X.Org Server Changelog

New in version 1.17.1

February 11th, 2015
  • xkb: Don't swap XkbSetGeometry data in the input buffer
  • xkb: Check strings length against request size

New in version 1.17.0 (February 4th, 2015)

  • dri2: Set vdpau driver name if ddx does not provide any driver name
  • randr: attempt to fix primary on slave output (v2)
  • config/udev: Respect seat assignments when assigned devices
  • os: Fix timer race conditions

New in version 1.16.3 (December 20th, 2014)

  • glx: Be more paranoid about variable-length requests [CVE-2014-8093 1/6]
  • glx: Be more strict about rejecting invalid image sizes [CVE-2014-8093 2/6]
  • glx: Additional paranoia in __glXGetAnswerBuffer / __GLX_GET_ANSWER_BUFFER (v2) [CVE-2014-8093 3/6]
  • glx: Fix image size computation for EXT_texture_integer [CVE-2014-8098 1/8]
  • glx: Add safe_{add,mul,pad} (v3) [CVE-2014-8093 4/6]
  • glx: Integer overflow protection for non-generated render requests (v3) [CVE-2014-8093 5/6]
  • glx: Length checking for RenderLarge requests (v2) [CVE-2014-8098 3/8]
  • glx: Top-level length checking for swapped VendorPrivate requests [CVE-2014-8098 4/8]
  • glx: Request length checks for SetClientInfoARB [CVE-2014-8098 5/8]
  • glx: Length-checking for non-generated vendor private requests [CVE-2014-8098 6/8]
  • glx: Length checking for non-generated single requests (v2) [CVE-2014-8098 7/8]
  • glx: Pass remaining request length into ->varsize (v2) [CVE-2014-8098 8/8]
  • Add -iglx & +iglx to Xserver.man
  • unchecked malloc may allow unauthed client to crash Xserver [CVE-2014-8091]
  • dix: integer overflow in ProcPutImage() [CVE-2014-8092 1/4]
  • dix: integer overflow in GetHosts() [CVE-2014-8092 2/4]
  • dix: integer overflow in RegionSizeof() [CVE-2014-8092 3/4]
  • dix: integer overflow in REQUEST_FIXED_SIZE() [CVE-2014-8092 4/4]
  • dri2: integer overflow in ProcDRI2GetBuffers() [CVE-2014-8094]
  • dbe: unvalidated lengths in DbeSwapBuffers calls [CVE-2014-8097]
  • Xi: unvalidated lengths in Xinput extension [CVE-2014-8095]
  • xcmisc: unvalidated length in SProcXCMiscGetXIDList() [CVE-2014-8096]
  • Xv: unvalidated lengths in XVideo extension swapped procs [CVE-2014-8099]
  • dri3: unvalidated lengths in DRI3 extension swapped procs [CVE-2014-8103 1/2]
  • present: unvalidated lengths in Present extension procs [CVE-2014-8103 2/2]
  • randr: unvalidated lengths in RandR extension swapped procs [CVE-2014-8101]
  • render: unvalidated lengths in Render extn. swapped procs [CVE-2014-8100 2/2]
  • xfixes: unvalidated length in SProcXFixesSelectSelectionInput [CVE-2014-8102]
  • Add request length checking test cases for some Xinput 1.x requests
  • Add request length checking test cases for some Xinput 2.x requests
  • Add REQUEST_FIXED_SIZE testcases to test/misc.c
  • fb: Fix Bresenham algorithms for commonly used small segments.
  • render: check request size before reading it [CVE-2014-8100 1/2]
  • glx: Length checking for GLXRender requests (v2) [CVE-2014-8098 2/8]
  • present: Support PresentOptionCopy
  • glx: check return from __glXGetAnswerBuffer
  • dbe: Call to DDX SwapBuffers requires address of int, not unsigned int [CVE-2014-8097 pt. 2]
  • glx: Can't mix declarations and code in X.org sources [CVE-2014-8098 pt. 9]
  • Missing parens in REQUEST_FIXED_SIZE macro [CVE-2014-8092 pt. 5]
  • dix: GetHosts bounds check using wrong pointer value [CVE-2014-8092 pt. 6]
  • present: Avoid crashes in DebugPresent(), a bit more info.
  • present: Fix use of vsynced pageflips and honor PresentOptionAsync. (v4)
  • glx: Fix mask truncation in __glXGetAnswerBuffer [CVE-2014-8093 6/6]

New in version 1.17 RC1 (October 29th, 2014)

