initial mode setting, + which makes this function incorrectly fail. Using scrn->virtualX should work + in all cases though. + + Bug 19017 reports a crash in xf86CrtcSetModeTransform when doing a modeset + for output probing, long before the screen array is initialized; that was + caused by a work-around to set pScreen->width/height so that xf86CrtcFitsScreen + could find the right values. + + Signed-off-by: Keith Packard + +commit 29a5b0596b396d3e4a8a014cacd3e3ef77467ab7 +Author: Keith Packard +Date: Mon Dec 15 11:38:00 2008 -0800 + + When disabling SIGIO tracking, use SIG_IGN instead of SIG_DFL. + + This avoids a race condition for drivers which mis-order the fd close and + signal disable. + + Signed-off-by: Keith Packard + +commit 1ba4cbb15919759aadd71960c5c057af9ba94fe3 +Author: Keith Packard +Date: Mon Dec 15 11:36:43 2008 -0800 + + Clean up rotation data when crtc is turned off + + The shadow frame buffer and other data used for rotation need to be freed + when the crtc is disabled, not just when rotation is disabled. + + Signed-off-by: Keith Packard + +commit 2a61397d17339113b9e37995b06ca543589814ce +Author: Cooper Yuan +Date: Mon Dec 15 10:29:34 2008 -0500 + + Fix typo in xf86PickCrtcs() + +commit 3db60add6671f766f9360e7d03492dfd51eba225 +Author: Jon TURNEY +Date: Mon Dec 15 12:42:10 2008 +0000 -commit 2a822c231145afd2b99498d5d4bf96be7dff92c5 + Cygwin/X: Update the native icon to one based on the X.Org logo + + Update the native icon for the X server to one created from the X.Org logo by Colin Harrison. + +commit 450739efa904a0fce5024372da7236bb7b84b67a +Author: Jeremy Huddleston +Date: Sat Dec 13 00:45:27 2008 -0800 + + XQuartz: Updated menu item ordering for better HIG compliance + (cherry picked from commit 4c9bb241ce5fb856fab20ae96fa89bd2b71ef242) + +commit aee27be2785d8b09ac10f5be040b4d573190de0e +Author: Jeremy Huddleston +Date: Wed Dec 10 23:04:30 2008 -0800 + + XQuartz: Tiger fix, don't call Xplugin code in the Appkit thread if Xplugin isn't threadsafe. + (cherry picked from commit 748d9e5bd756513d42c4046f3b31e1fdc55bccb6) + +commit a939368ab8140d48c1da4ba0bb229d13b221189c Author: Jeremy Huddleston Date: Mon Dec 15 03:18:13 2008 -0800 @@ -2788,701 +14693,1006 @@ Date: Mon Dec 15 03:18:13 2008 -0800 Signed-off-by: Peter Hutterer Signed-off-by: Jeremy Huddleston - (cherry picked from commit a939368ab8140d48c1da4ba0bb229d13b221189c) -commit a067bd907e9d3e55edd6ca61279968b6cb96236c +commit f1c9b5ab230cbb4124d8d476ae4886d05022adcb +Author: Ian Romanick +Date: Sun Dec 14 18:58:33 2008 -0800 + + GLX: Changes resulting from changes to Mesa generator scripts / data + + Several recent Mesa commits (listed below) make modifications to the + protocol generator data and scripts. This commit represents the + changes to the generated files resulting from the previous changes. + + - 0f73302d24f4201813da2939742c5bcb6964b3b1 + GLX: Fix protocol for glTexSubImage#D + + - 1709ab01ef24279c782e420568e9257b4b92b224 + Return 0 as the request size when the pixels parameter is NULL + + - 63cca2ba10ce7dcc8481cfa4be3872dfc269dded + GLX: Include glapi.h before glapitable.h + + This is the server-side part of the fix for bugzilla #11003. + +commit 6383bc93b2ae76361a58653bc7e291df0d6ddfec +Author: Peter Hutterer +Date: Sat Dec 13 15:44:52 2008 +1000 + + dix: Init DeviceEnterLeave event's type before FixUpEventFromWindow (#19064) + + X.Org Bug 19064 + +commit 49d38ab2328c409b2a98465b52677af057121513 +Author: Eric Anholt +Date: Tue Dec 2 15:09:57 2008 -0800 + + Warning fix: Remove dead glXDisp{,Swap}_DrawArraysEXT definitions. + + Signed-off-by: Eric Anholt + Acked-by: Ian Romanick + +commit 2c5bfffc832e3818bbf0c0a96522865ce1ef2653 +Author: Eric Anholt +Date: Tue Dec 2 14:59:04 2008 -0800 + + Warning fix (GL likes to call strings GLubyte * instead of char *). + + Signed-off-by: Eric Anholt + Acked-by: Ian Romanick + +commit 7be6520d94df874c6bbd46d06a1830a12d0967f2 +Author: Stuart Bennett +Date: Fri Dec 12 00:08:44 2008 +0000 + + dolt: allow older versions of bash to compile the xserver (#19031) + + Signed-off-by: James Cloos + +commit aeff14d5f208d02f211b8b2a1a98999624e8c2cf Author: Peter Hutterer -Date: Mon Dec 1 21:20:48 2008 +1000 +Date: Tue Nov 25 20:23:50 2008 +1000 - mi: Clean up CopyGetMasterEvent, re-use the memory. + dix: don't alloc in ChangeMasterDeviceClasses. - Alloc an EventList once and then re-use instead of allocing a new event each - time we need a master event. - There's a trick included: because all the event processing handlers only take - an xEvent, init a size 1 EventList and squash the events into this one. + We mustn't realloc as we are inside a signal handler. With + SetMinimumEventSize, this code should never be hit anyway, as the event list + should have the required memory before this code is hit. - Events that have count > 1 must be squished into an xEvent array anyway before - passing into the event handlers, so we don't lose anything here. + Signed-off-by: Peter Hutterer + +commit fb2a8d0e59a3d187255538f6add22ec67551507a +Author: Peter Hutterer +Date: Mon Dec 8 15:58:15 2008 +1000 + + Xi: silence compiler warning + + Don't mix declarations and statements. + +commit d961abb59f0a8bee4bd11b1540aa43ff83cddbb8 +Author: Peter Hutterer +Date: Thu Dec 11 15:54:15 2008 +1000 + + dix: fix compiler warning (mixing declarations + statements) + + Signed-off-by: Peter Hutterer + +commit cb95642dc8edebb2935dd471f8b339cb98aa8481 +Author: Peter Hutterer +Date: Fri Nov 28 22:28:32 2008 +1000 + + Remove #define NEED_EVENTS and NEED_REPLIES + + A grep on xorg/* revealed there's no consumer of this define. + + Quote Alan Coopersmith: + "The consumer was in past versions of the headers now located + in proto/x11proto - for instance, in X11R6.0's xc/include/Xproto.h, + all the event definitions were only available if NEED_EVENTS were + defined, and all the reply definitions required NEED_REPLIES. + + Looks like Xproto.h dropped them by X11R6.3, which didn't have + the #ifdef's anymore, so these are truly ancient now." - Signed-off-by: Peter Hutterer - Acked-by: Daniel Stone - (cherry picked from commit d281866b74f7067f2704c278fe9720eafc0ee5ef) + Signed-off-by: Peter Hutterer + Signed-off-by: Adam Jackson -commit 3519f83f509dd0363f574d3a68b21df98d5f2732 -Author: Peter Hutterer -Date: Wed Dec 10 11:35:09 2008 +1000 +commit 920b7622be4a68ebb5d8432f55c5362fbe964fce +Author: Simon Thum +Date: Tue Nov 25 14:28:26 2008 +1000 - dix: move MAX_VALUATOR_EVENTS into include/input.h + xfree86: dump the ptraccel filter setup to the log on init. - Signed-off-by: Peter Hutterer - (cherry picked from commit 0b4fef6337d88ae8ef05b8b73941350a9007565c) + Signed-off-by: Peter Hutterer -commit 3ec3b502e3c6f0b74f8d96f7dda4f872156589a1 -Author: Francis Giraldeau -Date: Wed Dec 17 09:49:39 2008 +0200 +commit fbb57a2c54b141ccbb50833143b0b7cb2c4cb903 +Author: Simon Thum +Date: Tue Nov 25 13:43:28 2008 +1000 - Don't log audit messages when -audit 0 specified + dix: ptraccel - Add GetAccelerationProfile() - make the auditTrailLevel check more general and don't log when not - using DTRACE and -audit 0 is set. - (cherry picked from commit d61e902aab92c262e6c8ee9cd70aec4493cf6cae) + Signed-off-by: Peter Hutterer -commit 053f32d92864870cd5714befb419810976729df3 -Author: Colin Guthrie -Date: Tue Dec 16 17:44:27 2008 -0500 +commit beb749c87f383a62172a9ffa36373438e7a12d8e +Author: Adam Jackson +Date: Thu Dec 11 17:04:37 2008 -0500 - Fix compilation with -Werror=format-security - (cherry picked from commit 6d8ea5104cf97dbf64612f58fc06f94f869ed5ec) + Fix syncsrv.h guard define + +commit 9fba808b4f43f5dc69d85a48fa40cb83dd4d9f9d +Author: Colin Harrison +Date: Thu Dec 11 13:22:53 2008 -0500 + + xsync: Prototype fix. -commit 8ed4ca2801ece271a4ad7c73a7c0e1329f69ade8 +commit dce887ff53222652c397bc96a1dd3566f5847745 Author: Adam Jackson -Date: Tue Jun 17 16:10:51 2008 -0400 +Date: Thu Dec 11 13:16:02 2008 -0500 - XAA: Disable offscreen pixmaps by default. - - Say Option "XaaOffscreenPixmaps" to turn them back on. + xsync: build fix - Apropos of bugs #13795 and #15098. But this still isn't correct. - (cherry picked from commit 0c6987df3b9b3a37d201d740d8248c326449835e) + argh protocol header disaster -commit 32e81074b967716865aef08b66ec29caf0fec2c5 -Author: Robert Noland -Date: Tue Dec 16 00:07:42 2008 -0500 +commit d41b8960be8ad316504ef5657c0abfe6d76d3b5e +Author: Matthias Hopf +Date: Thu Dec 11 18:48:02 2008 +0100 - Support -sharevts on FreeBSD - - Bugzilla #11179 - - Submitted-by: Nathan Whitehorn + randr: Fix initial panning border copy - Signed-off-by: Robert Noland - (cherry picked from commit 7aa29b9d66c3cd0f8af4fafbe92efd0c0556d225) + Thanks to Julien Cristau for finding this. -commit 4c68079c2ae8a2f04363623c44d17bc8c255c3bf -Author: Ian Romanick -Date: Sun Dec 14 18:58:33 2008 -0800 +commit fd77ce9f884f5ac4d36736f3a99ba86101f133ea +Author: Matthias Hopf +Date: Thu Dec 11 17:51:20 2008 +0100 - GLX: Changes resulting from