Age | Commit message (Collapse) | Author | Files | Lines |
|
It is distfile change only.
|
|
commit 09ef847d16333d6ae04c3f88ff7f535381cb9405
Date: Mon Jan 7 23:48:12 2008 +0100
Prevent cpp from prefixing errors with <stdin> when the
xrdb input is a file
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=458246
|
|
commit 427023df02bd38ff9dc2af2a2059466f37f4cfd8
Date: Wed Dec 19 15:17:34 2007 +1030
Fix build under autoconf 2.57
configure.ac specifies a max version of 2.57 but AC_CHECK_HEADERS_ONCE
was only introduced to autoconf in 2.59c.
Drop the _ONCE to make autoconf happy
commit e09956f244099ddd36b1a2cd5d7800d5fc7120c1
Date: Thu Dec 6 05:30:43 2007 -0500
Select correct client window in WMs using virtual roots
|
|
10.16.4:
Video Overlay: Handle clipping correctly with AutoPaint colorkey.
10.16.3:
minor Xv fixes
10.16.2:
More xorg and xfree86 backwards compatibility
Fix VMWARE_CTRL and Xinerama Initialization
Prepare for 10.16.2
|
|
2.0.0:
Remove unused ->device
Un-duplicate some init code.
vesa 2.0.0
1.99.1:
Don't disable FB access when it's already disabled.
Remove all trace of mfb
Bug #10004: Fix palette initialization in 8bpp modes.
Fix boolean inversion in VBE screen clear.
Remove some useless module loading cruft.
Prefer a more modern depth/bpp selection.
Add semi-reasonable mode validation.
Clean up memory model checking, and properly refuse unsupported models.
Update copyright.
vesa 1.99.1
Bug #11090: xf86-video-vesa COPYING file
Add conditional support for pci-rework branch.
Correct '#ifdef' to '#ifndef'. Oops.
Use XSERVER_LIBPCIACCESS to autodetect libpciaccess usage.
Use pci_device_map_range instead of pci_device_map_memory_range
Rename .cvsignore to .gitignore
Add *~ to .gitignore to skip patch/emacs droppings
fixup pciaccess version detect
Planar video is gone
Makefile.am: nuke RCS Id
|
|
Death to RCS tags.
Fix distcheck
Finish making xf1bpp/xf4bpp optional.
tseng 1.2.0
Replace XFree86 references in man page
renamed: .cvsignore -> .gitignore
blind pciaccess conversion
oops forgot pcirename file
Makefile.am: nuke RCS Id
|
|
1.3.0:
Death to RCS tags.
Fix distcheck
trident 1.3.0
initial trident pciaccess support
Makefile.am: nuke RCS Id
1.2.4:
changes unknown
|
|
0.10.0:
Death to RCS tags.
Fix distcheck
sis 0.10.0
inital SiS pci access patch
fix warnings in build
fix thinko between open/fopen
another open/fopen confusion
fixup type punning
convert all hostbridge read/writes to accessor functions
finish off read/write long conversion
switch bios reading over
use map domain memory with pciinfo
fixup unmap memory
fix some thinkos in the pciaccess patch, this now works on real hardware
Require pciaccess >= 0.10.0 for pci_device_map_range()
fixup pciaccess version detect
Makefile.am: nuke RCS Id
switch vga over to pciaccess
0.9.4:
bump Xv image height up from 1080 to 1088 for 315 chips
Define SIS_*_VERSION using PACKAGE_VERSION*
Convert callers of LookupWindow() to dixLookupWindow().
Revert "Convert callers of LookupWindow() to dixLookupWindow()."
Rename .cvsignore to .gitignore
Add *~ to .gitignore to skip patch/emacs droppings
Fix xserver version detection.
Get rid of the XFree86Server macro.
Bump to 0.9.4.
|
|
Death to RCS tags.
Fix distcheck
siliconmotion 1.6.0
more clock fixes
hostdata blit UTS works, but SW is faster
minor fixup
define SILICONMOTION_VERSION* using PACKAGE_VERSION*
smi: add pciaccess support
Fix UTS issues with flash
Makefile.am: nuke RCS Id
|
|
2.2.1:
Don't try to call vbeModeInit(), it doesn't exist and never will.
Fake sync ranges for panels with no EDID.
savage 2.2.1
no, really, savage 2.2.1
EXA upload to screen must honor pitch of the source data
savage: fix typos in error message
savage/pciaccess: don't try to map aperture larger than region window
2.2.0:
Death to RCS tags.