  • modesetting: match PCI class 3, any subclass
  • xfree86: Remove stale /usr/share/X11/xorg.conf.d hard-code from the xorg.conf man page
  • Adam Jackson (131):
  • Implement ->driverFunc
  • modesetting: Don't (brokenly) double-track software cursor
  • Use own thunk function instead of shadowUpdatePackedWeak
  • dix: Remove an ancient IBM configuration
  • dix: Remove some XFree86 3.x leftovers
  • dix: (Don't) change BITMAP_SCANLINE_UNIT on Linux s390{, x}
  • dix: Remove some dead macros
  • dix: Remove some cfb leftovers
  • dix: Default GLYPHPADBYTES to 4
  • dix: Remove wacky sparc special casing (v2)
  • dix: Remove a weird case of little-endian s390
  • dix: Remove arch awareness from servermd.h (v2)
  • xfree86: Remove i2c multimedia modules
  • xfree86: Remove #include "compiler.h" from places that don't need it
  • xfree86: Remove nds32_flush_icache
  • xfree86: Remove MMIO_MOVE32
  • xfree86: Undefine NO_INLINE
  • xfree86: Unspecialize gcc variants of unaligned memory access
  • xfree86: Remove unused unaligned int64 helpers
  • xfree86: Move generic unaligned helpers into int10 code
  • xfree86: Remove remaining unused unaligned accessors
  • xfree86: Remove a few random ppc decls
  • xfree86: Undef FAKEIT
  • xfree86: Undef __HIGHC__
  • xfree86: Remove an unlikely bit of #pragma
  • xfree86: Undef GCCUSESGAS
  • xfree86: Remove a useless !__SUNPRO_C guard
  • xfree86: Remove pre-2.6 Linux ppc support
  • xfree86: Simplify a bunch of OS and arch conditionals
  • xfree86: Clean up powerpc barrier decls
  • xfree86: Remove MMIO_ONB* and friends
  • xfree86: Pull generic barrier() definition up to top level
  • xfree86: Clean up some silly __sparc macro usage
  • xfree86: Unify the ppc/sparc mmio-swap-or-not conditionals
  • mi: Remove semi-arbitrary arch awareness in packed coordinate macros
  • bsd: Remove some reference arm code behind #if 0
  • dix: Make some LBX-era code more obvious
  • kdrive: Remove some dead bits of the man page
  • mi: Remove apparently unused miSegregateChildren
  • miext/shadow: Remove ancient backwards-compatibility hack
  • miext/shadow: Remove shadowInit
  • os: Remove deprecated malloc/free wrappers
  • os: Remove LocalClientCred
  • render: Remove unused CopyPicture
  • saver: Don't open-code IsMapInstalled
  • vgahw: Nuke unused vgaCmap.c
  • xfree86: Remove xf86ConfigActivePciEntity
  • xfree86: Remove some unused pre-pciaccess types
  • xfree86: Remove deprecated PCI types
  • xfree86: Remove unused xf86{Map,Unmap}LegacyIO
  • xfree86: Remove protectMem from VidMemInfo
  • xfree86: Remove pointless xf86RandRCreateScreenResources
  • xfree86: Remove useless #include "fb.h"
  • xfree86: Remove unused HardEdges
  • xkb: Remove some fascinating paranoia from event emission
  • xquartz: Remove unused DRI{,Post}ValidateTree
  • dri1: Remove unused DRI{,Post}ValidateTree
  • xwayland: Snap damage reports to the bounding box
  • xfree86: Remove xf86{Map,Unmap}VidMem
  • xfree86: Remove {set,undo}WC from VidMemInfo
  • xfree86: Remove xf86LinearVidMem
  • int10: Port off xf86ReadBIOS
  • xfree86: Remove xf86ReadBIOS
  • xfree86: Remove xf86CheckMTRR
  • xfree86: Remove the memory of MTRR awareness
  • xfree86: Clean up some more alpha awareness
  • loader: Remove compatibility with ancient XFree86 version numbering
  • ephyr: Properly implement hardware cursors (v3)
  • loader: Deobfuscate RTLD_* macro stuff
  • mi: Remove Windows 8-bpp colormap compatibility
  • xfree86: Remove xv clip notify driver hook
  • xfree86: Remove Option "TextClockFreq"
  • xfree86: Remove spurious xf86ConfigError
  • xfree86: Remove useless back-pointer to pScrn from colormap code
  • xfree86: Remove DisplayID support
  • xfree86: Remove xf86Info.log (v2)
  • xfree86: Remove unused xf86Info.useDefaultFontPathFrom
  • xfree86: configServerFlags never fails, make it return void
  • xfree86: Remove pointless xf86PrintMarkers
  • x86emu: Undefine _NO_INLINE
  • exa, kdrive: Remove redundant BitsPerPixel macros
  • input: Remove unused SetKeySymsMap
  • loader: Remove some baklava code
  • vfb: Don't reimplement micmap so much
  • xfree86: Remove some can't-happen printf from xf86CreateRootWindow
  • kdrive: Remove vestigial reference to fbInitValidateTree
  • xquartz: Remove useless DRIWindowExposures
  • rootless: remove redundant RootlessShapedWindowIn
  • xwin: Remove native GDI engine (v2)
  • xwin: Remove primary DirectDraw engine
  • dix: Drop the third argument from WindowExposuresProcPtr
  • mi: Deobfuscate miWindowExposures
  • mi: Deobfuscate miOverlayWindowExposures
  • mi: Simplify composite border clip redirection interface
  • mi: Drop plane argument from miHandleExposures
  • mi: Move pScreen->SendGraphicsExpose up to dix
  • dix: Remove DIXsaveUnder bit from the Window
  • dix: Lower backStorage to a bit instead of a pointer
  • dix: Remove an empty if
  • dix: Remove some pointless casting of NULL
  • dix: Remove an obfuscatory macro
  • mi: miSlideAndSizeWindow -> miResizeWindow
  • fb: fb{Map,Unmap}Window -> fb{Realize,Unrealize}Window
  • misc: Fold together some redundant conditionals