changes to Mesa generator scripts / data - - Several recent Mesa commits (listed below) make modifications to the - protocol generator data and scripts. This commit represents the - changes to the generated files resulting from the previous changes. - - - 0f73302d24f4201813da2939742c5bcb6964b3b1 - GLX: Fix protocol for glTexSubImage#D + randr: Oops, miscalculated panning rectangle's coordinates + +commit f8e52f1b6d7c59d007de99a1c9c69c053d4f3cbe +Author: Matthias Hopf +Date: Thu Dec 11 16:48:40 2008 +0100 + + randr: Update initial screen size if panning information is present + +commit 24d6b7d1c5fc5d07c2eb06a9fc4406e393d59b17 +Author: Matthias Hopf +Date: Thu Dec 11 16:48:11 2008 +0100 + + randr: Fix error message for bad panning config + +commit fdbbe65a7e777b7777bfae5a161efb89d4fb9a8d +Author: Matthias Hopf +Date: Thu Dec 11 14:56:51 2008 +0100 + + randr: Add monitor option "Panning" for initial panning configuration + +commit 27261a950d91c352eac25a3036656c3e4f81fb12 +Author: Paulo Cesar Pereira de Andrade +Date: Thu Dec 11 14:43:04 2008 -0200 + + Modify sdksyms.sh to receive $top_srcdir as first argument. - - 1709ab01ef24279c782e420568e9257b4b92b224 - Return 0 as the request size when the pixels parameter is NULL + If the basename of header file processed by cpp matches $top_srcdir, + check for extern symbols in the output, and add to the xorg_symbols + vector. + Possibly a better solution then using this script would be to somehow + tell the linker to not drop any symbols from the binary being generated. + +commit 1f4fb0225b278d1cf4145aebeb0bdd23dc8f62d5 +Author: Adam Jackson +Date: Wed Dec 10 16:13:20 2008 -0500 + + xsync: Fix wakeup storm in idletime counter. - - 63cca2ba10ce7dcc8481cfa4be3872dfc269dded - GLX: Include glapi.h before glapitable.h + Wakeup scheduling only considered the threshold values, and not whether + the trigger was edge or level. - This is the server-side part of the fix for bugzilla #11003. - (cherry picked from commit f1c9b5ab230cbb4124d8d476ae4886d05022adcb) + See also: + https://bugzilla.redhat.com/show_bug.cgi?id=474586 + http://svn.gnome.org/viewvc/gnome-screensaver/trunk/src/test-idle-ext.c?view=markup -commit 75bad86992338f5ce0ade6986b758cd6af2a5b9a -Author: Cooper Yuan -Date: Mon Dec 15 10:29:34 2008 -0500 +commit 1a99110f0c221b79045ea26d61c4a1ec1e0d7341 +Author: Adam Jackson +Date: Wed Dec 10 14:18:59 2008 -0500 - Fix typo in xf86PickCrtcs() - (cherry picked from commit 2a61397d17339113b9e37995b06ca543589814ce) + xsync: ANSI cleanups -commit f427d3aaa7ca58822033ac5486b46671a937b68a -Author: Jeremy Huddleston -Date: Sat Dec 13 00:45:27 2008 -0800 +commit e0d8f6a8084a3fe6c549c1dee11e4502e316c382 +Author: Adam Jackson +Date: Wed Dec 10 13:32:04 2008 -0500 - XQuartz: Updated menu item ordering for better HIG compliance - (cherry picked from commit 4c9bb241ce5fb856fab20ae96fa89bd2b71ef242) - (cherry picked from commit 73987010b2ef9c67b6614e226c6fae65d834d8f3) + xsync: Use a local header for server API definitions -commit d0eb4337e3468a5854e4801fcb1d2bc7d5c4d81f -Author: Jeremy Huddleston -Date: Wed Dec 10 23:04:30 2008 -0800 +commit 1208a1dbcaecbc218a77bf51068f9c0c768e16c5 +Author: Adam Jackson +Date: Wed Dec 10 11:47:35 2008 -0500 - XQuartz: Tiger fix, don't call Xplugin code in the Appkit thread if Xplugin isn't threadsafe. - (cherry picked from commit 748d9e5bd756513d42c4046f3b31e1fdc55bccb6) - (cherry picked from commit fd31984e0c0f9a37087cd1cffaa3ba116b12c2e5) + xsync: remove cast abuse. -commit d9e1fdcd40bf12378c9fa05a8adbe3aefca65676 +commit 119d9c46e841f5fa35610f557e6fa1ec58587c24 Author: Jeremy Huddleston Date: Wed Dec 10 01:56:01 2008 -0800 XQuartz: Fix path to executable (cherry picked from commit 7e9d3698e070a0c63dd2556651373c3aca58e2fe) - (cherry picked from commit 5926b213b39a90601c73f026dc0699723f5ed10d) -commit e93daf4e7140115f9485c8c0cdc358d50c9575be +commit c46b564c47fde3474e948aa1b188a975836cba47 Author: Jeremy Huddleston Date: Wed Dec 10 00:00:29 2008 -0800 XQuartz: fixed make dist (cherry picked from commit 2d52367ab92429e58596d1a1ed3ef52a0a38a7aa) - (cherry picked from commit 13eff12902be1b25d0ccc2089e08305f88949f32) -commit 1cf76f23c7d210d75a401568c5ad788a2fd268d1 +commit 23156a21d8e2f1be7d19d69cb91dae1b885ef671 +Author: Jeremy Huddleston +Date: Tue Dec 9 23:48:18 2008 -0800 + + XQuartz: Make debugging output for invalid depths a bit more detailed + (cherry picked from commit 609fb166b7062c76f0561df12ffe893811f6ac8f) + +commit 76351d2faf9ceb79aaa00b374c203b8b279c58f3 Author: Jeremy Huddleston Date: Tue Dec 9 23:47:32 2008 -0800 XQuartz: Avoid using login /bin/sh blech. Just use a bash script to start the app, so it will inherit the right environment (cherry picked from commit f4b7ad9cc6b0c99fc7ee8516c4bf858ece938148) - (cherry picked from commit fdf64256127b2661bd6aa81ac694350028d36c43) -commit 22d3ecda18ba799966d08237d8041ad80b20d0f2 +commit e5ce6e198f4c245f4dc840c5e9e90a7ef80fb6e9 Author: Jeremy Huddleston Date: Mon Dec 8 21:59:49 2008 -0800 XQuartz: unset DISPLAY if we didn't get a launchd socket handoff (cherry picked from commit b959727f38733009c6381cc8ca06b5984257bdac) - (cherry picked from commit 9cbed0a325175e7ddb751db54fe6c0f5a5cedd16) -commit 9eb5f3092984206ac7cab9a9fdb4b19448db3a24 +commit ecaaf0462d6c70cabfcaf6868a526e5415505e2c Author: Jeremy Huddleston Date: Mon Dec 8 21:00:11 2008 -0800 XQuartz: unsetenv(DISPLAY) if we're not org.x.X11 Also some prefix related fixes. (cherry picked from commit fd459e96581a883e30323c840b71004aa0747169) - (cherry picked from commit 99b2cbf061a9d074e66e6220dc08f8b4624ea6bb) -commit 19e6f73710776aee0b76dcbd62596a448029611e +commit ea94995982d1cc585768d52fec75cae018457fa5 Author: Jeremy Huddleston Date: Sat Dec 6 12:01:28 2008 -0800 XQuartz: Removed some debug spew (cherry picked from commit 370dcf11b360b87aa24cace35eb615419057fda0) - (cherry picked from commit 8065953ea8c3b7d10c775f6b7fec629bb5a2c83c) - -commit ced368287bad65c126cd6a5304969d0a0f9dc4eb -Author: Jeremy Huddleston -Date: Sat Dec 6 14:11:25 2008 -0800 - - XQuartz: darwinPointer reports the actual pixel position now rather than a relative position - (cherry picked from commit a41e7f75decd340d064fdc0d2c4fe6c88d7dbc82) - (cherry picked from commit 70930f6d31cc2ca16b40c17e101b106506a8337a) -commit b5687a9d129f6eea7ebe590120f9d90f9d7ed48b -Author: Jeremy Huddleston -Date: Fri Dec 5 12:42:31 2008 -0800 +commit 86c64ddf21763972aa7fc8c5770259123c9907b3 +Author: Adam Jackson +Date: Mon Dec 8 17:42:47 2008 -0500 - XQuartz: Corrected name/command labels in the customization widget - (cherry picked from commit a689c23f17eb445c36b97eb617ef4b8bd157985f) - (cherry picked from commit 9ac2e68d86ed1eb6e3f6c900c60908813eca140e) + randr: clear primaryOutput when the output is deleted -commit 53d93d0d29f0bef6b2b7e312fb09e498b68636f9 -Author: Jeremy Huddleston -Date: Tue Dec 2 14:54:10 2008 -0800 +commit fe65f400ed16cb39db8c9518b9446f590c34db1a +Author: Adam Jackson +Date: Mon Dec 8 17:40:07 2008 -0500 - Removed unised option from configure.ac for launchd - (cherry picked from commit 87e4de0a9a8af6e5e0cbba74e585761f1160a6ab) - (cherry picked from commit 451050b1e2dc0f2f6356d74ddb6f52183a794e8f) + randr: use primary output for RRFirstOutput() -commit df3f23fe8d13be74e58c894c44683024a190b0f6 -Author: Jeremy Huddleston -Date: Tue Dec 2 14:53:21 2008 -0800 +commit cdcb516e561e2f65eb2fa523ca001c57674d5caf +Author: Adam Jackson +Date: Mon Dec 8 17:37:17 2008 -0500 - XQuartz: pbproxy: Simplify linking - (cherry picked from commit 909cc5c4dca0f63e90505575bbd454b46a4670cc) - (cherry picked from commit cbb9ee57f8f29d2a1c39946381471fcd3b8e495e) + randr: Mangle GetScreenResources sort order based on primary output -commit 6e8626f2eb31a3699bf49597974e01ea4757f31e -Author: Peter Hutterer -Date: Mon Dec 8 15:58:15 2008 +1000 +commit d7b316e82bc7051f8829b4f4a640f50ae91c2db9 +Author: Adam Jackson +Date: Mon Dec 8 17:28:55 2008 -0500 - Xi: silence compiler warning - - Don't mix declarations and statements. - (cherry picked from commit fb2a8d0e59a3d187255538f6add22ec67551507a) + randr: Mangle compat Xinerama reply based on primary output -commit e56059dfa5f2835625b9f74a2af1f128d6eaf265 -Author: Peter Hutterer -Date: Thu Dec 4 09:47:25 2008 +1000 +commit 9d58d2a319059989ccdfa758f586149ccdc16df6 +Author: Adam Jackson +Date: Tue Dec 9 10:51:37 2008 -0500 - xfree86: fix compiler warning (use of uninitialized variable) - - drv and idev are only set for SDs, but are only dereferenced for SDs too, so - initializing them to NULL is safe. - - Signed-off-by: Peter Hutterer - (cherry picked from commit 8e3279134987a45f2a89c963ef2d33bc3d3c8179) + randr: Add [GS]etOutputPrimary -commit 0d12c44d832b98da10dccc3b8bac7676d8ea2c96 +commit d281866b74f7067f2704c278fe9720eafc0ee5ef Author: Peter Hutterer -Date: Mon Dec 1 16:57:06 2008 +1000 +Date: Mon Dec 1 21:20:48 2008 +1000 - xfree86: don't render SW cursors for devices attached to VCP (#16805) - - When leaving 3D games such as