Remove use of deprecated {DE,}ALLOCATE_LOCAL.
savage 2.2.0
fix segfault from pScrn->currentMode
fix the build for the non-pci-rework case (compile tested only)
Fix reversed logic in SavageMapMem()
Fix broken downscale past 2:1 - MM8190 and MM8198 can be used together
for arbitrary downscale
BCI can only handle widths that are multiple of 16, so only use BCI for these
Prevent use of BCI for YV12 -> YUY2 conversion from scribbling past the
end of the allocated video buffer.
Reset lastKnownPitch to 0 right after enabling streams.
Do not wait for vertical retrace anymore on old streams engine.
Make sure MMIO region is mapped VIDMEM_MMIO, not VIDMEM_FRAMEBUFFER
Fix ambiguous operator precedence
Implement driver option IgnoreEDID
Minimal change set to convert driver to the new PCI interfaces. Enjoy.
dos2unix
Merge PCI-rework support and "classic" PCI interface support into one tree
Update to use new libpciaccess interfaces for mapping with MTRR support.
Fix odd typo.
Rename .cvsignore to .gitignore
Add *~ to .gitignore to skip patch/emacs droppings
Include <unistd.h> to get a declaration for usleep()
fixup pciaccess version detect
Makefile.am: nuke RCS Id
Bug #11139: Remove some dead code.
|
|
1.10.1:
Bug #14298: Fix mode validation to be less ridiculous.
Nuke the useless UseFB option.
Bug #13243: Re-enable 24+32 framebuffer.
Fix typo in S3VValidMode.
1.10.0:
Death to RCS tags.
Fix distcheck
s3virge 1.10.0
Replace references to XFree86/XF86Config in man page
renamed: .cvsignore -> .gitignore
Define S3VIRGE_VERSION_* using PACKAGE_VERSION*
s3virge: pciaccess conversion
oops fixed pointed out by pcpa
Makefile.am: nuke RCS Id
Clean up: Remove some mga remains
|
|
Death to RCS tags.
Fix distcheck
s3 0.6.0
renamed: .cvsignore -> .gitignore
Was previously done in monolith Imakefile as:
Define VERSION* using PACKAGE_VERSION*
s3: pciaccess conversion
Add non-vbe DPMS support for TRIOs, adjust max clock on Trio64V2
|
|
Death to RCS tags.
Fix includes to silence a warning.
rendition 4.2.0
Man page updates - replace references to XFree86 & changes for XF86 4.0
RENDITION_VERSION* using PACKAGE_VERSION_*
pciaccess: make auto-detect work
pciaccess: do unmap path
Supply NULL driverFunc pointer in DriverRec.
Add prototype for renditionGetRec.
Move renditionTypes, renditionChipsets, and renditionPCIchipsets earlier
in file.
Port driver to PCI-rework interfaces.
Makefile.am: nuke RCS Id
|
|
1.2.1:
fbpseudocolor no longer exists, don't try to use it.
neomagic 1.2.1
Compile warning fix.
Don't include fbpseudocolor.h, as it is no longer available.
1.2.0:
Death to RCS tags.
neomagic 1.2.0
Replace references to XFree86/XF86Config in man page
renamed: .cvsignore -> .gitignore
Define NEO_*_VERSION using PACKAGE_VERSION_*
neomagic: port to libpciaccess
neo: finish pciaccess conversion
finish pciaccess conversion
Makefile.am: nuke RCS Id
|
|
Update PInS defaults for G200SE-A.
Filter high-bandwidth modes on G200SE.
No depth 15 on G200SE-A.
Updates to memory sizing on G200SE.
Adjust memory timings on G200SE-A.
Change initial memory sizing on G200SE.
Disable I2C on G200SE for now.
Death to RCS tags.
mga 1.4.9
cfb8_32: Gut a remnants of support for 24-bit with 8-bit overlay
Initialize default BIOS values from a data table instead of from code
Get memory probe size and stride (offset) values values from a data table
instead of from code
I2C: First round of clean-ups
I2C: Second (and final) round of clean-ups
G200EV: Enable support for G200EV
Makefile.am: nuke RCS Id
Replace ALLOCATE_LOCAL/DEALLOCATE_LOCAL with xalloc/xfree.
For compatibility, accept SWcursor as option.
Bug #15836: Fix DDC on G200SE.
Bug #16545: Add G200WB support.
|
|
Death to RCS tags.