  • xinerama: Fix access mode in GetImage's drawable lookup
  • render: Always store client clip as a region
  • mi: Mark the span blit routines as _X_COLD
  • mi: Unexport wide line details
  • mi: Unexport polygon edge table implementation details
  • mi: Unexport span group implementation details
  • mi: Unexport subpixel polygon implementation details
  • mi: Unexport arc fill implementation details
  • mi: Fold mispans.c into miwideline.c
  • mi: Fold mifpolycon.c into miarc.c
  • mi: Fold mipoly{con,gen}.c into mipoly.c
  • mi: Fold mipolyutil.c into mipoly.c
  • mi: Fold micursor.c into mipointer.c
  • fb: Eliminate fbLaneTable, staticize fb{8,16,32}Lane
  • fb: FB_SHIFT is 5 (and FB_UNIT is 32)
  • fb: Remove even/odd stipple slow-pathing
  • fb: Move fbStipple*Bits near their only consumer
  • fb: Remove unused fbReduceRasterOp
  • fb: Remove even/odd tile slow-pathing
  • fb: Hide glyph implementation details
  • fb: Hide some zero-width line details
  • fb: Hide Bresenham line details
  • fb: Hide fbDots
  • fb: Hide fbPush{Pattern,Fill}
  • dix: Always store GC client clip as a region (v2)
  • dix: Retype and repack GC
  • composite: Wrap GetSpans
  • Alan Coopersmith (5):
  • Make list tests actually test lists
  • Abort xi2 tests if ActivateDevice() fails
  • Use unique display name for each xi2 test program [v3]
  • Allocate enough room for both reset & flags attributes
  • If fork fails in System(), don't fallthrough to exec()
  • Alan Hourihane (14):
  • Initial commit of new modesetting driver
  • Run indent
  • Plug in hardware cursor support
  • Bring inline with current modesetting-101 branch for connector/encoder changes
  • Add DRI2 support.
  • formatting fixes
  • more formatting fixes
  • yet more
  • Add EXA winsys for gallium pipe driver interface.
  • Fix copy surface
  • NULL checks
  • add --with-gallium-dir to specify location of gallium tree
  • More error checking
  • fixes for gallium stride changes
  • Alex Deucher (1):
  • modesetting: query cursor size from the kernel
  • Alex Orange (1):
  • fb: Fix Bresenham algorithms for commonly used small segments.
  • Alon Levy (1):
  • modesetting: add virtual connector support
  • Andreas Hartmetz (1):
  • glamor: Don't free memory we are going to use.
  • Axel Davy (1):
  • Fix present_pixmap when using present_notify_msc
  • Carlos Garnacho (1):
  • xwayland: Only disable/enable devices on capabilities change
  • Chris Wilson (2):
  • Add missing GTF modes
  • Xext/shm: Detach SHM segment after Pixmap is released
  • Colin Walters (2):
  • autogen.sh: Implement GNOME Build API
  • xf86platformBus: Add assertion to avoid (fatal) compiler warning
  • Daniel Martin (3):
  • dix: Remove XpExtension leftovers from protocol.txt
  • security: Remove XpExtension from SecurityTrustedExtensions
  • xfree86: Delete file modinit.h
  • Daphne Pfister (1):
  • os: Regenerate os/oscolor.c from app/rgb #52289
  • Dave Airlie (61):
  • drop exa
  • modesetting: restart driver effort from other codebases
  • cleanup some the drmmode_display header file
  • modesetting: rewrite probing based on fbdev.
  • gut some more unneeded code
  • update authors/copyright
  • drop loading dri2
  • port damage tracking code from st/xorg
  • fix server recycling
  • add -Wall, cleanup warnings
  • remove unused debug
  • add shadowfb support, default to on.
  • add initial man page
  • check drm support dumb buffer capability
  • fix make distcheck
  • fix fb_id for dirty reporting
  • configure: don't require xvmc.
  • fixup device open
  • fix some whitespace
  • modesetting: add cursor fallback if kernel rejects cursor.
  • cleanup dumb cap fetch
  • use a cap to decide if shadow is preferred or not.
  • shadowfb: dump shadowfb state at startup
  • modesetting: fix stride if kernel modifies it.
  • modesetting: move shadow stuff to other structure
  • modesetting: fix shadow resizing.
  • modesetting: disable dirty updates for ENOSYS
  • modesetting: fix warnings, remove dead code.
  • xf86-video-modesetting 0.2.0
  • modesetting: fix build against older Xext
  • drop use of alloca, just use malloc/free
  • dirty: check malloc return
  • cursor: hide cursors on LeaveVT
  • modesetting: add set/drop master around VT switch
  • modesetting: move opening kernel device to before setting depth
  • modesetting: attempt to work out if we want 24 or 32bpp
  • modesetting: make sure the pci device corresponds to the drm device
  • bump version to 0.3.0
  • modesetting: add compat header file
  • modesetting: convert to new scrn conversion APIs.
  • modesetting: fix make distcheck
  • modesetting: bump to latest X server compat api.
  • modesetting: workaround kernel bug reporting 0x0 as valid mins
  • modesetting: pci probing requires interface version 1.4
  • modesetting 0.4.0: bump configure.ac
  • modesetting: drop useless xf86PciInfo include
  • modesetting: fix warning about close being undefined.
  • modesetting: add platform bus support
  • modesetting: add output slave support.
  • modesetting: remove alloca usage again