quake3 or sauerbraten, a cursor may stay on the - screen. This is caused by one run of SW rendering for the SD, even though the - SD was attached to the VCP and thus has HW rendering capabilities. + mi: Clean up CopyGetMasterEvent, re-use the memory. - Check for the SD's attachment (like in all other functions) before deciding on - SW or HW rendering. + Alloc an EventList once and then re-use instead of allocing a new event each + time we need a master event. + There's a trick included: because all the event processing handlers only take + an xEvent, init a size 1 EventList and squash the events into this one. - X.Org Bug 16805 + Events that have count > 1 must be squished into an xEvent array anyway before + passing into the event handlers, so we don't lose anything here. - Signed-off-by: Peter Hutterer - (cherry picked from commit bb072019fa8dd292a50ef433d05caeefd1304a73) + Signed-off-by: Peter Hutterer + Acked-by: Daniel Stone -commit 63b3bf85ed4e11ace433e6aa21b311cce7144007 -Author: Peter Hutterer -Date: Wed Dec 3 15:06:37 2008 +1000 +commit 0b4fef6337d88ae8ef05b8b73941350a9007565c +Author: Peter Hutterer +Date: Wed Dec 10 11:35:09 2008 +1000 - xkb: Allow NULL as rulesFile in XkbSetRulesDflts. - - If no rules file is given, simply re-use the previous one. If no RF is given - the first time this function is called, use the built-in default. - This includes fixing the built-in default to something that actually exists. + dix: move MAX_VALUATOR_EVENTS into include/input.h - Signed-off-by: Peter Hutterer - (cherry picked from commit 463e02e7de5da3e582a3a049110a476713c7210e) + Signed-off-by: Peter Hutterer -commit 9a44cc619f3296b0faca97b0959e19c208f143bb +commit b36c398b11321a908cfe217108b26a32ffc1d850 Author: Peter Hutterer -Date: Mon Dec 8 14:16:00 2008 +1000 +Date: Mon Dec 8 16:41:20 2008 +1000 - Xi: don't update VCP's valuators from DeviceValuator events #18882 + dix: purge dead device-based window access code. - The VCP doesn't need to update the valuators anyway since it cannot send XI - events. Just skip that bit. - - X.Org Bug 18882 - - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit c7fffe9e1354f816b6e2adedff045999fba9f1be -Author: Stuart Bennett -Date: Fri Dec 12 00:08:44 2008 +0000 +commit aa71ac1dd4697738752c73fe4c106b665ea3ec25 +Author: Paulo Cesar Pereira de Andrade +Date: Wed Dec 10 00:32:34 2008 -0200 - dolt: allow older versions of bash to compile the xserver (#19031) + Add dependency tracking to sdksyms.c and export composite wrapper. - Signed-off-by: James Cloos - (cherry picked from commit 7be6520d94df874c6bbd46d06a1830a12d0967f2) + Thanks to David Miller for noticing a make problem with sdksyms.c + not being regenerated when sdksyms.sh is updated. This is not yet the + best solution; automake generates dependency for sdksyms.o, but the + build really should also regenerate sdksyms.c when sdksyms.o needs to + be regenerated. + Export the symbols in miext/cw/cw.h. These symbols are in libxaa, and + at least the nvidia driver uses them. Maybe cw.h should be installed + in the sdk. -commit ca56d764d2be28c64fe15c9e37d534ef00117ad2 -Author: Adam Jackson -Date: Wed Dec 10 16:13:20 2008 -0500 +commit 090f63123975220819d531f569df6e5787607ec6 +Author: Peter Hutterer +Date: Mon Dec 8 11:53:20 2008 +1000 - xsync: Fix wakeup storm in idletime counter. + mi: always update the sprite for master devices. - Wakeup scheduling only considered the threshold values, and not whether - the trigger was edge or level. + Follow-up to 9ce995373e4a. This re-enables cursor rendering if the MD is + controlled through software (e.g. synergy). - See also: - https://bugzilla.redhat.com/show_bug.cgi?id=474586 - http://svn.gnome.org/viewvc/gnome-screensaver/trunk/src/test-idle-ext.c?view=markup - (cherry picked from commit 1f4fb0225b278d1cf4145aebeb0bdd23dc8f62d5) + Reported by John Tapsell: "I use Xorg with no mouse attached, but use synergy + to control the mouse. The commit means that I no longer have a visible mouse + cursor. The mouse cursor is still 'there' in terms that I can click buttons + etc with it, but it's just not visible." + + Signed-off-by: Peter Hutterer -commit a82f10c5dd9fa74ff18759ab288bbd9c8b7ac4de -Author: Adam Jackson -Date: Mon Dec 8 17:42:47 2008 -0500 +commit 7c8720c1433d2c3b85bbf4b811cc54c2df4c0080 +Author: Paulo Cesar Pereira de Andrade +Date: Mon Dec 8 22:11:50 2008 -0200 - randr: clear primaryOutput when the output is deleted - (cherry picked from commit 86c64ddf21763972aa7fc8c5770259123c9907b3) + Correct wrong symbol reference on sparc. + + The awk script was incorrectly referencing the struct name, and + not the struct variable. + Also added some comments to sdksyms.sh, for the reason it generates + the "symbol table" and add a message to the generated file, telling + is was automatically generated. -commit 2bc53ce66828b6c177e3298fa2f326c77c93e136 +commit 58a27d2932164e43c0db42b1286ec2f95250b420 Author: Adam Jackson -Date: Mon Dec 8 17:40:07 2008 -0500 +Date: Mon Dec 8 16:28:00 2008 -0500 - randr: use primary output for RRFirstOutput() - (cherry picked from commit fe65f400ed16cb39db8c9518b9446f590c34db1a) + Default to x86emu even on i386 linux + + vm86 mode is a bad idea anyway, and using the emulator everywhere + means we get a consistent set of bugs. -commit f0234a9eb88ed103bca7db73a833c472ab95b48f -Author: Adam Jackson -Date: Mon Dec 8 17:37:17 2008 -0500 +commit bbf811514d3cdf84790bad5b852942a4e636902b +Author: Sascha Hlusiak +Date: Mon Dec 8 12:24:39 2008 +0100 - randr: Mangle GetScreenResources sort order based on primary output - (cherry picked from commit cdcb516e561e2f65eb2fa523ca001c57674d5caf) + ddxCtrls.c: XkbDDXUsesSoftRepeat always returns 1 now + + We'd like to do soft repeat in the server for all keys. Remove obscure check, that'd + prevent the server from autorepeating when delay is set to exactly 660ms and rate is + set to exactly 25 (interval=40). + + Signed-off-by: Daniel Stone -commit 2ef02833d614c42693e019a444560e84f501b5dc -Author: Adam Jackson -Date: Mon Dec 8 17:28:55 2008 -0500 +commit 8c1dd40a04693f09f4fcea6e7f905af34c7589de +Author: Benjamin Close +Date: Mon Dec 8 14:49:38 2008 +1030 - randr: Mangle compat Xinerama reply based on primary output - (cherry picked from commit d7b316e82bc7051f8829b4f4a640f50ae91c2db9) + Don't use gnu specific extensions to awk when builing symbols + + Traditional posix awk doesn't know about \W and whilst we check that + awk exists in configure.ac we don't check which awk we are using. + This corrects symbol generation for posix only awk. -commit 0bdfdaa7df8105c7ffc3248a4fdd5f64da67103c -Author: Adam Jackson -Date: Tue Dec 9 10:51:37 2008 -0500 +commit 39db182b637041255ed6dac739ff77c8e4e07c30 +Author: Peter Hutterer +Date: Mon Dec 8 12:12:39 2008 +1000 - randr: Add [GS]etOutputPrimary - (cherry picked from commit 9d58d2a319059989ccdfa758f586149ccdc16df6) + xfree86: init EQ before trying to initialise the devices (#18890) + + The kbd driver may send events during device initialisation, and these events + need the EQ set up already. + + X.Org Bug 18890 + + Signed-off-by: Peter Hutterer -commit 09039fb89f3fd047f10b575e019bba6762448456 -Author: Keith Packard -Date: Mon Dec 15 19:53:45 2008 -0800 +commit 78a62d7713c708d067d8824ec41b0a0225c1997f +Author: Tom Jaeger +Date: Mon Dec 8 11:38:12 2008 +1000 - Patch brown-bag fix for bug 19017 (scrn->virtualX/virtualY 0 before PreInit) - - When a driver uses a crtc during device detection, the scrn has not yet been - configured and virtualX/virtualY are still zero. This caused the X server - to try and allocate a shadow frame buffer, which couldn't work. + Xi: XIGetDevice needs to ignore the MORE_EVENTS flag. - Detect this by checking for zero virtualX/virtualY values. + Reported in X.Org Bug 18882, Comment 5. + - Signed-off-by: Keith Packard - (cherry picked from commit 063eb6743cd0d98dd52d1a9559b804381ee5144d) + Signed-off-by: Peter Hutterer -commit ae2cbbff9fa0f458e133bf1894d8a7f9fb3fbb72 -Author: Keith Packard -Date: Mon Dec 15 11:39:24 2008 -0800 +commit ee1a6c28418a6dad6c89f79a994f27bfbaa77368 +Author: Peter Hutterer +Date: Fri Dec 5 10:12:57 2008 +1000 - Use scrn->virtualX/virtualY in xf86CrtcFitsScreen. Fix bug 19017. - - pScreen->width/height are not initialized when doing initial mode setting, - which makes this function incorrectly fail. Using scrn->virtualX should work - in all cases though. + dix: fix calculation of valuator events. - Bug 19017 reports a crash in xf86CrtcSetModeTransform when doing a modeset - for output probing, long before the screen array is initialized; that was - caused by a work-around to set pScreen->width/height so that xf86CrtcFitsScreen - could find the right values. + Follow-up to 4971315296cb. countValuatorEvents was copied from GKVE where it + was obviously broken but nobody noticed. GPE had the correct version, but that + one got lost during de-duplication. Restoring the correct calculation - if we + have 6 valuators, we