Fix distcheck
i128 1.2.0
Replace references to XFree86/XF86Config in man page
renamed: .cvsignore -> .gitignore
Define I740_*_VERSION* using PACKAGE_VERSION_*
i740: pciaccess conversion
Makefile.am: nuke RCS Id
Minor fixes to compile with pciaccess.
|
|
Death to RCS tags.
Fix distcheck
i128 1.3.0
Replace references to XFree86/XF86Config in man page
renamed: .cvsignore -> .gitignore
Define I128_*_VERSION* using PACKAGE_VERSION_*
i128: pciaccess conversion
i128: drop ansi C wrapper
i128: fixup typo
Makefile.am: nuke RCS Id
|
|
1.2.1:
Fix regression from pciaccess conversion (bug#16060)
Bump to 1.2.1
1.2.0:
Fix distcheck
glint 1.2.0
Fix --disable-dri for glint driver
Define GLINT_*_VERSION* using PACKAGE_VERSION_*
half attempt at pciaccess conversion for glint - doesn't build yet
fix typos in dac code from last commit
hey look a flying pig....
fix regression in tinderbox
Rename .cvsignore to .gitignore
Add *~ to .gitignore to skip patch/emacs droppings
Add missing #includes to fix compiler warnings.
Makefile.am: nuke RCS Id
|
|
1.2.1:
Bug #10715: Fix depth configuration for alpine chips.
cirrus 1.2.1
1.2.0:
Replace references to XFree86/XF86Config in man page
renamed: .cvsignore -> .gitignore
Define CIR_*_VERSION* using PACKAGE_VERSION_*
cirrus initial pci access support
test for 1/4 bpp in case server is configured with them off
make dist check pass
fixup iomapsize so we only map the 4k not 16k which is an error
cirrus: set default depth to 24bpp
Makefile.am: nuke RCS Id
|
|
Death to RCS tags.
Fix distcheck
chips 1.2.0
Define CHIPS_*_VERSION* using PACKAGE_VERSION_*
chips: pciaccess conversion
add another missing rename file
chips: remove ALLOCATE_LOCAL
chips: optionalise 1/4bpp support
Rename .cvsignore to .gitignore
Add *~ to .gitignore to skip patch/emacs droppings
Makefile.am: nuke RCS Id
Fix Xv with scaling
|
|
Fix distcheck.
ast 0.85.0
Define AST_*_VERSION* using PACKAGE_VERSION_*
ast: pciaccess conversion
Rename .cvsignore to .gitignore
Add *~ to .gitignore to skip patch/emacs droppings
Compile warning fixes.
compiler warning fixed
modified: ChangeLog
Make sure used symbols are of public visibility
Fixed can't get correct MCLK in some platforms
|
|
Death to RCS tags.
ark 0.7.0
Define VERSION* using PACKAGE_VERSION*
ark: add basic pciaccess support
Rename .cvsignore to .gitignore
Add *~ to .gitignore to skip patch/emacs droppings
|
|
Remove useless mfb.h include.
Death to RCS tags.
apm 1.2.0
Replace references to XFree86/XF86Config in man page
renamed: .cvsignore -> .gitignore
Define APM_*_VERSION using PACKAGE_VERSION_*
pcirename conversion
xf1/4 bpp conversion
further build fixes
Naming change: Security*Access -> Dix*Access
Makefile.am: nuke RCS Id
|
|
with this enabled for a while now with no untoward effects.
|
|
This is version update only, no functional change in this ruby extention.
|
|
Vinay Bondhugula (2):
Compatibility with older versions of X
Don't call conversion_proc for older Xservers
|
|
its own, starting with "VirtualEvent" (Which is correctly set to LASTEvent,
defined in include/X11/X.h). In xproto-7.0.13, a new event "GenericEvent" was
added - making the defined array broken for all of Tk's internal events.
The easy fix is to just add in the missing event into the hard-coded array.
This patch was reported here: http://bugs.gentoo.org/show_bug.cgi?id=225999
A cleaner fix, but a much bigger patch, is listed in the Tk bug tracker. I'm
punting that Tk will have been updated with the fix before X.h grows another
event.
http://sourceforge.net/tracker/index.php?func=detail&aid=2010422&group_id=12997&atid=112997
Bump PKGREVISION.
|
|
Patch #236 - 2008/07/27
* correct memory reallocation when handling a paste of UTF-8 text from
patch #225 changes (report/patch by Max Mikhanosha).
* correct allocation of temporary buffer in xtermFindShell in case the
user's $PATH contains no ":" (report/analysis by Victor Stinner,
Freedesktop.Org Bugzilla #16790).