  • modesetting: fix crashes caused by udev race conditions
  • modesetting: bump to 0.6.0
  • modesetting: provide dummy hooks for shadow
  • modesetting: fix adjust frame crash
  • modesetting: change output names for secondary GPUs
  • modesetting: move closing fd to after we check outputs
  • modesetting: try and use hotspot cursor support
  • modesetting: fix cursor rendering with hotspots.
  • modesetting: fix use after free.
  • modesetting: fix build regression against older servers.
  • pixmap: fix reverse optimus support with multiple heads
  • David Ung (1):
  • randr: Fix logic in RRPointerToNearestCrtc
  • Egbert Eich (1):
  • BellProc: Send bell event on core protocol bell when requested
  • Eric Anholt (52):
  • xorg: Remove duplicated definitions of some XV-related structs.
  • kdrive: Do a little more cleanup from the XV struct deduplication.
  • kdrive: Remove dead generic XV adaptors code.
  • kdrive: Simplify the adaptor setup interface.
  • glamor: Split the XV code into XF86-dependent parts and generic.
  • kdrive: Mark XV names const to avoid warnings.
  • ephyr: Add support for XV using glamor.
  • glamor: Share code for put_image handling.
  • xephyr: Allow initializing glamor with gles2 (on GLX).
  • glamor: Drop unnecessary glTexParameteri() in SetSpans().
  • glamor: Use MIN/MAX macros to clean up glamor_transfer.c
  • glamor: Drop unnecessary glTexParameteri() in upload of texture data.
  • glamor: Drop constant arguments to glamor_solid().
  • glamor: Remove always-true yInverted flag.
  • glamor: Drop dead get/pub sub pixmap functions.
  • glamor: Drop dead glamor_download_pixmap_to_cpu()
  • glamor: Drop dead glamor_restore_pixmap_to_texture().
  • glamor: Remove a dead prototype.
  • glamor: Drop the "are we doing a series of blits or draws" logic.
  • xv: Remove dead VIDEO_INVERT_CLIPLIST from the xorg and kdrive DDXes.
  • xv: Remove dead VIDEO_NO_CLIPPING from the xorg and kdrive DDXes.
  • xv: Move common code for adaptor cleanup to xvmain.c
  • xv: Drop the ddQueryAdaptors() interface.
  • xv: Move the DDX XV screen private allocation into the DDXes.
  • xv: Drop unused XvdiPreemptVideo().
  • xv: Drop unused XvdiVideoStopped().
  • xv: Move CloseScreen setup from a DIX hook to normal wrapping.
  • xv: Move xf86 XV color key helper to core.
  • xv: Fix malloc-failure cases in the fill color key helper.
  • kdrive: Remove dead KXVPaintRegion().
  • kdrive: Don't bother explicitly clearing new window privates to NULL.
  • xv: Remove the no-op AllocatePort/FreePort interfaces.
  • xv: Drop the ClientPtr from the interface to the DDX.
  • kdrive: Remove a dead struct.
  • glamor: Don't try to set up core fonts textures when we won't use them.
  • glx: Disable indirect GLX contexts by default.
  • modesetting: Merge modesetting's COPYING into the xserver's.
  • modesetting: unifdef MODESETTING_OUTPUT_SLAVE_SUPPORT
  • modesetting: xserver already checks XEXTPROTO >= 7.2.99.901.
  • modesetting: Drop use of "pointer" for "void *"
  • modesetting: Update for the platform device API change.
  • modesetting: Connect the driver to the build.
  • modesetting: Drop use of compat-api.h
  • modesetting: Drop remaining old-xserver compatibility.
  • modesetting: Run x-indent-all.sh.
  • modesetting: Fix some weird formatting after x-indent-all.sh.
  • modesetting: Deduplicate some scrn setup.
  • modesetting: Add support for rendering using glamor.
  • dri2: Import a copy of Mesa's PCI ID -> driver name mappings.
  • dri2: Automatically fill in the driver name if the DDX doesn't provide it.
  • modesetting: Export two functions I want to reuse from DRI2/Present.
  • modesetting: Add support for DRI2 with glamor.
  • Gaetan Nadon (14):
  • Fill the COPYING file with license text
  • Add contact information to the README file
  • man: missing AM_V_GEN and hard-coded sed command name
  • Fix typo in .gitignore
  • make: remove redundant AUTOMAKE_OPTIONS
  • Remove redundant EXTRA_DIST for the README file.
  • Add missing targets for ChangeLog and INSTALL
  • Replace obsolete AM_CONFIG_HEADER with AC_CONFIG_HEADERS
  • Remove redundant AC_PROG_CC
  • Remove unneeded AM_PROG_CC_C_CO
  • make: remove empty variable assignment
  • make: add all warnings according to the platform
  • config: replace deprecated AC_HELP_STRING with AS_HELP_STRING
  • config: layout and comment the top portion of configure.ac
  • Hans de Goede (1):
  • Add support for server managed fds
  • James Simmons (1):
  • don't map cursors in sw cursor mode
  • Jasper St. Pierre (2):
  • xwayland-input: Fix a crasher for a race with the Wayland compositor
  • xwayland: Implement throttling for surfaces based on the frame event
  • Jeremy Huddleston Sequoia (3):
  • mieq: Fix a crash regression in mieqProcessDeviceEvent
  • ListenOnOpenFD: Remove Resets since this is intended to be for hotplugging connections
  • XQuartz: Better support turning off "Displays have separate Spaces" on OS X Mavericks
  • Joerg Sonnenberger (1):
  • Fix format string.
  • Jon TURNEY (56):
  • hw/xwin/glx: When generating shims, limit the considered features to GL version

