* update applications for 4.01
* various fixes in windows port
* add some GTK enumerations and update stock icon list
* add properties GtkTreeView.enable_{tree,grid}_lines
* add properties GtkEntry.{primary,secondary}_icon_{stock,name,pixbuf},
see examples/entry2.ml for usage
* add tags in GtkMovementStep
* fix compatibility with ocaml 4.01 (?lab for non-optional arguments)
* add handling of new modifiers
* detect findlib during configuration
* support DESTDIR with findlib-install
* have old-install copy the META file too
* cleanup the two phases of findlib-install
The DRI3 extension provides mechanisms to translate between direct
rendered buffers and X pixmaps. When combined with the Present
extension, a complete direct rendering solution for OpenGL is
provided.
The direct rendered buffers are passed across the protocol via
standard POSIX file descriptor passing mechanisms. On Linux, these
buffers are DMA-BUF objects.
DRI3 also includes a mechanism to translate between Linux Futexes
and X Sync extension Fences. This provides a synchronization
mechanism which can be used to serialize access to shared render
buffers.
The Present extension provides a way for applications to update
their window contents from a pixmap in a well defined fashion,
synchronizing with the display refresh and potentially using a more
efficient mechanism than copying the contents of the source pixmap.
1.8.0:
Consider this the NetBSD release, bringing the driver closer to sync with
the NetBSD source directory. Non-BSD platforms should not be affected by
these changes.
Peter Hutterer (1):
keyboard 1.8.0
Thomas Klausner (25):
Added some missing keysyms to the ADB keymap.
Map the Sun keyboard "help" key.
Added keysym for Power keys.
Added keysyms for extra keys on Sun Type 6 USB keyboard.
Fixes for the ADB map.
More ADB fixes.
Add translation table for AT/PS2 pseudo XT scancodes.
Use existing wscons USB keymap for WSKBD_TYPE_MAPLE on dreamcast.
Add a couple mappings for keys found on Sun keyboards.
Fix scan code 50 key on non-US-USB keyboards.
Add a basic map for an LK201 keyboard.
Set the version for the keyboard.
Fix a shadowing warning, and add const.
Recognize WSKBD_TYPE_MAPLE for dreamcast.
Add default device for Protocol "wskbd".
Add WSKBD_TYPE_LK201.
Make OpenKeyboard() warn instead of error out.
Fix wskbd handling when VT switching.
Improve auto-detection.
Move WSCONS case inside switch statement.
Add some NetBSD feature detection code.
Do not assume type in WS_*_MAP_SIZE macros.
Factor out common code.
Check open() return value.
Whitespace fixes.
1.7.0:
No big changes, but since we removed AM_MAINTAINER_MODE might as well bump
the minor release number.
Adam Jackson (1):
configure: Drop AM_MAINTAINER_MODE
Colin Walters (1):
autogen.sh: Implement GNOME Build API
Peter Hutterer (2):
Use sigsafe logging for keyboard debug messages
keyboard 1.7.0
pkgsrc changes:
- remove obsolete post-3.2.2 fixes
- adjust patches per changes in original files
Changes from doc/en/ReleaseNote:
ver 3.3.0
* Support framebuffer on NetBSD/x68k. (Thanks to NetBSD/x68k developers)
* Support Android. (Experimental)
* Update the colormap instead of searching the closest color when
OSC 4 ; c ; spec changes RGB of the color number on framebuffer.
* Support OSC ? 80 h (disable sixel scrolling).
* Support U+XXXX-XXXX key in 'font' configuration file.
* Add "unicode_noconv_areas" (a.k.a --ucsnoconv) option.
* Add "allow_scp" (a.k.a --scp) option.
* Remove SWITCH_OSC52. (Use "proto:allow_osc52=switch" instead.)
* OSC 5383 reqests challenge sequence.
* Bug fixes:
#70 (Thanks to Thomas Wolff)
Fix the bug which disabled redrawing the margin area on framebuffer.
Fix the bug of the transparent color of sixel graphics on win32.
OSC 4 has an effect on RGB of the window background color.
Send a sixel next line character (-) following a sixel dump.
Fix the bug which disabled Mod+XXX shortcut key settings on win32.
Fix the bug which erased the end character of each line in scrolling if DECSLRM
set the right margin at the right end is fixed.
Fix the bug of DL/IL which scrolled the screen instead of inserting or deleteing
lines from the cursor position.
