Age | Commit message (Collapse) | Author |
|
seems to mitigate some hangs seen by weerd@ and matthieu@
|
|
the mesa update has gone in
|
|
ok matthieu@
|
|
- INSTALL often got doubled by patch
- empty AUTHORS files didn't get removed.
|
|
ok matthieu@
|
|
|
|
an be reverted when a mesa update goes in.
from jsg@
|
|
It fixes an endianess problem with kms and the in tree x86-video-ati
among other things.
It has been tested on various i386/amd64/sparc64/macppc machines
on a range of hardware from r100->northern islands.
|
|
versions expected by the time_t bump, this time done correctly
ok matthieu@
|
|
|
|
|
|
|
|
the kernel before sending them up to X. Reduces staircacse effects
on diagonal freehand drawings. Patch from Henri Kemppainen, tested
by edd@. Thanks.
|
|
Some luna88k specific quirks are needed in
driver/xf86-video-wsfb/src/wsfb_driver.c for now.
ok matthieu@
|
|
the dri driver causes xorg to crash with r700 (PCI HD4350). Likely
due to the incomplete support for r600/r700 currently in the tree.
|
|
code in the KMS path. Previously disabled due to
not having the required libdrm parts.
ok matthieu@
|
|
not having the required libdrm parts. UMS will still be used if
KMS is not available.
ok matthieu@
|
|
|
|
Problem reported and fix tested by J. Scott Heppler. Thanks.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This is enough to fix the screen corruption seen on Xft fonts and other
Xrender drawings.
|
|
removed from X server 1.14.
|
|
|
|
|
|
ok mpi@ jcs@ matthieu@
|
|
|
|
|
|
discovered by Ilja van Sprundel.
CVE-2013-1981 X.org libX11 1.5.99.901 (1.6 RC1) integer overflows
CVE-2013-1982 X.org libXext 1.3.1 integer overflows
CVE-2013-1983 X.org libXfixes 5.0 integer overflows
CVE-2013-1984 X.org libXi 1.7.1 integer overflows
CVE-2013-1985 X.org libXinerama 1.1.2 integer overflows
CVE-2013-1986 X.org libXrandr 1.4.0 integer overflows
CVE-2013-1987 X.org libXrender 0.9.7 integer overflows
CVE-2013-1988 X.org libXRes 1.0.6 integer overflows
CVE-2013-1989 X.org libXv 1.0.7 integer overflows
CVE-2013-1990 X.org libXvMC 1.0.7 integer overflows
CVE-2013-1991 X.org libXxf86dga 1.1.3 integer overflows
CVE-2013-1992 X.org libdmx 1.1.2 integer overflows
CVE-2013-1994 X.org libchromeXvMC & libchromeXvMCPro in openChrome
0.3.2 integer overflows
CVE-2013-1995 X.org libXi 1.7.1 sign extension issues
CVE-2013-1996 X.org libFS 1.0.4 sign extension issues
CVE-2013-1997 X.org libX11 1.5.99.901 (1.6 RC1) buffer overflows
CVE-2013-1998 X.org libXi 1.7.1 buffer overflows
CVE-2013-1999 X.org libXvMC 1.0.7 buffer overflows
CVE-2013-2000 X.org libXxf86dga 1.1.3 buffer overflows
CVE-2013-2001 X.org libXxf86vm 1.1.2 buffer overflows
CVE-2013-2002 X.org libXt 1.1.3 buffer overflows
CVE-2013-2003 X.org libXcursor 1.1.13 integer overflows
CVE-2013-2004 X.org libX11 1.5.99.901 (1.6 RC1) unbounded recursion
CVE-2013-2005 X.org libXt 1.1.3 memory corruption
CVE-2013-2066 X.org libXv 1.0.7 buffer overflows
|
|
mibstore.h defines miInitializeBackingStore() as an empty stub, and
goes away in xserver 1.14.
|
|
ok matthieu@
|
|
ok matthieu@
|
|
appropriate wscons ioctls.
ok matthieu@
|
|
|
|
A recent kernel with kernel modesetting support is required.
Thanks to jsg@ and kettenis@ for their work.
|
|
big endian an apply the correct fix, working at depth 16 and 24, kindly
provided by Michel Dänzer who already committed it upstream.
No objection from the usual macppc users.
|
|
ok miod@
|
|
|
|
successfully tested by schwarze@
|
|
on ivy bridge, mostly based on upstream drm code:
- remove a workaround which was in itself causing issues
- switch the order of disabling fdi rx & tx
- disable DPLL_SEL when disabling the crtc
- add a few extra delays
Most of this concerns the gen5+ crtc disable function, and thanks
to a bunch of testers this is known to still work fine on gen 5/6/7.
ok kettenis@ matthieu@
|
|
ok matthieu@
|
|
chip generations. The driver is currently unable to restore text mode on
these chips, and attempting to restore text mode anyway results in the
display not being restored properly when switching back to X. The only way
of restoring the display after restarting X was to reboot the machine.
ok mpi@
|
|
doesn't actually seem to require any newer interfaces.
ok jsg@
|
|
|
|
ok jsg@
|