New in version 1.16.1 (September 22nd, 2014)

  • A single fix since Monday's 1.16.0.901, to address an issue when building Xwayland from the tarball.

New in version 1.15.0 (December 27th, 2013)

  • present: Handle PresentOptionAsync for copy mode
  • present: Don't abandon presents which are already queued for flip
  • present: Set complete notify mode to Skip as needed

New in version 1.15 RC5 (December 21st, 2013)

  • dri3: Guard against buggy clients
  • dri3: Don't use the major/minor version from the protocol headers
  • present: Don't use the major/minor version from the protocol headers
  • glx: Add null pointer protection to __glGetProcAddress
  • configure: Fix a typo near dri3/shmfence detection
  • Xi: Don't ActivateEarlyAccept POINTER_REGULAR listeners
  • darwin: Don't leave stdin/stdout closed
  • XQuartz: Use asl_log_descriptor to log stdout/stderr of child processes
  • configure.ac: Add PRESENT_LIB to XQUARTZ_LIBS
  • present: recursively set window pixmaps on flip
  • Xi: ungrab device when releasing a passive grab without ButtonReleaseMask (#71878)
  • dix: fix check for grab type

New in version 1.14.5 (December 13th, 2013)

  • xfree86: detach scanout pixmaps when detaching output GPUs
  • xfree86: detach scanout pixmaps when detaching output GPUs
  • randr: Fix a copypasta bug in CRTC confinement
  • randr: Fix a copypasta bug in CRTC confinement
  • xfree86: Find primary entity when bus types are nominally different
  • randr: don't directly set changed bits in randr screen
  • randr: make SetChanged modify the main protocol screen not the gpu screen
  • randr: only respected changed on the protocol screen
  • randr: report changes when we disconnect a GPU slave
  • Revert "Merge commit 'f98d6cfa0de41cf33e7555342753897c07af84ef' into server-1.14-branch"
  • randr: don't directly set changed bits in randr screen
  • randr: make SetChanged modify the main protocol screen not the gpu screen
  • randr: only respected changed on the protocol screen
  • randr: report changes when we disconnect a GPU slave
  • glx: Add support for the new DRI loader entrypoint.
  • XQuartz: Bump bundle version to 2.7.5
  • miext/damage: Partial revert of "Only wrap into the GC ops chain if there's a listener (v3)"
  • XQuartz: Update Info.plist copyright years
  • XQuartz: Don't sleep forever on exit if usleep() returns an error
  • xfree86: Fix build without libpciaccess
  • randr: send RRProviderChangeNotify event
  • randr: send RRResourceChangeNotify event
  • randr: deliver Output and Crtc events of attached output providers.
  • randr: send RRProviderChangeNotify event (abi unbreak)
  • randr: send RRResourceChangeNotify event (abi unbreak)
  • randr: deliver Output and Crtc events of attached output providers.
  • sync: fix corner-case in triggering idle alarms

New in version 1.15 RC3 (November 25th, 2013)

  • mieq: Bump default queue size to 512
  • Stop including inline assembly .il file for Solaris Studio builds
  • glx/glxdri2: Unwrap EnterVT/LeaveVT upon CloseScreen
  • test: build the touch test only when building Xorg
  • xfree86: return NULL for compat output if no outputs.
  • kdrive: handle WxH as valid geometry spec
  • present: Round fake MSC computations intead of truncating
  • present: Ignore event_id 0 from driver vblank notify
  • present: Signal destroyed flip window with vblank->window == NULL
  • present: Block for wait_fence in present_execute
  • present: When unflipping, copy to flip window rather than screen
  • xnest: Ignore GetImage() error in xnestGetImage() (v3)
  • Xext: Recover from miSyncInitFenceFromFD failure without crashing
  • miext: Ensure xshmfence is only called when driver supports it
  • ephyr: xcb_connect returns an error, not NULL
  • kdrive: modify ephyr events to use POINTER_DESKTOP and scale them to that
  • kdrive: fix cursor jumps on CursorOffScreen behavior
  • include: export key_is_down and friends
  • configure: allow for --enable-libunwind and --disable-libunwind

New in version 1.15 RC2 (November 14th, 2013)

  • present: Don't try to initialize when building without present support
  • Avoid conflicts with Solaris defines that clash with our names
  • Enable XTRANS_SEND_FDS on Solaris too.
  • ARM64: Add support for aarch64
  • configure.ac: Move GLX section after DRI
  • XQuartz: Update Info.plist copyright years
  • XQuartz: Don't sleep forever on exit if usleep() returns an error
  • xfree86: Fix build without libpciaccess
  • os: Actually use the computed clockid in GetTimeInMicros
  • Disable DRI3 and sync fence FD functions if xshmfence isn't available
  • hw/xfree86: Link libdri3 only when DRI3 is defined
  • Link with xshmfence, reference miSyncShmScreenInit in sdksyms
  • Use $GL_LIBS instead of -lGL for linking
  • Make XTrans FD passing support optional. Define only on Linux
  • Remove dix-config.h config variables from xorg-config.h
  • Require xextproto version 7.2.99.901
  • Xext: Enable MIT-SHM FD-passing request definitions only when possible
  • Require libXtrans version 1.3.2
  • Trap SIGBUS to handle truncated shared memory segments
  • Present: Check for window/fence destroyed when idling pixmaps
  • present: Change debug output a bit to help diagnose missing vblank signals
  • Get rid of the rest of the FD passing code when XTRANS_SEND_FDS isn't set
  • Proper spelling of MAP_ANONYMOUS is MAP_ANON.
  • xext: Fix shmint.h to not use headers outside of sdk_HEADERS
  • Update to version 1.14.99.902

New in version 1.15 RC1 (November 2nd, 2013)

  • Adam Jackson:
  • glx: Only take a Pixmap reference if creating the GLXPixmap succeeded
  • glx: Fix memory leak in context garbage collection (v2)
  • glx: Fix a corner case in DrawableGone
  • glx: Simplify glXDestroyContext
  • glx: Remove DRI1 AIGLX (v2)
  • glx: convert to direct GL dispatch (v2)
  • xfree86: Bump video/input/extension ABIs for 1.15
  • dri: Fix GLX_Y_INVERTED_EXT fbconfig attribute
  • glx: Lie about GLX_Y_INVERTED_EXT
  • Connor Behan:
  • xfree86: Find primary entity when bus types are nominally different
  • Daniel Czarnowski:
  • glx: Allow float renderType in glXCreateContextAttribsARB
  • glx: Correctly set render type enum
  • glx: Handle float config types in glxConvertConfigs
  • glx: Enable GLX_ARB_fbconfig_float for DRI2 drivers
  • Egbert Eich (6):
  • DDX/DPMS: Call dixSaveScreens() also when screen is turned on
  • DDX/modes: Add a sanity check when using screen sizes from EDID
  • DDX/Events: Distinguish between Input- and GeneralHandlers in xf86VTSwitch()
  • DDX/Randr: Avoid server crash when xrandr SetConfig is called while switched away
  • config/ACPI: Add option to disable building of ACPI support on Linux
  • config/APM: Add option to disable building of APM support on Linux
  • Eric Anholt:
  • glx: Add support for the new DRI loader entrypoint.
  • Gaetan Nadon:
  • config: replace deprecated use of AC_OUTPUT with AC_CONFIG_FILES
  • Jeremy Huddleston Sequoia (2):
  • XQuartz: Bump bundle version to 2.7.6
  • miext/damage: Partial revert of "Only wrap into the GC ops chain if there's a listener (v3)"
  • Keith Packard:
  • kdrive/ephyr: Don't discard one-time driver structure at server reset
  • Xext: Move MIT-SHM 'ShmDesc' to shmint.h
  • Clean up a couple of warnings in os/
  • Add a RegionDuplicate function
  • Add swapll to byte swap 64-bit datatypes
  • os: Add GetTimeInMicros
  • misync: Don't smash custom screen sync functions
  • Add interfaces to get FDs from clients over the socket
  • Add support for MIT-SHM AttachFd request
  • dri3: Add DRI3 extension
  • present: Add Present extension
  • Add dri3 and present directorys to distribution
  • Set version to 1.14.99.901 (1.15 RC1)
  • Michal Srb:
  • randr: send RRProviderChangeNotify event
  • randr: send RRResourceChangeNotify event
  • randr: deliver Output and Crtc events of attached output providers.
  • Peter Hutterer:
  • Xi: fix logic error when calculating emulated motion events
  • dmx: queue button events with a flags of 0
  • configure: remove a comment
  • os: use a constant for backtrace array size
  • sync: fix corner-case in triggering idle alarms
  • Strake:
  • unbreak KdEnableInput
  • Søren Sandmann Pedersen (3):
  • xf86AddBusDeviceToConfigure(): Store device in DevToConfig[i].pVideo
  • ephyr: hostx_screen_init(): Fix bits_per_pixel and bytes_per_line
  • ephyr: Ensure stride of private framebuffer is multiple of 4