Fix keypad '5' key (VK_CLEAR) on win32.
Fix the bug of ignoring the next character after sixel graphics data.
Fix the bug which makes the window smaller than necessary in resizing.
Support wsdisplay of the BBGGRR format on NetBSD/OpenBSD. (Thanks to @oshimyja san)
announcement. This has been fixed, the matching distfiles are now available.
The contents did not change at all (says Ben, and diff.
Set DIST_SUBDIR and update distinfo.
I am proud to announce xauth-1.0.8, a minor maintenance release
containing mostly bug fixes, where I want to highlight two of them:
* Fix a long standing problem that for ssh tunneled connections a
display variable of the form: localhost:<N>.<M> leads to correct
authorization when an X client is started but "xauth list $DISPLAY"
returns nothing.
* Fix for: If using GDM with XDMCP, then ssh is not able to start X11
clients on the remote side. You get a "No xauth data; using fake
authentication data for X11 forwarding." from SSH.
--Tilmann Bubeck
Alan Coopersmith (3):
Remove unused TLI ("STREAMSCONN") code from xauth
Remove unused DECnet ("DNETCONN") code from xauth
Assume signal handlers return void, as C89 requires
Dr. Tilmann Bubeck (11):
Fix various compiler warnings raised by modern gcc (4.8.1).
More fixes for compiler warnings regarding the use of "const".
Added black box tests for basic functionality started with "make check".
Rework match_auth_dpy() to be easier to read and maintain (hopefully).
Fix warning raised by "-Wshadow".
Added command "version" and command line arg "-V" to show version of xauth.
Improved README to explain "make check".
version bump to 1.0.8
Revert version bump back to 1.0.7 to prepare missing commit before 1.0.8
Rewrote test_xauth to C to make automake and "make distclean" happy.
Second version bump to 1.0.8
Egbert Eich (5):
Don't crash when running past the end of the chain
Look for FamilyLocal if inet or inet6 address is loopback
Look for FamilyLocal if inet or inet6 address is loopback
Handle v4-mapped inet6 addresses correctly
Make matching algorithm mimic XauGet*AuthByAddr
Jeremy Huddleston Sequoia (2):
Revert "Look for FamilyLocal if inet or inet6 address is loopback"
Silenece lack of noreturn warning
Jon TURNEY (1):
Include Xwinsock.h rather than sys/socket.h on WIN32
Thomas Klausner (1):
Use copystring to create non-const string.]
Alex Deucher (2):
radeon: add berlin pci ids
radeon: pad CS to 8 DW
Ben Widawsky (1):
libdrm: bump to 2.4.47
Chia-I Wu (1):
intel: silence valgrind warnings for unsynchronized maps
Damien Lespiau (3):
drm: Synchronize the stereo 3D mode flags from the kernel headers
drm: Sync the DRM_SET_CLIENT_CAP ioctl definition
drm: Introduce a drmSetClientCap() wrapper
Daniel Kurtz (1):
libdrm: Make some drm headers compatible with gcc -std=c89 -pedantic
Emil Velikov (6):
tests/*version: add missing string.h header
tests/modetest: silence compiler warning about uninitialised variables
freedreno: use local drm.h over the system one
freedreno: set automake options to include subdir-objects
intel: Update package name and description in libdrm_intel.pc
configure: add AC_LANG_SOURCE call within AC_COMPILE_IFELSE
Ilia Mirkin (2):
modetest: add a -D option to specify a device to be used
modetest: allow setting a scaling factor when showing plane
Imre Deak (1):
libdrm: add missing DRM_CAP_TIMESTAMP_MONOTONIC
Kristian Høgsberg (1):
intel: Set bo size from lseek if kernel supports it
Laurent Pinchart (23):
modetest: Fix warnings
modetest: Remove extern declarations of opt(arg|ind|err|opt)
modetest: Sort command line arguments
modetest: Add a command line parameter to select the driver
modetest: Add a command line parameter to drop master after mode set
modetest: Retrieve all resources in one go
modetest: Don't limit mode set and planes to two instances
modetest: Add a command line parameter to set properties
modetest: Allow specifying plane position
modetest: Print the plane ID when setting up a plane
modetest: Remove the -m argument
modetest: Create a device structure
modetest: Compute CRTC pipe number as needed
modetest: Remove the struct connector_arg encoder field
modetest: Store the crtc in the connector_arg structure
modetest: Store the mode in the crtc structure
modetest: Give the CRTC ID to the -P option
modetest: Split mode setting and plane setup
modetest: Rename struct connector_arg to struct pipe_arg
modetest: Support pipes with multiple connectors
modetest: Try all possible encoders for a connector
modetest: Fix line stride in SMPTE YUV packet pattern generator
modetest: Allocate NV buffers large enough for the two planes
Marek Olšák (1):
radeon: fix pitch alignment for non-power-of-two mipmaps on SI
Michel Dänzer (1):
radeon: Fix tiling mode index for 1D tiled depth/stencil surfaces on CIK
Rob Clark (5):
tests: add tilcdc to modetest/vbltest
freedreno: support either kgsl or msm
freedreno: add support for msm drm
tests: add msm
freedreno: add missing NOSYNC flag
From a4d9bf1259ad28f54b6d59a480b2009cc89ca623 Mon Sep 17 00:00:00 2001
From: Alan Coopersmith <alan.coopersmith@oracle.com>
Date: Mon, 16 Sep 2013 21:47:16 -0700
Subject: [PATCH] Avoid use-after-free in dix/dixfonts.c: doImageText()
Save a pointer to the passed in closure structure before copying it
and overwriting the *c pointer to point to our copy instead of the
original. If we hit an error, once we free(c), reset c to point to
the original structure before jumping to the cleanup code that
references *c.