want 1 valuator event, not 2. - Signed-off-by: Keith Packard - (cherry picked from commit fde2f961035609cfff8761ef1e0f23e6a9761be2) + Signed-off-by: Peter Hutterer -commit 6d671b2dee7d2e5481399f0bdb3323a0250a075d -Author: Keith Packard -Date: Mon Dec 15 11:38:00 2008 -0800 +commit 13de7511b17b57a28668e1a60b196ccfe61dbcbe +Author: Peter Hutterer +Date: Thu Dec 4 10:30:02 2008 +1000 - When disabling SIGIO tracking, use SIG_IGN instead of SIG_DFL. + xfree86: Only use the evdev ruleset on linux. - This avoids a race condition for drivers which mis-order the fd close and - signal disable. + As suggested by Julien Cristau - Signed-off-by: Keith Packard - (cherry picked from commit 29a5b0596b396d3e4a8a014cacd3e3ef77467ab7) + This is an follow-up to + commit 9c5dd7337fa93fb1650cc017e523b939dcbf482a + Author: Peter Hutterer + Date: Wed Dec 3 14:24:25 2008 +1000 + + Let the DDX decide on the XkbRulesDefaults. + + Signed-off-by: Peter Hutterer + Acked-by: Julien Cristau + Signed-off-by: Daniel Stone -commit 8f17a31aa721e42925cd6b4616ccad90497818f3 -Author: Keith Packard -Date: Mon Dec 15 11:36:43 2008 -0800 +commit 8e3279134987a45f2a89c963ef2d33bc3d3c8179 +Author: Peter Hutterer +Date: Thu Dec 4 09:47:25 2008 +1000 - Clean up rotation data when crtc is turned off + xfree86: fix compiler warning (use of uninitialized variable) - The shadow frame buffer and other data used for rotation need to be freed - when the crtc is disabled, not just when rotation is disabled. + drv and idev are only set for SDs, but are only dereferenced for SDs too, so + initializing them to NULL is safe. - Signed-off-by: Keith Packard - (cherry picked from commit 1ba4cbb15919759aadd71960c5c057af9ba94fe3) + Signed-off-by: Peter Hutterer -commit be21199f550d9a19c10df16ed23c6f9ed94e55da -Author: Matthias Hopf -Date: Thu Dec 11 18:48:02 2008 +0100 +commit 22d4ddcc3d63b7fbf2a23f5162075e4ee06db781 +Merge: d2dad38 c3bf15b +Author: Matthieu Herrb +Date: Sun Dec 7 23:56:15 2008 +0100 - randr: Fix initial panning border copy + Merge branch 'master' of git://anongit.freedesktop.org/git/xorg/xserver + +commit d2dad384f5725a36e593fd75f03d712f53b20620 +Author: Matthieu Herrb +Date: Sun Dec 7 23:52:26 2008 +0100 + + Fix build in separate build directory. + +commit c3bf15ba85fd3ab2e1fe809428b75bf34db25cc8 +Author: Paulo Cesar Pereira de Andrade +Date: Sun Dec 7 18:52:23 2008 -0200 + + Require macros 1.2.0 or newer for XORG_CHANGELOG and XORG_CWARNFLAGS. + +commit 5a8068b390b7d30eb526e954443d8e43a199b971 +Author: Paulo Cesar Pereira de Andrade +Date: Sun Dec 7 18:31:32 2008 -0200 + + Convert libx86emu.a to a "libtool convenience library". + +commit ed4a17274015ecd8040ae85bd5cd9d1dbcc9460a +Author: Paulo Cesar Pereira de Andrade +Date: Sun Dec 7 14:59:25 2008 -0200 + + Export some symbols from libxf86config when installing it. - Thanks to Julien Cristau for finding this. + These are private symbols, but used by the X Server. + The newly exported symbols were not added to the sdk headers. + Optionally, libxf86config could be compiled without hidden symbols + when being installed. + Thanks to Maarten Maathuis for noticing the problem. -commit 19a978359d7358501cccb79965b78a3f8ec18743 -Author: Matthias Hopf -Date: Thu Dec 11 17:51:20 2008 +0100 +commit bf4c29ab48a166eb158cf4be7c597982d65ef214 +Author: Paulo Cesar Pereira de Andrade +Date: Sun Dec 7 05:12:41 2008 -0200 - randr: Oops, miscalculated panning rectangle's coordinates + Correct xf86acpiDisableFlag symbol. + + It is declared as + ifdef HAVE_ACPI + Bool xf86acpiDisableFlag = FALSE; + endif + in hw/xfree86/common/xf86Globals.c + but not protected by the ifdef in the sdk header xf86Priv.h, what + caused a build failure in the tinderbox, due to the address of the + symbol being taken (to ensure it is available) in sdksyms.c. -commit ab56607a0855e88de2ad2d7eb43e1d806db0c47b -Author: Matthias Hopf -Date: Thu Dec 11 16:48:40 2008 +0100 +commit b1dac41fb3853ca8182048ea57b88b6e84ecceb3 +Author: Paulo Cesar Pereira de Andrade +Date: Sun Dec 7 02:22:19 2008 -0200 + + Use libtool convenience libraries and better "symbol" table. + + All .a libraries were converted to .la, and instead of linking the + Xorg binary with a mix of .a and .la, and adding some libraries more + then once in the command line, etc, now it generates a single libxorg.la + from all the required convenience libraries, and links with a dummy + xorg.c (that should usually be the file with the main function...). + This removes the requirement of some things like libosandcommon and + libinit, that existed to circumvent problems when linking multiple + .a and .la in the final Xorg binary. + + The "symbol table" is now generated dynamically, by a shell script, + with an embedded gawk parser that parses cpp output. The new file + sdksyms.sh is generated by hand by analyzing all Makefile.am's and + making it create a sdksyms.c file, that includes all sdk headers that + will add symbols for the Xorg binary. Module headers aren't read, and + a in 2 files it was required to add a "ifndef XorgLoader" around + declarations shared between the Xorg binary and libextmod. A few + other changes were added to other sdk headers, like preventing + multiple inclusion, or including other headers to satisfy dependencies. + + This should be a lot more portable, and better (hopefully properly) + using libtool to generate convenience libraries. + +commit ccd2c668c13863ed704f86b29fc6b3ee628e56a2 +Author: Jeremy Huddleston +Date: Sat Dec 6 14:11:52 2008 -0800 - randr: Update initial screen size if panning information is present + XQuartz: darwinPointer reports the actual pixel position now rather than a relative position + (cherry picked from commit a41e7f75decd340d064fdc0d2c4fe6c88d7dbc82) -commit 657d963a663a189d3c8068a39528b2d73af5fe70 -Author: Matthias Hopf -Date: Thu Dec 11 16:48:11 2008 +0100 +commit 7ebd0c7e8d42a13079957d9bacf5cb30d7855a59 +Author: Jeremy Huddleston +Date: Fri Dec 5 12:42:31 2008 -0800 - randr: Fix error message for bad panning config + XQuartz: Corrected name/command labels in the customization widget + (cherry picked from commit a689c23f17eb445c36b97eb617ef4b8bd157985f) -commit 44b89dc0a5aa23df69539754fb76c67c310530db -Author: Matthias Hopf -Date: Thu Dec 11 14:56:51 2008 +0100 +commit 027ff97a1354ab4c83fecb615f6bc2a6b739b871 +Author: Alan Coopersmith +Date: Fri Dec 5 12:00:08 2008 -0800 - randr: Add monitor option "Panning" for initial panning configuration + Make sure _X_EXPORT is defined in edid.h + + Fixes build error in xf86-video-intel utilities -commit 523aae1fa6d8002e55e85aee49f113b7eb9a6df3 -Author: Keith Packard -Date: Tue Dec 9 22:10:56 2008 -0800 +commit 466b0fca9ba5d5e7fb36e47a6d1bd60218f51c75 +Author: Paulo Cesar Pereira de Andrade +Date: Fri Dec 5 17:01:09 2008 -0200 - Bump version to 1.5.99.3 (1.6 beta3) + Add back a simplified version of the loader static address tables. - Signed-off-by: Keith Packard + If not taking the symbol addresses, linkage will break badly, as not + all symbols will be present, and it also requires changing library order, + and/or making some changes like the "libosandcommon". + + This table should be modified to be generated automatically, as + it is required to "fool" the compiler/loader into adding all required + symbols to the X Server. -commit 4810226dfc52ef798b507d284030b4b3aec020a6 +commit 44bef8b850c5a78a3d3eca5f0d92b71bdd0a87e2 Author: Matthias Hopf Date: Fri Dec 5 15:37:15 2008 +0100 randr: Update SProcRandrVector for panning - (cherry picked from commit 44bef8b850c5a78a3d3eca5f0d92b71bdd0a87e2) + +commit 44c8c3cf7de589fb8e987f4ab931294eaf0b405f +Author: Paulo Cesar Pereira de Andrade +Date: Fri Dec 5 04:26:57 2008 -0200 + + Remove static symbol address tables in hw/xfree86/loader/*sym.