New in version 1.14.4 (November 1st, 2013)

  • Avoid use-after-free in dix/dixfonts.c: doImageText() [CVE-2013-4396]
  • Allow disabling XFree86-DGA, DRI, VidModeExtension extensions
  • DIX/Xi: Pass correct client to CheckDeviceGrabAndHintWindow()
  • test: include dix-config.h in hashtabletest.c
  • xfree86: improve check for posix saved ids
  • dix: add missing include for DeleteWindowFromAnySelections
  • test: fix the gcc diagnostics pragma
  • dix: check for grab type before checking XI2 mask
  • Xi: allow for XIAllowEvent requests larger than XI < 2.2 size (#68554)
  • dix: only allow button and key events to freeze a sync'd pointer
  • include: change grabtypes to start at 1
  • dix: don't attempt to deliver an event for a different grabtype
  • dix: only deliver for the current grab type
  • sync: compress two if statements
  • sync: always call BracketValues when recalculating upper/lower brackets
  • sync: supply the counter from IdleTimeBlockHandler
  • dix: provide accessor methods for the last device event time
  • sync: if the idle time was reset, force alarms to trigger (#70476)
  • sync: always set the brackets (#59644)
  • sync: split updating and triggering a counter up

New in version 1.14.4 RC1 (October 28th, 2013)

  • Alan Coopersmith:
  • Avoid use-after-free in dix/dixfonts.c: doImageText() [CVE-2013-4396]
  • Allow disabling XFree86-DGA, DRI, VidModeExtension extensions
  • Egbert Eich:
  • DIX/Xi: Pass correct client to CheckDeviceGrabAndHintWindow()
  • Julien Cristau:
  • test: include dix-config.h in hashtabletest.c
  • xfree86: improve check for posix saved ids
  • dix: add missing include for DeleteWindowFromAnySelections
  • Peter Hutterer:
  • test: fix the gcc diagnostics pragma
  • dix: check for grab type before checking XI2 mask
  • Xi: allow for XIAllowEvent requests larger than XI < 2.2 size (#68554)
  • dix: only allow button and key events to freeze a sync'd pointer
  • include: change grabtypes to start at 1
  • dix: don't attempt to deliver an event for a different grabtype
  • dix: only deliver for the current grab type
  • sync: compress two if statements
  • sync: always call BracketValues when recalculating upper/lower brackets
  • sync: supply the counter from IdleTimeBlockHandler
  • dix: provide accessor methods for the last device event time
  • sync: if the idle time was reset, force alarms to trigger (#70476)
  • sync: always set the brackets (#59644)
  • sync: split updating and triggering a counter up

New in version 1.14.99.3 (October 21st, 2013)

  • Adam Jackson:
  • input: calloc minimization for xi2mask_new
  • Alan Coopersmith:
  • Skip damage calls if DamageCreate fails in exa functions
  • DMX glxproxy: Don't allocate & copy data just to free it unused
  • Update GLX dependencies now that DRI & DRI2 are builtins, not modules
  • Avoid use-after-free in dix/dixfonts.c: doImageText() [CVE-2013-4396]
  • Emil Velikov:
  • configure.ac: Include missing proto declarations in SDK_REQUIRED_MODULES
  • Jeremy Huddleston Sequoia:
  • rootless: Use miCopyRegion instead of fbCopyRegion
  • XQuartz: Dont't set screen on __GLXConfig
  • Jon TURNEY:
  • hw/xwin: Fix for "glx: Remove screen number from __GLXconfig"
  • Keith Packard:
  • Update to version 1.14.99.3
  • Michele Baldessari:
  • Fix Xephyr compilation when DEBUG is enabled
  • Xephyr: restore cursor visibility
  • Peter Hutterer:
  • dix: only allow button and key events to freeze a sync'd pointer
  • include: change grabtypes to start at 1
  • dix: don't attempt to deliver an event for a different grabtype
  • dix: only deliver for the current grab type
  • sync: compress two if statements
  • sync: always call BracketValues when recalculating upper/lower brackets
  • sync: supply the counter from IdleTimeBlockHandler
  • dix: provide accessor methods for the last device event time
  • sync: if the idle time was reset, force alarms to trigger (#70476)
  • sync: always set the brackets (#59644)
  • sync: split updating and triggering a counter up
  • Pino Toscano:
  • xfree86/hurd: include
  • os: move for any !win32 system

New in version 1.14.3 RC2 (September 2nd, 2013)