Since one of the errors being checked for is whether the server was
able to malloc(c->nChars * itemSize), the client can potentially pass
a number of characters chosen to cause the malloc to fail and the
error path to be taken, resulting in the read from freed memory.
Since the memory is accessed almost immediately afterwards, and the
X server is mostly single threaded, the odds of the free memory having
invalid contents are low with most malloc implementations when not using
memory debugging features, but some allocators will definitely overwrite
the memory there, leading to a likely crash.
Reported-by: Pedro Ribeiro <pedrib@gmail.com>
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
Reviewed-by: Julien Cristau <jcristau@debian.org>
Bump PKGREVISION.
xf86-video-savage-2.3.7 provides a small bugfix for video playback
on MX/IX/SuperSavage (not sure if anyone noticed).
Otherwise just the usual catch-up to the ever-changing xserver, build
and run on 1.14, and default to EXA now that the xserver does not
support XAA anymore.
== Changes since 2.3.6 ==
Alexander Grothe (1):
Drop miInitializeBackingStore() call
Dave Airlie (2):
savage: move compat-api.h down to proper place.
savage: fix default accel arch when built with no XAA
Tormod Volden (5):
Really fix default accel arch when built with no XAA
exa: Use exaGetPixmapFirstPixel() instead of devPrivate.ptr
Fix building with TRACEON defined
Fix video window height on MX/IX/SuperSavage
xf86-video-savage: bump to version 2.3.7
The release 2.10 of xkeyboard-config did not really happen because of
the technical issues, so I am happy to present 2.10.1
Most of the work in that release was made by Benno Shulenberg, who did
some massive cleanups of the codebase.
Other people contributed various layouts fixes and code improvements.
Traditional credits go to the TP for the quality translations.
xf86-video-r128 6.9.2 includes a fix for EXA color errors on PowerPC,
plus updates to build with xserver 1.13 and 1.14.
Note that also this xserver patch is needed for 1.13 and 1.14:
http://lists.x.org/archives/xorg-devel/2013-September/037971.html
All credits go to Connor Behan for keeping this driver alive!
Connor Behan (4):
Remove call to obsolete miInitializeBackingStore()
Swap pixmap bytes for a solid picture on big endian host
Minor spacing changes
Specify NULL for the platform probe function
Tormod Volden (1):
r128: bump to version 6.9.2
This release revises and extends the DEC NRCS (national replacement
character set) feature, adding several character sets and making
it usable in a UTF-8 environment. There are other fixes for a few
control sequences, improvements in font handling, and improvements
to configurability.
Adam Jackson (2):
Don't bother to include vbe.h
Remove mibstore.h
Alan Coopersmith (1):
Cleanup leading/trailing whitespace in source files
Thomas Klausner (1):
Use uint32_t for a variable that's passed to a function expecting a uint32_t.
Y.C. Chen (6):
Support 1360x768 and 1600x900 Resolutions
Modify for DRAM Initial Settings
Fixed S3 Resume Failed Issue
Fixed Display Abnormal Issue with GOP Driver
Support ShadowFB if no XAA
Bump version to 0.98.0