{c,h} + + Those tables were once used to decide what symbols are visible to + modules, but they were outdated. The only real usage was that, since + it was taking the address of symbols, linkage should fail if the + symbols were not available. + Now the proper way to make symbols available to modules should + be to use the _X_EXPORT macro, or not compile with hidden symbols, + so that all symbols would be available. + All symbols in the tables were revised to ensure they are exported, + and only symbols that were not exported are ClientSleepUntil() and + DuplicateModule(), that were not in the sdk for quite some time + already, and should not have any users outside of the X Server + (and/or builtin modules). + +commit 091a50d1260c70055aba1fa6a2d1b1a36de1114e +Author: Paulo Cesar Pereira de Andrade +Date: Thu Dec 4 18:23:25 2008 -0200 + + Move _X_EXPORT attribute to header file. - Signed-off-by: Keith Packard + The attribute should be set on header files to make it easier to + manage what symbols are available to modules. + _X_EXPORT should be used in sources only for special cases, like + symbols that must be visible by non video/input driver/modules but + should not be "advertised" in the sdk. -commit 798a4878a4ffb03024e77ea6ddbd9b170b9c3fe4 +commit e5ab9e66628cde081757cf2a1013a78e927a622e Author: Matthias Hopf Date: Thu Dec 4 18:13:40 2008 +0100 randr: Allow panning to be disabled per axis - (cherry picked from commit e5ab9e66628cde081757cf2a1013a78e927a622e) - - Signed-off-by: Keith Packard -commit 5fa2cce83cc2df560ce62ec4bbf88233ee70e64a +commit bad118ace6c5bae5a5ed8a35129c90c38f1c1932 Author: Matthias Hopf Date: Thu Dec 4 16:55:14 2008 +0100 randr: Rework panning area verification - (cherry picked from commit bad118ace6c5bae5a5ed8a35129c90c38f1c1932) - - Signed-off-by: Keith Packard -commit 97e8a75ce3c70e7a83028b256b6884084f5e196b +commit 219c26ce0c65625d55cfd943ec66fe94a1a0ddfd Author: Matthias Hopf Date: Thu Dec 4 16:28:40 2008 +0100 randr: Don't change panning parameters if verification fails. - (cherry picked from commit 219c26ce0c65625d55cfd943ec66fe94a1a0ddfd) - - Signed-off-by: Keith Packard -commit 6189f0832ecd6930608f5b253e534bb104174694 +commit 18a8bac1a1567b6215928f96870554ea63f39aab Author: Matthias Hopf Date: Thu Dec 4 16:30:38 2008 +0100 randr: Rename pan() to set_origin(), and xf86CrtcPan() to xf86CrtcSetOrigin() - (cherry picked from commit 18a8bac1a1567b6215928f96870554ea63f39aab) - - Signed-off-by: Keith Packard -commit 68858b202a0c32e8cc3fb22ca4714c1baa05b44c +commit 825b2c2f4a59ac4852f90bbbddf18ab832297fdd Author: Matthias Hopf Date: Thu Dec 4 16:11:21 2008 +0100 randr: Nuke config-timestamp for panning - (cherry picked from commit 825b2c2f4a59ac4852f90bbbddf18ab832297fdd) - - Signed-off-by: Keith Packard -commit 7cbded550d3bebc35e2322a89d0afbb57b37b3da +commit eeeb98d1df59baaaec954b6318d788a37e388d11 Author: Matthias Hopf Date: Fri Nov 28 17:51:20 2008 +0100 randr: Protocol bits for panning support - (cherry picked from commit eeeb98d1df59baaaec954b6318d788a37e388d11) - - Signed-off-by: Keith Packard -commit bc05471d1184ebf72b793b1dceff9856eba616d2 +commit b929d721efdb17bcc94b9984c4f34d0df3d267d5 Author: Matthias Hopf Date: Fri Nov 28 17:49:31 2008 +0100 randr: Panning support - (cherry picked from commit b929d721efdb17bcc94b9984c4f34d0df3d267d5) - - Signed-off-by: Keith Packard -commit f7e2fee8fe9b47f1e898feb8419aacd9a32c1f01 +commit 834cbc16f3eb539704faade7bff347b161ce69d9 Author: Matthias Hopf Date: Fri Nov 28 17:39:23 2008 +0100 randr: Crtc interface update for panning support. - (cherry picked from commit 834cbc16f3eb539704faade7bff347b161ce69d9) - - Signed-off-by: Keith Packard -commit 27ca32eb05101e0e5624422dd25e1aa95b129d77 +commit a475eb9feec75e9ce1e316da0f1679acd7dd3aa8 Author: Matthias Hopf Date: Fri Nov 28 17:38:52 2008 +0100 randr: Weird enough, crtc->version was never set upon creation. Fix that. - (cherry picked from commit a475eb9feec75e9ce1e316da0f1679acd7dd3aa8) + +commit 9db84fac0cc767e23986223d22bc085cde0cc86f +Author: Jeremy Huddleston +Date: Tue Dec 2 14:54:10 2008 -0800 + + Removed unised option from configure.ac for launchd + (cherry picked from commit 87e4de0a9a8af6e5e0cbba74e585761f1160a6ab) + +commit fa4f2527a5002711fe47c02b14097032fd845dc4 +Author: Jeremy Huddleston +Date: Tue Dec 2 14:53:21 2008 -0800 + + XQuartz: pbproxy: Simplify linking + (cherry picked from commit 909cc5c4dca0f63e90505575bbd454b46a4670cc) + +commit 126d8a4c8a72b039903023dce0da64b251080e1d +Author: Paulo Cesar Pereira de Andrade +Date: Thu Dec 4 02:43:29 2008 -0200 + + Update .gitignore. - Signed-off-by: Keith Packard + Add files left by make distcheck and remove files that no longer + are generated. -commit 0d27dd23cf08a2dc3581f9db81371c2cdb798db0 -Author: Adam Jackson -Date: Mon Dec 8 16:28:00 2008 -0500 +commit 9826b95c198e74f89680cc247b9bd9dd884d72be +Author: Paulo Cesar Pereira de Andrade +Date: Thu Dec 4 02:05:05 2008 -0200 - Default to x86emu even on i386 linux + Correct make distcheck by removing dolt files. + +commit 5dbe70dd52fd8daf8d0797951cc5a758d2c3b44c +Author: Paulo Cesar Pereira de Andrade +Date: Thu Dec 4 01:57:46 2008 -0200 + + Remove dummylib. - vm86 mode is a bad idea anyway, and using the emulator everywhere - means we get a consistent set of bugs. - (cherry picked from commit 58a27d2932164e43c0db42b1286ec2f95250b420) + Only cvt required it, and only used XNF{,c}alloc + via xnf{,c}alloc macros. + Based on patch by Eric Anholt. + +commit f841d4e3cccbde02e91c948f5ffb9e32c8c3b3cc +Author: Julien Cristau +Date: Wed Dec 3 23:02:03 2008 +0100 + + Move RELEASE_DATE closer to AC_INIT so it's more likely to be updated + +commit 75b02dd0be2ef64c0f8e3138c90b5c5e48b17913 +Author: Paulo Cesar Pereira de Andrade +Date: Wed Dec 3 14:46:30 2008 -0200 + + Include if _X_EXPORT is not defined. + +commit dd128ddcdcbe254a9cdd973590f6a979a7f0427e +Author: Timo Aaltonen +Date: Wed Dec 3 18:40:29 2008 +0200 + + If AEI is on, disable 'vmmouse' in addition to 'kbd' and 'mouse'. - Signed-off-by: Keith Packard + Signed-off-by: Timo Aaltonen + Signed-off-by: Peter Hutterer -commit ff9b55d8cbc19e0e31a91034e332058acd967cd1 -Author: Sascha Hlusiak -Date: Mon Dec 8 12:24:39 2008 +0100 +commit 0b5ecabfb803cd820338fb0364521fe39b05578b +Author: Julien Cristau +Date: Wed Nov 26 21:19:55 2008 +0100 - ddxCtrls.c: XkbDDXUsesSoftRepeat always returns 1 now + randr: add swapped dispatch for RR[GS]etCrtcTransform - We'd like to do soft repeat in the server for all keys. Remove obscure check, that'd - prevent the server from autorepeating when delay is set to exactly 660ms and rate is - set to exactly 25 (interval=40). + Fix a memory leak in ProcRRGetCrtcTransform() while I'm at it. - Signed-off-by: Daniel Stone - (cherry picked from commit bbf811514d3cdf84790bad5b852942a4e636902b) + Signed-off-by: Julien Cristau + Cc: Keith Packard + +commit 110a71d11ab7a1a55a6a24d792457fdef0b0746d +Author: Kristian Høgsberg +Date: Wed Dec 3 11:22:38 2008 -0500 + + Test for DRI2 extension in dri_internal.h and only enable AIGLX DRI2 if found. + +commit fd2d40b7ec5d685dac55453eb1f2da672dc83126 +Author: Peter Hutterer +Date: Wed Dec 3 11:31:48 2008 +1000 + + Xi: change XIUnRegisterPropertyHandler to XIUnregisterPropertyHandler + + CamelCase can be taken too far, and AFAICT there's no consumers of that + function yet anyway. + + Signed-off-by: Peter Hutterer + +commit 49f77fff1495c0a2050fb18f9b1fc627839bbfc2 +Author: Paulo Cesar Pereira de Andrade +Date: Wed Dec 3 05:43:34 2008 -0200 + + Rework symbol visibility for easier maintenance + + Save in a few special cases, _X_EXPORT should not be used in C source + files. Instead, it should be used in headers, and the proper C source + include that header. Some special cases are symbols that need to be + shared between modules, but not expected to be used by external drivers, + and symbols that are accessible via LoaderSymbol/dlopen. + + This patch also adds conditionally some new sdk header files, depending + on extensions enabled. These files were added to match pattern for + other extensions/modules, that is, have the headers "deciding" symbol + visibility in the sdk. These headers are: + o Xext/panoramiXsrv.h, Xext/panoramiX.h + o fbpict.h (unconditionally) + o vidmodeproc.h + o mioverlay.h (unconditionally, used only by xaa) + o xfixes.h (unconditionally, symbols required by dri2) + + LoaderSymbol and similar functions now don't have different prototypes, + in loaderProcs.h and xf86Module.h, so that both headers can be included, + without the need of defining IN_LOADER. + + xf86NewInputDevice() device prototype readded to xf86Xinput.h, but + not exported (and with a comment about it). + +commit 0b8f8b24f718820a72ebdc52423c2e6a44e848c5 +Author: Stuart Bennett +Date: Tue Dec 2 22:52:53 2008 -0800 + + xf86Cursors: xf86_reload_cursors shouldn't unconditionally show hwcursor (#14820) + + Also, no need to call ShowCursor when SetCursorPosition already does it + Based on a previous patch by Maarten Maathuis Signed-off-by: Keith Packard -commit b0c64725526338cbc02708415e1aa9497cdc3a38 +commit 9c5dd7337fa93fb1650cc017e523b939dcbf482a Author: Peter Hutterer -Date: Wed Nov 26 14:15:04 2008 +1000 +Date: Wed Dec 3 14:24:25 2008 +1000 - xfree86: don't FatalError on "too many input devices". + Let the DDX decide on the XkbRulesDefaults. - Just ignore devices after MAXDEVICES has been reached, but warn the user that - the devices are ignored. + Rather than assuming rules in the CoreKeyboardProc, init the default rules in + InitCoreDevices, then re-use them later. + + In the xfree86 DDX, set the rules to "base" or "evdev", depending on whether + we'll load kbd or evdev. + + If we create a new MD, use pc105,us as default and re-use the rules file used + previously. Signed-off-by: Peter Hutterer - Signed-off-by: Keith Packard -commit db5abde7ea0e482041d16d7d5f3715cd4f6222d3 +commit 463e02e7de5da3e582a3a049110a476713c7210e Author: Peter Hutterer -Date: Fri Dec 5 10:12:57 2008 +1000 +Date: Wed Dec 3 15:06:37 2008 +1000 - dix: fix calculation of valuator events. + xkb: Allow NULL as rulesFile in XkbSetRulesDflts. - Follow-up to 4971315296cb. countValuatorEvents was copied from GKVE where it - was obviously broken but nobody noticed. GPE had the correct version, but that - one got lost during de-duplication. Restoring the correct calculation - if we - have 6 valuators, we want 1 valuator event, not 2. + If no rules file is given, simply re-use the previous one. If no RF is given + the first time this function is called, use the built-in default. + This includes fixing the built-in default to something that actually exists. Signed-off-by: Peter Hutterer - (cherry picked from commit ee1a6c28418a6dad6c89f79a994f27bfbaa77368) + +commit e670fd889607fa712876218882cd4a9b46937661 +Author: Peter Hutterer +Date: Wed Dec 3 11:55:13 2008 +1000 + + dix: fix GetMaximumEventsNum(), may return a DCCE event too. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit d679cf70a79aa53e823f4fa51a7ab19837f26525 -Author: Tom Jaeger -Date: Mon Dec 8 11:38:12 2008 +1000 +commit 85d84c7cf2d368d1803dc2cdd03d6c9df8cc0430 +Author: Eric Anholt +Date: Tue Dec 2 17:33:43 2008 -0800 - Xi: XIGetDevice needs to ignore the MORE_EVENTS flag. + Fix GLX after 180bad84774493d48f2793a6281d825560944863. - Reported in X.Org Bug 18882, Comment 5. - + Sigh. + +commit 4a61ad427c634bcc38a31ce0f14ff5c2d3f706c7 +Author: Alan Coopersmith +Date: Tue Dec 2 14:35:45 2008 -0800 + + Update sample xorg.conf file + +commit c1db925d10fd37077bed90612ed95c3fd20cd2e2 +Author: Alan Coopersmith +Date: Tue Dec 2 14:32:48 2008 -0800 + + Add atKeynames.h to libdmxinput_a_SOURCES so it's included in tarballs - Signed-off-by: Peter Hutterer - (cherry picked from commit 78a62d7713c708d067d8824ec41b0a0225c1997f) + Reported by geearu in #xorg-devel + +commit 8561514574b3540c729bcc3acca9c943adcdc778 +Author: Alan Coopersmith +Date: Tue Dec 2 13:32:34 2008 -0800 + + Don't need to check uid/euid for every commandline argument - Signed-off-by: Keith Packard + Check uid/euid only when handling the arguments that are restricted + to root/non-setuid users -commit 480caacbdd20d47770cc9a698be9f7c98e87783d -Author: Peter Hutterer -Date: Mon Dec 8 12:12:39 2008 +1000 +commit 6de6ffff35ac03d49fa61de195d4a0605e0ef8bf +Author: Alan Coopersmith +Date: Tue Dec 2 13:19:08 2008 -0800 + + Remove unused config flags from FlagValues & FlagOptions + + FLAG_DISABLEMODINDEV, FLAG_MODINDEVALLOWNONLOCAL, & FLAG_HANDLE_SPECIAL_KEYS + +commit bb072019fa8dd292a50ef433d05caeefd1304a73 +Author: Peter Hutterer +Date: Mon Dec 1 16:57:06 2008 +1000 - xfree86: init EQ before trying to initialise the devices (#18890) + xfree86: don't render SW cursors for devices attached to VCP (#16805) - The kbd driver may send events during device initialisation, and these events - need the EQ set up already. + When leaving 3D games such as quake3 or sauerbraten, a cursor may stay on the + screen. This is caused by one run of SW rendering for the SD, even though the + SD was attached to the VCP and thus has HW rendering capabilities. - X.Org Bug 18890 + Check for the SD's attachment (like in all other functions) before deciding on + SW or HW rendering. - Signed-off-by: Peter Hutterer - (cherry picked from commit 39db182b637041255ed6dac739ff77c8e4e07c30) + X.Org Bug 16805 - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 991f0be6e2bfe3cc3df55b2fc75fe3900d5e0b94 -Author: Maarten Maathuis -Date: Sat Nov 29 14:30:55 2008 +0100 +commit 3a690598cf18c4cdc6aadd10a1ecf0772cacd34b +Author: Alan Coopersmith +Date: Tue Dec 2 12:59:57 2008 -0800 - randr: Avoid needlessly creating a shadow framebuffer. - - - pScreen->width and height were zero, so it didn't "fit" the screen. - (cherry picked from commit ffb484f7ef84099019b196ef97bfb2355eb6d52a) + Remove unused HandleSpecialKeys config option - Signed-off-by: Keith Packard + Was only used by "keyboard" driver - the last place the value was checked + was in xf86PostKbdEvent, which was removed in commit + 60ea7b51fe2b8a19a08b63db48504971a7a50ee6 -commit d6129ebf8dad9cfbc13dc8db7c780b023bf9a60b -Author: Timo Aaltonen -Date: Wed Dec 3 18:40:29 2008 +0200 +commit 1dfed222e93f4684c2a450944a9a0ea9e085c43f +Author: Bernhard Rosenkraenzer +Date: Tue Dec 2 09:01:56 2008 +1000 - If AEI is on, disable 'vmmouse' in addition to 'kbd' and 'mouse'. + Xext: fix MultiBuffer compilation error with TryClientEvents. (#18835) - Signed-off-by: Timo Aaltonen - Signed-off-by: Peter Hutterer - (cherry picked from commit dd128ddcdcbe254a9cdd973590f6a979a7f0427e) + X.Org Bug 18835 - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit 90900976ceee294fef4ae39e10869057c50f2021 -Author: Julien Cristau -Date: Wed Nov 26 21:19:55 2008 +0100 +commit 9ce995373e4aa4b1bf51b4adafa2324f781ec1cd +Author: Peter Hutterer +Date: Mon Dec 1 21:32:39 2008 +1000 - randr: add swapped dispatch for RR[GS]etCrtcTransform - - Fix a memory leak in ProcRRGetCrtcTransform() while I'm at it. + mi: UpdateSprite only if the device is attached. + +commit d507f60689f4e14383b0d24e63afc8cf836360d5 +Author: Peter Hutterer +Date: Wed Nov 26 14:15:04 2008 +1000 + + xfree86: don't FatalError on "too many input devices". - Signed-off-by: Julien Cristau - Cc: Keith Packard - (cherry picked from commit 0b5ecabfb803cd820338fb0364521fe39b05578b) + Just ignore devices after MAXDEVICES has been reached, but warn the user that + the devices are ignored. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit de83a342f483fbf3cc841929876002a3c47e3496 -Author: Kristian Høgsberg -Date: Wed Dec 3 11:22:38 2008 -0500 +commit 9bf761468fa3baf4c5fa40efa717b4b581d920f9 +Author: Peter Hutterer +Date: Fri Nov 28 11:44:58 2008 +1000 - Test for DRI2 extension in dri_internal.h and only enable AIGLX DRI2 if found. - (cherry picked from commit 110a71d11ab7a1a55a6a24d792457fdef0b0746d) - - Signed-off-by: Keith Packard + Xi: fix xi_filters size. -commit 3df1db67e37094ef263134a86ee82a36fc52f990 +commit 260e48c252c95fd1b1ba9ad7478791f6b9a67e2d Author: Peter Hutterer -Date: Wed Dec 3 11:31:48 2008 +1000 +Date: Fri Nov 28 11:29:06 2008 +1000 - Xi: change XIUnRegisterPropertyHandler to XIUnregisterPropertyHandler - - CamelCase can be taken too far, and AFAICT there's no consumers of that - function yet anyway. + dix: remove confusing (and wrong) comment. VCP is not the only pointer. Signed-off-by: Peter Hutterer - Signed-off-by: Keith Packard -commit ba666a969e938daa2176c12c4959393789b0a98d -Author: Keith Packard -Date: Tue Dec 9 20:36:41 2008 -0800 +commit 5d02e580d7c4144ea6f0984240fc913e05fec877 +Author: Peter Hutterer +Date: Fri Nov 28 11:46:15 2008 +1000 - Revert "dmx: claim we support XI 2." - - This reverts commit 8da8a0fec4b1b9d9208635dedb2f449dc99e0004. + dix: use UpdateFromMaster in GetProximityEvents. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit b268458eab2f213ec14dfe8013aa714c187e3aab -Author: Keith Packard -Date: Tue Dec 2 23:43:02 2008 -0800 +commit 95fc59a199f99bf167fbb09297a9bb0e33e31869 +Author: Peter Hutterer +Date: Fri Nov 28 14:55:15 2008 +1000 - Bump version to 1.5.99.2 (1.6 beta 2) - - Signed-off-by: Keith Packard + xkb: Extra sanity checks to prevent dev->key == NULL dereferencing. -commit d68632e2213adc0945dcdfca2e93f68aeab97457 -Author: Alan Coopersmith -Date: Tue Dec 2 14:32:48 2008 -0800 +commit a425abf0eaa61ee5ccb2f32a1af24edd190f7889 +Author: Peter Hutterer +Date: Wed Nov 26 12:20:00 2008 +1000 - Add atKeynames.h to libdmxinput_a_SOURCES so it's included in tarballs + xkb: don't attempt to filter events for devices without key classes. - Reported by geearu in #xorg-devel - (cherry picked from commit c1db925d10fd37077bed90612ed95c3fd20cd2e2) + Reported by Magnus Kessler. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer -commit f5ef57e5fef0be80d660c26ed370f608b4444a5e -Author: Adam Jackson -Date: Mon Dec 1 11:36:06 2008 -0500 +commit a85f0d6b98237d8a196de624207acf1983a1859a +Author: Peter Hutterer +Date: Wed Nov 26 11:15:05 2008 +1000 - randr: Don't send output property events on server exit + Xi: fix use of button->down - bitflags instead of int arrays. - If the Window resource type is already gone, there's no point in trying - to send events, all it can do is access already-freed memory. + The device's button down state array was changed to use DOWN_LENGTH and thus + bitflags for each button in cfcb3da7. - Relevant thread: + Update the DBSN events to copy this bit-wise state. + Update xkb and Xi to check for the bit flag instead of the array value. - http://lists.freedesktop.org/archives/xorg/2008-November/040443.html - (cherry picked from commit b0d371ab0a6efd4956c3677faa20b2ac15c33765) + Reported by ajax. - Signed-off-by: Keith Packard + Signed-off-by: Peter Hutterer + +commit 180bad84774493d48f2793a6281d825560944863 +Author: Paulo Cesar Pereira de Andrade +Date: Tue Dec 2 02:50:45 2008 -0200 -commit 40116cd5d0c205f4e8ab340dca80e38f8b78dddb + Add visibility flags to XSERVER_CFLAGS. + + This is done to actually change DIX_CFLAGS, as not all "modules" use + XORG_CFLAGS. + Also export the symbols that are required by other modules after + the