  • Keith Packard (1):
  • Xi: Allow clients to ask for 2.3 and then 2.2 without failing
  • Maarten Lankhorst (2):
  • Xi: Clamp XIClient maximal version to XIVersion
  • test/xi2: fix protocol-xiqueryversion test
  • Peter Hutterer (4):
  • dix: check the xi2mask, not the grab type for touch listeners
  • dix: set the valuator mask to ensure XI 1.x events have data
  • test: fix XIQueryVersion check for new behaviour
  • Revert "test: fix XIQueryVersion check for new behaviour"

New in version 1.15 Snapshot (April 25th, 2013)

  • Aaron Plattner (1):
  • fb: Rename wfbDestroyGlyphCache
  • Alan Coopersmith (8):
  • Handle failure to create counter in init_system_idle_counter
  • Stop leaking overlayWin in PanoramiXCompositeGetOverlayWindow error paths
  • Free keymap on error in Xephyr's hostx_load_keymap
  • Make xf86ValidateModes actually copy clock range list to screen pointer
  • Avoid NULL pointer dereference in xf86TokenToOptinfo if token not found
  • Avoid memory leak on realloc failure in localRegisterFreeBoxCallback
  • xf86XvMCScreenInit: Avoid leak if dixRegisterPrivateKey fails
  • Avoid memory leak in ddc resort() if find_header() fails
  • Andreas Wettstein (1):
  • xkb: Fixes to LatchMods/LatchGroup
  • Bryce Harrington (6):
  • xfree86: (Cleanup) Close fd if drm interface 1.4 could not be set.
  • xfree86: Track error code and add label for error handling.
  • xfree86: Provide more details on failure
  • xfree86: Keep trying to set interface on drm for 2 seconds.
  • xfree86: Fix race condition failure opening drm.
  • xfree86: Be verbose if waiting on opening the drm device
  • Daniel Martin (2):
  • ephyr: Add -resizeable option
  • ephyr: Fix crash on 24bpp host framebuffer
  • Dave Airlie (1):
  • xf86: fix flush input to work with Linux evdev devices.
  • Geert Uytterhoeven (1):
  • kdrive/fbdev: revive randr new screen size logic
  • Jeremy White (1):
  • Define prototypes for hw/xfree86/modes/xf86Modes.c only in xf86Modes.h.
  • Keith Packard (2):
  • Bump release to 1.14.99.0
  • Bump to 1.14.99.1
  • Marcin Slusarz (1):
  • os: use libunwind to generate backtraces
  • Peter Harris (1):
  • xkb: Set nIndicators in XkbGetIndicatorMap
  • Peter Hutterer (20):
  • randr: fix "set but unused" warnings
  • xfree86: drop unused prevSIGIO
  • fb: drop two unneeded shadowing variables
  • Xext: renaming shadowing variable
  • Xext: rename two shadowing variables
  • xkb: remove unused variable 'names'
  • xfree86: remove redundant declaration of inputInfo
  • dix: FreeAllAtoms() on reset
  • dix: only show the cursor if a window defines one (#58398)
  • os: document pnprintf as sigsafe snprintf
  • kdrive: fix "set but not used" warnings
  • xephyr: fix "set but not used warnings"
  • xfixes: ifdef PanoramiXFixes* (#62015)
  • Xi: return !Success from DeliverTouchEmulatedEvent if we didn't deliver
  • Xi: use a temp variable for the new listener
  • dix: fix a comment
  • Xi: compress two if statements with the same body
  • dix: update coords for touch events in PlayReleasedEvents
  • Xi: add a comment to make a condition a bit clearer
  • Revert "kdrive: fix "set but not used" warnings"
  • Piotr Dziwinski (1):
  • glx: fix uninitialized var in __glXDRIscreenProbe
  • Robert Morell (2):
  • configure.ac: Require inputproto 2.3
  • list.h: Make xorg_list_init inline
  • Tomasz Lis (1):
  • Full support of sRGB capable fbconfigs.
  • Zack Rusin (1):
  • GLX/DRI2: Do not expose INTEL_swap_event without swap control

New in version 1.14.1 (April 18th, 2013)

  • xf86: fix flush input to work with Linux evdev devices.
  • Xi: add a comment to make a condition a bit clearer
  • dix: fix a comment
  • configure.ac: Require inputproto 2.3
  • list.h: Make xorg_list_init inline

New in version 1.14.0 RC2 (February 14th, 2013)