* modify CF_XOPEN_SOURCE to add case for DragonFly BSD, to fix new
compile problem exposed by fix for fd_mask (patch by Hasso Tepper).
* add configure-check for ncurses use_extended_names, (report by
Martin Mokrejs).
* correct computation for toolbar height; layout manager already takes
into account borderWidth resource.
* implement VT320-style SCS (select character set) for ISO Latin-1
supplemental.
* fixes for vt100-style character sets in UTF-8 mode (Ubuntu #230919).
* fix to make luit work with xterm's -ls option (report/patch by
Marius Tolzmann).
* update config.guess, config.sub
|
|
changes:
Xinerama/RANDR fixes: this time for sure. It should now work to add/remove
monitors or resize screens at any time.
New hack, skytentacles.
New version of gleidescope.
Added the -log option to the xscreensaver daemon, since a truly shocking
number of Linux users seem to have no idea how to redirect output to a
file.
Added -duration arg to most image-loading hacks, so that they pick a new
image every few minutes.
Added an ATM crash to BSOD.
|
|
Adam Jackson (2):
C sucks: define XEventClass in terms of unsigned int, not CARD32.
Typo fix.
Peter Hutterer (2):
Add DeviceControlChanged define.
inputproto 1.4.4
|
|
Adam Jackson (1):
libSM 1.1.0
Alan Coopersmith (1):
Fix --without-libuuid to not call PKG_CHECK_MODULES for libuuid
Dan Williams (1):
Bug #14949: Don't use gethostbyname() for client IDs.
James Cloos (2):
Add *~ to .gitignore to skip patch/emacs droppings
Replace static ChangeLog with dist-hook to generate from git log
|
|
This release adds chip names for the GeForce 9 mobile chips and the GeForce GTX
GPUs. It also adds code to read DDC-based EDIDs for LVDS panels that have such
a thing.
- -- Aaron
Aaron Plattner (9):
GeForce 9 mobile chips.
GeForce GTX 280 and 260 chip names.
Replace copyright notices with stock MIT X11 boilerplate.
Add new chips to the man page and fix capitalization of "Quadro".
Add a note that MODE_PANEL really means "larger than
BIOS-programmed panel size".
G80: Handle extended I2C ports and LVDS panels with DDC-based EDIDs.
Fix build by using CARD32 instead of uint32_t, like we do
everywhere else.
More G8x chips.
Bump to 2.1.10.
|
|
James Cloos (4):
Rename .cvsignore to .gitignore
Add *~ to .gitignore to skip patch/emacs droppings
Replace static ChangeLog with dist-hook to generate from git log
xwd 1.0.2
Jeremy Huddleston (1):
Added missing AM_PROG_CC_C_O
Kim Woelders (2):
Fix window selection by pointer.
Remove unused code, ANSIfy, add const.
Niveditha Rau (1):
Add -silent to options in man page synopsis
|
|
Main new features included within this release are
- New Intel 4 series chipset support
- Improved 965 exa render performance
- Integrated HDMI support
- SDVO-HDMI support (e.g ASUS P5E-VM board)
Also a long list of bugfixes:
http://lists.freedesktop.org/archives/intel-gfx/2008-July/000192.html
|
|
- ANSIfy.
- Use standard headers instead of rolling custom wrong decls.
- Don't use void main.
- Fix y2038 bugs.
- Prune dead code.
- Stop using outdated timezone interfaces.
- Don't need libcompat any more.
- Generally clean up build.
Should fix broken build on Linux, probably other non-BSD OSes as well.
PKGREVISION++.
|
|
|
|
Thu Jul 24 02:20:00 2008
|
|
define an INT32 type (and incompatible ones even, since Xmd's is
unsigned long on ILP32." (from GIT)
Bump PKGREVISION.
|
|
Qt 4.4 introduces many new features as well as many improvements and
bugfixes over the 4.3.x series. For more details, see the online
documentation which is included in this distribution. The
documentation is also available at http://doc.trolltech.com/4.4
The Qt version 4.4 series is binary compatible with the 4.3.x series.
A long list of changes can be found in the changes-4.4.0 file in the
distribution.
Update OKed (and initial work provided) by adam@..
|
|
want -pthread in XTHREAD_CFLAGS, as this will end up linking everything
against libpthread. Thus, in keeping with our gcc specfile, define
_REENTRANT in XTHREAD_CFLAGS instead. XTHREADLIB is now
-pthread, not -lpthread.