change. + +commit 8c560422b44e012053612754430d2b87dc44ed59 Author: Alan Coopersmith Date: Mon Dec 1 20:02:34 2008 -0800 @@ -3502,75 +15712,20 @@ Date: Mon Dec 1 20:02:34 2008 -0800 The Xorg/xorg.conf sections on input device selection could use further updates to better match the current state of HAL-enabled configuration. - (cherry picked from commit 8c560422b44e012053612754430d2b87dc44ed59) - - Signed-off-by: Keith Packard -commit 4e5be5ad368b140d189fb69d394cf0c901f34d83 +commit 6a1850b8c677e2a2993f6f6b731ee3d35aa55d09 Author: Alan Coopersmith Date: Mon Dec 1 19:19:04 2008 -0800 Correct warning for unknown GlxVisuals option in conf file - (cherry picked from commit 6a1850b8c677e2a2993f6f6b731ee3d35aa55d09) - - Signed-off-by: Keith Packard - -commit ca744c371a138cfd19e826ba7f3847f32139ad64 -Author: Jay Cotton -Date: Mon Dec 1 16:06:09 2008 -0800 - - Sun bug 6618220: Xorg server core dump in xf86RandRModeRefresh(NULL) - - - (cherry picked from commit 5f3f14179edf48aad518f6f707bfdc37c27267c6) - - Signed-off-by: Keith Packard - -commit 78af5d81e818f4ac2fb43d9683fa634b3d1f7fc5 -Author: Alan Coopersmith -Date: Mon Dec 1 16:28:45 2008 -0800 - - Update See Also lists in Xorg & xorg.conf man pages - - Remove xorgconfig & xorgcfg - Update driver lists: - - Remove vga - - i810->intel - - via->openchrome - (cherry picked from commit 24c562f04b41d219c34f5fa3f963564accf329f2) - - Signed-off-by: Keith Packard - -commit e40841ce5afa0678b74b9eff37c7d304758309b6 -Author: Bernhard Rosenkraenzer -Date: Tue Dec 2 09:01:56 2008 +1000 - - Xext: fix MultiBuffer compilation error with TryClientEvents. (#18835) - - X.Org Bug 18835 - - Signed-off-by: Peter Hutterer - (cherry picked from commit 1dfed222e93f4684c2a450944a9a0ea9e085c43f) - -commit 0d902b05ad910cada46dcc2b85d8b0148aed35e6 -Author: Stuart Bennett -Date: Tue Dec 2 22:52:53 2008 -0800 - xf86Cursors: xf86_reload_cursors shouldn't unconditionally show hwcursor (#14820) - - Also, no need to call ShowCursor when SetCursorPosition already does it - Based on a previous patch by Maarten Maathuis - - Signed-off-by: Keith Packard - (cherry picked from commit 0b8f8b24f718820a72ebdc52423c2e6a44e848c5) - -commit cbf4fec52960ffd74b78ea45d0b917c76f553ebf +commit afeacb8e74b2a1e366e5ca9f0e58805d8d96c457 Author: Kristian Høgsberg Date: Mon Dec 1 20:53:17 2008 -0500 Bump dri2proto requirement to 1.99.3. -commit 670ac5d097c64d4f09ee3861d1ff21d5d6af3fdf +commit ad01e86b5c7c528adec8a1f95ecaa294f58a8922 Author: Kristian Høgsberg Date: Mon Dec 1 20:52:41 2008 -0500 @@ -3579,114 +15734,218 @@ Date: Mon Dec 1 20:52:41 2008 -0500 For this first iteration of DRI2 we don't have any vsync functionality in place yet, so back out the support in DRI2 and the protocol for now. -commit 862d5b2dc55e678921b3db710b2821b66d55ea2f -Author: Peter Hutterer -Date: Wed Nov 26 12:20:00 2008 +1000 +commit 24c562f04b41d219c34f5fa3f963564accf329f2 +Author: Alan Coopersmith +Date: Mon Dec 1 16:28:45 2008 -0800 - xkb: don't attempt to filter events for devices without key classes. - - Reported by Magnus Kessler. + Update See Also lists in Xorg & xorg.conf man pages - Signed-off-by: Peter Hutterer + Remove xorgconfig & xorgcfg + Update driver lists: + - Remove vga + - i810->intel + - via->openchrome -commit 41471afa7774a6598344ea7b488f993a0bd32e7f -Author: Peter Hutterer -Date: Fri Nov 28 14:55:15 2008 +1000 +commit bd147e6f4b03f344c967c88fd433877b14fe1912 +Author: Alan Coopersmith +Date: Mon Dec 1 16:08:55 2008 -0800 - xkb: Extra sanity checks to prevent dev->key == NULL dereferencing. + Remove hack to only load font modules if magic strings appear in font path -commit b983c4fa1ec13f6096badf9dd5ebbf55880bb6c1 -Author: Peter Hutterer -Date: Wed Nov 26 21:35:34 2008 +1000 +commit 5f3f14179edf48aad518f6f707bfdc37c27267c6 +Author: Jay Cotton +Date: Mon Dec 1 16:06:09 2008 -0800 - Require inputproto 1.5 - - Also claim to now support XI 1.5. + Sun bug 6618220: Xorg server core dump in xf86RandRModeRefresh(NULL) - Signed-off-by: Peter Hutterer - -commit f245c46d291dc22f40fb839bf21d2620091a84ef -Author: Peter Hutterer -Date: Fri Nov 28 14:21:05 2008 +1000 - - dix: don't detach SDs during grabs. + -commit 71f289f7d1b4fe79f25b229830047e71f50d8610 -Author: Peter Hutterer -Date: Fri Nov 28 14:05:58 2008 +1000 +commit 09df7cc5ad7b72d8a23c3e22fc718aad8c16f4d3 +Author: Kristian Høgsberg +Date: Mon Dec 1 12:41:10 2008 -0500 - dix: Don't deliver XI events from MDs. - - Restore the XI 1.x event model: VCP/VCK deliver core events only, SDs device - events only. + Avoid dereferencing NULL pScreen in xf86CrtcSetModeTransform(). - Signed-off-by: Peter Hutterer - -commit c910886737bf6ce7a4bd44d2149a1f0a810453d2 -Author: Peter Hutterer -Date: Fri Nov 28 13:49:40 2008 +1000 - - Xi: Purge XGE remainders, we don't have any XGE events anymore. + We can get there during PreInit as we set a mode for load detection. + At that time there's no pScreen anywhere, so just skip the optimization + then. -commit be4bf65d00a28cc77b939f00f115528a9024ab89 -Author: Peter Hutterer -Date: Fri Nov 28 09:22:04 2008 +1000 +commit b0d371ab0a6efd4956c3677faa20b2ac15c33765 +Author: Adam Jackson +Date: Mon Dec 1 11:36:06 2008 -0500 - input: don't switch MDs' classes around between SDs. + randr: Don't send output property events on server exit - Server 1.6 uses the X Input 1.x input model, where the core devices (VCP and - VCK) do not generate XI events. They don't have to swap device classes but - instead stay at the default number of classes at all times. - This means we can get rid of the DeviceClassesChangedEvents as well. - -commit aba1cbaadcde50a7a25f8aee06b66eec67a9145e -Author: Peter Hutterer -Date: Fri Nov 28 09:19:49 2008 +1000 - - dix: No DeviceEnterLeave events in server 1.6 - -commit 9739fac04520e3e0c338f74cedb58855a17fc899 -Author: Peter Hutterer -Date: Fri Nov 28 09:38:49 2008 +1000 - - Xi: don't allow VCP/VCK be OpenDevice'd, but allow all SDs. + If the Window resource type is already gone, there's no point in trying + to send events, all it can do is access already-freed memory. + + Relevant thread: - Reverting to traditional XI behaviour. + http://lists.freedesktop.org/archives/xorg/2008-November/040443.html -commit 7330cb56501defb0002815767d2dc7f5b2b8ca1c -Author: Peter Hutterer -Date: Fri Nov 28 09:35:48 2008 +1000 +commit 16b11cd03d8c5def07f0e598f237f71a37883a46 +Author: Paulo Cesar Pereira de Andrade +Date: Sun Nov 30 02:59:34 2008 -0200 - Xi: don't list attachment in ListInputDevices, but list all SDs. - - In XI2, we only list the VCP and the VCK as well as floating SDs to non-XI2 - clients. This is not the case here, we just list all devices. + Correct static symbol XkmReadTOC and first pass on compile warning fixes. - Signed-off-by: Peter Hutterer + The warnings corrected were only the ones that should correct + real problems. The most common one is 64 bit integers as + "printf %l" arguments. + Note that there is a patch related to this at: + http://bugs.freedesktop.org/show_bug.cgi?id=18204 -commit 1d307ed7f268d3bb511985c18d580ab8a104ed41 -Author: Peter Hutterer -Date: Fri Nov 28 09:34:56 2008 +1000 +commit fb22d4d928bc5d1a893494a059359da2ed2ac639 +Author: Paulo Cesar Pereira de Andrade +Date: Sun Nov 30 01:33:20 2008 -0200 - Xi: remove the GetExtensionVersion hack - don't reply with supported version. + Remove declarations of symbols that are never defined. - XI2 abuses the GEV request to reply with the min/major version of the - supported extension if the length for the name is 0. Don't do that, yet. - - Signed-off-by: Peter Hutterer + These symbols were removed from the X Server, or never declared. + One symbol that may need special attention is XkbBuildCoreState(), + that doesn't have a prototype anywhere, but is called from + xkb/xkbEvents.c:XkbFilterEvents(), and also used by the macros + XkbStateFieldFromRec() and XkbGrabStateFromRec() defined in + include/xkbstr.h. + fb/wfbrename.h also may need some cleanup, as it makes several + "renames" of non existing symbols. -commit cfe7301bcf134ade26a81764be2ca5890eb29d5d -Author: Peter Hutterer -Date: Wed Nov 26 21:40:05 2008 +1000 +commit d6cbd4511e35a89a0353f11834c6fdb8d4d2189f +Author: Paulo Cesar Pereira de Andrade +Date: Sat Nov 29 23:56:06 2008 -0200 + + Export symbols defined in the sdk. + + This is the biggest "visibility" patch. Instead of doing a "export" + symbol on demand, export everything in the sdk, so that if some module + fails due to an unresolved symbol, it is because it is using a symbol + not in the sdk. + + Most exported symbols shouldn't really be made visible, neither + advertised in the sdk, as they are only used by a single shared object. + + Symbols in the sdk (or referenced in sdk macros), but not defined + anywhere include: + XkbBuildCoreState() + XkbInitialMap + XkbXIUnsupported + XkbCheckActionVMods() + XkbSendCompatNotify() + XkbDDXFakePointerButton() + XkbDDXApplyConfig() + _XkbStrCaseCmp() + _XkbErrMessages[] + _XkbErrCode + _XkbErrLocation + _XkbErrData + XkbAccessXDetailText() + XkbNKNDetailMaskText() + XkbLookupGroupAndLevel() + XkbInitAtoms() + XkbGetOrderedDrawables() + XkbFreeOrderedDrawables() + XkbConvertXkbComponents() + XkbWriteXKBSemantics() + XkbWriteXKBLayout() + XkbWriteXKBKeymap() + XkbWriteXKBFile() + XkbWriteCFile() + XkbWriteXKMFile() + XkbWriteToServer() + XkbMergeFile() + XkmFindTOCEntry() + XkmReadFileSection() + XkmReadFileSectionName() + InitExtInput() + xf86CheckButton() + xf86SwitchCoreDevice() + RamDacSetGamma() + RamDacRestoreDACValues() + xf86Bpp + xf86ConfigPix24 + xf86MouseCflags[] + xf86SupportedMouseTypes[] + xf86NumMouseTypes + xf86ChangeBusIndex() + xf86EntityEnter() + xf86EntityLeave() + xf86WrapperInit() + xf86RingBell() + xf86findOptionBoolean() + xf86debugListOptions() + LoadSubModuleLocal() + LoaderSymbolLocal() + getInt10Rec() + xf86CurrentScreen + xf86ReallocatePciResources() + xf86NewSerialNumber() + xf86RandRSetInitialMode() + fbCompositeSolidMask_nx1xn + fbCompositeSolidMask_nx8888x0565C + fbCompositeSolidMask_nx8888x8888C + fbCompositeSolidMask_nx8x0565 + fbCompositeSolidMask_nx8x0888 + fbCompositeSolidMask_nx8x8888 + fbCompositeSrc_0565x0565 + fbCompositeSrc_8888x0565 + fbCompositeSrc_8888x0888 + fbCompositeSrc_8888x8888 + fbCompositeSrcAdd_1000x1000 + fbCompositeSrcAdd_8000x8000 + fbCompositeSrcAdd_8888x8888 + fbGeneration + fbIn + fbOver + fbOver24 + fbOverlayGeneration + fbRasterizeEdges + fbRestoreAreas + fbSaveAreas + composeFunctions + VBEBuildVbeModeList() + VBECalcVbeModeIndex() + TIramdac3030CalculateMNPForClock() + shadowBufPtr + shadowFindBuf() + miRRGetScreenInfo() + RRSetScreenConfig() + RRModePruneUnused() + PixmanImageFromPicture() + extern int miPointerGetMotionEvents() + miClipPicture() + miRasterizeTriangle() + fbPush1toN() + fbInitializeBackingStore() + ddxBeforeReset() + SetupSprite() + InitSprite() + DGADeliverEvent() + + SPECIAL CASES + o defined as _X_INTERNAL + xf86NewInputDevice() + o defined as static + fbGCPrivateKey + fbOverlayScreenPrivateKey + fbScreenPrivateKey + fbWinPrivateKey + o defined in libXfont.so, but declared in xorg/dixfont.h + GetGlyphs() + QueryGlyphExtents() + QueryTextExtents() + ParseGlyphCachingMode() + InitGlyphCaching() + SetGlyphCachingMode() + +commit ffb484f7ef84099019b196ef97bfb2355eb6d52a +Author: Maarten Maathuis +Date: Sat Nov 29 14:30:55 2008 +0100 - Xi: remove all new XI2 protocol requests and their handling. - - This commit reverts to XI 1.4 requests, plus the input device property - requests. + randr: Avoid needlessly creating a shadow framebuffer. - Signed-off-by: Peter Hutterer + - pScreen->width and height were zero, so it didn't "fit" the screen. -commit db115e78705e59a376c6c425e7cb97cfb14ff2ac +commit 01eaebdc98f30fdf543af6337cdf012d3ff16f09 Author: George Staplin Date: Fri Nov 28 13:57:45 2008 -0700 @@ -3699,7 +15958,7 @@ Date: Fri Nov 28 13:57:45 2008 -0700 Use xalloc and xfree instead of malloc and free. (cherry picked from commit 755f53eb40c4329d8c82a31cb910c31fdd4ea12e) -commit d4fb734c631a20e93da22773e676f747c9f645ba +commit 3d527b91b4bfa31e78d5fc7a1447a4026876f14e Author: Jeremy Huddleston Date: Fri Nov 28 12:59:09 2008 -0800 @@ -3709,42 +15968,135 @@ Date: Fri Nov 28 12:59:09 2008 -0800 (cherry picked from commit fd4710aff3723d5f3422cf6a6530172eafc556d9) (cherry picked from commit 635019ad18db921fec99256294debd8571074abf) -commit e9ee5d4f61794d8d12109b44daa92868cb834954 +commit 71d3df7cc37403b5842227a035d9b995fd920e0b Author: Jeremy Huddleston Date: Fri Nov 28 11:18:14 2008 -0800 XQuartz: Avoid some warning messages being spewed to system.log by AppKit (cherry picked from commit 780eff230ce41ad785e54d61a82c731269ae0446) -commit 2ca2742aa588d48348971e7e7c4b75cae4a4f5ad +commit a9796c7bc43223df44f12a7be08e361aea963ec1 Author: Jeremy Huddleston Date: Fri Nov 28 01:45:37 2008 -0800 XQuarz: Setup our PATH and PWD earlier, so our initial client benefits from it as well... (cherry picked from commit cbae2b447357b3fed6ff19414c60dd3792045600) -commit added4cff7069f0c9d6b7047c6435d1599003ccb +commit ef4179f43a84d90f867b95ee5974ad26884253b5 Author: Jeremy Huddleston Date: Wed Nov 26 12:15:58 2008 -0800 XQuartz: Fix dead-acute on Greek keyboards (cherry picked from commit 807f2ec35c5f95b56b3e2b5eac51aec300fe5cb6) -commit 202908c6139b806bc599cd74f3b68172c14aa7b5 +commit 800f5c681bed300d8caf99935e8f80ea6c1993ba Author: Jeremy Huddleston Date: Wed Nov 26 12:07:52 2008 -0800 XQuartz: Fix Czech keyboard dead-acute (cherry picked from commit 771df2786bc60389489f0967e705c7c95bdda085) -commit c7c8eeaedc8dc24c3ed7930d4322eb811977536e +commit ac57bb36d56a7a4d41add8d5a206ff37544a1819 Author: Jeremy Huddleston Date: Tue Nov 25 20:14:37 2008 -0800 XQuartz: Fix an uninitialized keyboard_type on Tiger (cherry picked from commit 27b1a5eb3482052253ebdce1a09aedf05ac1b099) -commit f0e7a792bedbf9fe9844298ac74445a8116c52aa +commit 45c8bd0fe54273039fdaa1eeeafb81b5774f2c75 +Author: Paulo Cesar Pereira de Andrade +Date: Fri Nov 28 02:01:57 2008 -0200 + + Enable compiling the X Server and modules with hidden symbols by default. + + Note that it checks if support for symbol visibility is available by + the compiler, not if it is functional. It may have problems on non x86 + architectures. + To disable the feature, just pass the option "--disable-visibility" + to the X Server configure script. + Unless using an alternate build schema, drivers built from a git + checkout will use the same "visibility" related CFLAGS used to compile + the X Server. + +commit 31285d063ec4623cb0764437d6d57e244f20460d +Author: Paulo Cesar Pereira de Andrade +Date: Fri Nov 28 01:55:11 2008 -0200 + + Make visible symbols required by xorg modules. + + This patch exports all symbols required by the compilable + (in a x86 linux computer) xorg/driver/* modules. + Still missing symbols worth mentioning are: + + sunleo + miFindMaxBand no longer available + + intel (uxa/uxa-accel.c) + fbShmPutImage no longer available (and should have been static) + + mga + MGAGetClientPointer (should come from matrox's libhal) + + This is not a definitive "visibility" patch, as all it does is to + export missing symbols, but the modules that current don't compile, + may require more symbols once fixed, and third party drivers should + also require more symbols exported. + A "definitive" patch should export symbols defined in the sdk. + +commit 6f18ea7e656b79e58f66f51067334414767b413f +Author: Julien Cristau +Date: Wed Nov 26 19:07:04 2008 +0100 + + Typo fix + +commit efefc03f9b26738b69abb570c1af2167cc99c4d8 +Author: Julien Cristau +Date: Sun Oct 26 13:19:48 2008 +0100 + + Add missing include + + ../../../../hw/xfree86/xaa/xaaTables.c:9:14: warning: symbol 'byte_expand3' was not declared. Should it be static? + ../../../../hw/xfree86/xaa/xaaTables.c:53:14: warning: symbol 'byte_reversed_expand3' was not declared. Should it be static? + +commit 74bc792e1814849b8eee9e448e36c3568b821e6c +Author: Paulo Cesar Pereira de Andrade +Date: Thu Nov 27 05:34:14 2008 -0200 + + Export symbols required by the vesa and fbdev drivers. + + This is a minimal set of patches, to have an usable X Server when + compiling it with symbols hidden by default. + +commit 74e2669103b59d51b24ddc48f378b1d04a3effd4 +Author: Paulo Cesar Pereira de Andrade +Date: Thu Nov 27 04:33:44 2008 -0200 + + Export symbols also defined in libXfont. + + libXfont has stubs for these symbols, so, when compiling with hidden + symbols by default, these symbols must be visible in the X Server, or + the stubs in libXfont will be used. + +commit 87a7fb7438b70d07ebf240c530b8548a01021d48 +Author: Paulo Cesar Pereira de Andrade +Date: Thu Nov 27 00:12:59 2008 -0200 + + Rework code using return value of LoaderSymbol as a function pointer. + + The patch removes all macros in the format + define xf86_sym ((type (*)(argument-list))LoaderSymbol("sym")) + creates a new macro in the format + define xf86_sym sym + and ensures "sym" is a "visible" symbol. + The patch doesn't add or remove features, and is source and binary + compatible with previous shared objects (with the difference that it + requires the dlloader). + These symbols are a special case, as, due to the fact that LoaderSymbol + was being used to reference them, they are not easily found by "automated" + tools that check for missing symbols. And now it also have the benefit + that the compiler/loader "knows what is going on". + +commit fc708394318ed92c612e2938b335c08c1ffebb28 Author: Keith Packard Date: Wed Nov 26 15:49:02 2008 -0800 @@ -3753,7 +16105,12 @@ Date: Wed Nov 26 15:49:02 2008 -0800 This depends on randrproto 1.2.99.1 Signed-off-by: Keith Packard - (cherry picked from commit fc708394318ed92c612e2938b335c08c1ffebb28) + +commit f710ce269c020a39f58b2bfbd0fe5192a3279c72 +Author: Alan Hourihane +Date: Wed Nov 26 20:02:44 2008 +0000 + + bump master to 1.6.99.1 (now the 1.6 branch is created) commit f3edc1fb0210149f35eab4e413700b5c4ac48214 Author: Keith Packard -- cgit v1.2.3