  • Aaron Plattner (4):
  • dix: Make small bitfields that store enums unsigned
  • config/udev: fix "removing GPU device" format string mistake
  • xf86: use nt_list_for_each_entry_safe to walk InputHandlers in xf86Wakeup
  • randr: unref the provider shared pixmap the appropriate number of times
  • Alan Coopersmith (1):
  • Xext: avoid null-pointer dereference in XTestFakeInput (#59937)
  • Andreas Wettstein (2):
  • xkb: Do not use base group as an array index.
  • xkb: Fix repeat behaviour of redirect and message actions
  • Benjamin Tissoires (1):
  • dix: fix error logging occuring in signal context of GetTouchEvents
  • Bryce Harrington (2):
  • xfree86: Use fbdev/vesa driver on Oaktrail, Medfield, CDV rather than -intel
  • xfree86: Man page shouldn't say Device is mandatory anymore
  • Carlos Garnacho (3):
  • mi: Ensure pointer emulating touch events update the sprite
  • render: Unwrap early on the animated cursor BlockHandler
  • dix: Set focus field on XI2 crossing events
  • Dave Airlie (3):
  • xf86: bump input ABI version to 19
  • xserver: fix build regression since 91ab237358c6e33da854914d3de493a9cbea7637
  • randr: bump advertised RandR version to 1.4
  • Jasper St. Pierre (1):
  • protocol-versions: Bump minor version of XI
  • Jon TURNEY (10):
  • hw/xwin: Bring the X screen window to the front on a single left-click on the tray icon
  • hw/xwin: Fix some comments in winkeybd.c
  • hw/xwin: Process one Windows message per wakeup, rather than all of them.
  • hw/xwin: Fix MinGW build of winSetAppModelID.c
  • hw/xwin: Move reshape code from winUpdateWindowPosition() to the map event handler
  • hw/xwin: Add a new WM_WM_HINTS_EVENT event to update window style
  • hw/xwin: Make sure that WM_WM_HINTS_EVENT does nothing for override-redirect windows
  • hw/xwin: Ensure full styling is applied when the window is mapped
  • hw/xwin: Use ITaskBarList interface to ensure show-on-taskbar state is updated correctly
  • hw/xwin: Stop assuming WS_EX_APPWINDOW style in WM_SHOWWINDOW
  • Keith Packard (5):
  • input: Pull TouchListener declaration to top-level
  • input: Record grab pointer in TouchListener
  • hw/xfree86: Require only one working CRTC to start the server.
  • os: Round fraction in pnprintf %f format
  • Update to version 1.3.99.902 (1.14 RC2)
  • Peter Hutterer (26):
  • test/xi2: fix compiler warning
  • dix: add resource type to touch listeners
  • dix: check for the right device's xi2 mask
  • dix: remove already-moved hunk
  • dix: don't filter RawEvents if the grab window is not the root window (#53897)
  • Fix two typos "requires an string value"
  • xfree86: set event->detail for DGA pointer events
  • xfree86: update the device state for all DGA events (#59100)
  • xfree86: don't access the old input handler after freeing it
  • dix: typo fix in comment
  • test/signal-logging: simplify tests using sprintf
  • os: silently ignore length modifiers in pnprintf
  • os: add support for %f to pnprintf
  • dmx: don't include dmx-config.h from xdmxconfig (#37502)
  • Xi: limit valuator copy to valuator array size (#59939)
  • include: fix typo in list description
  • Xext: pass the current screen to miProcessDeviceEvent() from xtest calls
  • Xext: if a root window is given in XTestFakeInput, move to that
  • dix: fix ptraccel debugging printfs
  • dix: add some more info to a ptraccel debug msg
  • dix: use BUG_RETURN_VAL for an error message
  • dix: unify prefix for ptraccel debugging in DebugAccelF macro
  • dix: pre-scale relative events from abs devices to desktop ratio (#31636)
  • dix: when shutting down slave devices, shut down xtest devices last
  • dix: support the transformation matrix for relative devices.
  • Short-cut the input device cleanup process during AbortServer()
  • Ryan Pavlik (2):
  • include: Add RELOCATE_PROJECTROOT to xwin-config.h header
  • hw/xwin: Add missing include xwin-config.h to winglobals.h
  • Sybren van Elderen (1):
  • dix: when scaling from desktop coord, take the total desktop size (#51904)
  • Ted Felix (1):
  • xfree86: bail on misformed acpi strings (#73227)

New in version 1.13.1 (December 14th, 2012)

  • xf86: select a fake output for headless servers
  • Xi: Fix modifier swapping in XIPassiveGrabDevice
  • randr: call RRProviderInit in the proper place.
  • cursor: Revise edge cases for the pointer moving towards barriers
  • xf86: fix compat output selection for no output GPUs
  • When resetting device idle time, reset XIAll(Master)Devices too
  • xkb: fill in keycode and event type for slow keys enablement
  • hw/dmx: add update_desktop_dimensions() call
  • Sync TouchListener memory allocation with population in TouchSetupListeners()
  • Xi: Update the device after delivering the emulated pointer event(#56558)
  • Xi: Set modifier mask on touch events
  • End physically active touches when device is disabled
  • Xi: Don't check for TOUCH_END, it's never set
  • Touch: Fix duplicate TouchBegin selection with virtual devices
  • dri2: invalidate drawable after sharing pixmap
  • xf86/platform: scan pci after probing devices
  • config/udev: ignore change on drm devices
  • xf86: fix multi-seat video device support. (v2)
  • Correct description of -displayfd option in man page.
  • hw/xwin: Only add GLX extension once.
  • Fix compilation of Xorg DDX without XF86VIDMODE
  • Fix 'make distcheck' for hw/xwin
  • EXA: Track source/mask pixmaps more explicitly for Composite fallback regions.
  • dix: set the device transformation matrix
  • dix: fix crash on XI 1.x grabs on disabled devices. (#54934)
  • xfree86: add xf86UpdateDesktopDimensions()
  • dix: fix crash on shutdown if a disabled device is still grabbed (XI1 grab)
  • xkb: ProcesssPointerEvent must work on the VCP if it gets the VCP
  • Xi: don't deliver TouchEnd to a client waiting for TouchBegin (#55738)
  • dix: fix zaphod screen scrossing (#54654)
  • dix: fix scale_to_desktop for edge ABS events

New in version 1.10.2 (May 30th, 2011)

  • Aside from two crash fixes for the XQuartz DDX, it is identical to RC2.

New in version 1.7.2 RC2 (November 20th, 2009)

  • randr: Fill in errorValue when verifying outputs/crtcs/modes
  • Move FD_CLR above pInfo->read_input
  • Xming: Add mouse motion events with flag POINTER_SCREEN in winEnqueueMotion()
  • loader: actually stat something that has some chance of existing. (v2)
  • XQuartz: Controller thread launches clients
  • XQuartz: Don't weed out duplicates in generated keymap
  • XQuartz: Use dixLookupResourceByType instead of LookupIDByType
  • XQuartz: Cleanup X11Controller.m compilation warnings.
  • XQuartz: Buildfix for Leopard and older
  • XQuartz: pbproxy: Wait for the server to finish starting up, so display is valid.
  • XQuartz: Explicitly pass a bellProc to make XBell() work again.
  • Cygwin/X: Remove a couple of extraneous '
  • ' in logged version info
  • Avoid a null dereference if IFF_BROADCAST is set but there is no broadcast address
  • xfree86: set a sane umask before opening the log
  • dix: increase default number of buttons to 10.