While here, update to head of libX11-1.1 branch in GIT (with the
exception of some unicode changes which caused mkpatches to choke).
changes:
- IM: Respect XMODIFIERS for Thai locale (bug #15719)
- Bug #15884: Remove useless sleep()'s from the connection code.
- NLS: Make UTF-8 the default for Russian
- ConnDis: properly cast 'addr' before accessing it as a byte array.
- added error check in Xcms color file parser; closes bug #15305
- Fix missing error condition in cmsColNm.c, also.
Bump PKGREVISION.
|
|
commit 9ab3d5cca3cf9a76c6873007b8d4345c838cc297
Bug #16149: Make the field not be named 'delete' in C++ mode.
Thanks for the upward compatibility, Bjarne.
commit 3243afaa593f95bb89b1381dac2b920111ce36b1
Correct a copy'n'paste error in RRGetOutputInfo spec text.
commit d1c5739ae19f7cc01a6a5603d3fae93e1f9ee3d1
Delete duplicate B32's in xRRScreenChangeNotifyEvent
Caused compiler warnings:
"randrproto.h", line 567: warning: syntax error:
empty member declaration
"randrproto.h", line 568: warning: syntax error:
empty member declaration
|
|
|
|
|
|
Patch stolen from libXv's b3.mk.
|
|
This fixes build problems on darwin
|
|
|
|
Correct the sse2 detection logic in configure/configure.ac.
Bump PKGREVISION.
|
|
It create broken binary package for SSE-less machine, noticed in PR 39117.
Such a tuning parameters should be defined in /etc/mk.conf.
Bump PKGREVISION.
|
|
(just a reminder: when updating this package, be sure to adjust
files/pixman.sym appropriately)
commit 411c0e990f7a96d4e15f2cbbe07d3b50b6a20f95
Make a couple of functions static
commit 96f57c07f24cd6d86c0aad624bd1a16b85f08e04
Speed up fbOver
Use FbByteMulAdd to operate on two components at a time and force
the function to be inlined.
commit dcbe4b3f80fb8dc6a83efa7e66a891f8b102608d
fix --enable-{mmx,sse2,vmx}
If --enable-foo is used, don't treat it as --disable-foo, and
error out if the appropriate compiler support isn't detected.
commit e21f00db50c30e80c8969fbf63982aeb7434dc1d
Add pixman-1-uninstalled.pc file
commit 1570746350baf82d36aeb7ea1702350700b795d7
Use -no-undefined -- fixes bug 15927
commit 99fead412e7602daf193035831b952e252dac0e1
Use <inttypes.h> on most types of Unix.
Fix for bug 15560.
commit 2070f10db927a97d7d38024e607093f5a6e00291
Remove unused pixman-combine.c
commit 9d54568d5f93701313fdbb49c1ef9dc79fe5850c
Fix SSE2 bug where x888 pixels were treated as 8888
commit 7dfd023e94d9ab5940d7062c2e5cde11ab02f43e
fix Altivec detection
the old code used to cause infinite looping on G3 machines.
commit 0a92401678286eb438fe24979fd032efba540a0a
Fix fbCompositeSrc_x888xnx8888mmx to properly ignore source alpha
commit 687176023caf53f2fe234827d152f270048b1cd3
make --{en,dis}able-gtk work correctly
--enable-gtk had the same effect as --disable-gtk.
Now we check for it by default, and error out if we can't
find it and it was explicitly enabled.
commit 9d97716d2fd91d5611a5e7a7b29b38887b672e9f
Fix bug 16310 in the SSE2 fast path in function
fbCompositeSolidMask_nx8888x0565Csse2
commit eb53d111bd8651cd00c1b728b09e8be09b8482f7
Add a lossy 32-bit generic pixel fetch function for wide surfaces.
The transformed fetch path currently only works at 32-bit precision.
Until a wide version of that function is added, we need to have a
path to perform a wide FetchPixel and then contract it down to
a8r8g8b8.
Also, use the right format when expanding the result in
fbFetchTransformed64. fbFetchTransformed returns a8r8g8b8 results.
commit 3c43b869f3821495978c61b9195f0b0e9e9e1245
Unexport or delete various functions
- Delete pixman_region_append()
- Delete pixman_region_empty()
- Make pixman_region_validate() static and don't export it.
Reported by Julien Cristau.
commit 36b05f2cac7fa28e31131ca0d6b8b133ccd95ab8
Only export the 16 bit version of pixman_region_set_static_pointers()
|
|
From upstream commit 55678c64bc6e3ed613ea6db14c105c18a0cf28ce.
|