import xorg-x11-drv-intel-2.99.917-38.20180618.el8

This commit is contained in:
CentOS Sources 2019-08-02 12:22:04 -04:00 committed by Stepan Oksanichenko
commit f1ca0dd029
10 changed files with 809 additions and 0 deletions

2
.gitignore vendored Normal file
View File

@ -0,0 +1,2 @@
SOURCES/intel-gpu-tools-1.9.tar.bz2
SOURCES/xf86-video-intel-20180618.tar.bz2

View File

@ -0,0 +1,2 @@
396e9f50283681d50364b86a7af9988c56148ac8 SOURCES/intel-gpu-tools-1.9.tar.bz2
7e044e8985f88104f40cc9a96bce001edff8a184 SOURCES/xf86-video-intel-20180618.tar.bz2

View File

@ -0,0 +1,101 @@
From 9c1151b4d65c356f0d25d952fe1a10c89fdb834a Mon Sep 17 00:00:00 2001
From: Adam Jackson <ajax@redhat.com>
Date: Mon, 5 Mar 2018 11:03:18 -0500
Subject: [PATCH] Fix build on F28 and later
Newer gcc complains that it doesn't know how to always-inline memcpy():
/usr/include/bits/string_fortified.h:31:1: error: inlining failed in
call to always_inline 'memcpy': target specific option mismatch
This is because we need to wrap our declaration of memcpy() in the
options we're trying to push at the top of blt.c. So: include compiler.h
to define sse2, then push options, then include everything else.
However if you do that, the word 'nonnull' collides with the usage in
<X11/Xfuncproto.h>. I'm too lazy to fix that properly, just expand it to
the __attribute__ form in the few places we say it.
Signed-off-by: Adam Jackson <ajax@redhat.com>
---
src/sna/blt.c | 6 ++++--
src/sna/compiler.h | 2 --
src/sna/gen6_common.h | 6 +++---
3 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/src/sna/blt.c b/src/sna/blt.c
index cb90437a..fb357d35 100644
--- a/src/sna/blt.c
+++ b/src/sna/blt.c
@@ -29,13 +29,15 @@
#include "config.h"
#endif
-#include "sna.h"
-#include <pixman.h>
+#include "compiler.h"
#if defined(sse2)
#pragma GCC push_options
#pragma GCC target("sse2,inline-all-stringops,fpmath=sse")
#pragma GCC optimize("Ofast")
+
+#include <pixman.h>
+#include "sna.h"
#include <xmmintrin.h>
#if __x86_64__
diff --git a/src/sna/compiler.h b/src/sna/compiler.h
index 0f3775ec..3c176a16 100644
--- a/src/sna/compiler.h
+++ b/src/sna/compiler.h
@@ -39,7 +39,6 @@
#define pure __attribute__((pure))
#define tightly_packed __attribute__((__packed__))
#define flatten __attribute__((flatten))
-#define nonnull __attribute__((nonnull))
#define page_aligned __attribute__((aligned(4096)))
#else
#define likely(expr) (expr)
@@ -52,7 +51,6 @@
#define pure
#define tighly_packed
#define flatten
-#define nonnull
#define page_aligned
#endif
diff --git a/src/sna/gen6_common.h b/src/sna/gen6_common.h
index b53ec0c9..119a2d5d 100644
--- a/src/sna/gen6_common.h
+++ b/src/sna/gen6_common.h
@@ -133,7 +133,7 @@ inline static bool force_blt_ring(struct sna *sna, struct kgem_bo *bo)
return false;
}
-nonnull inline static bool
+__attribute__((nonnull)) inline static bool
prefer_blt_ring(struct sna *sna, struct kgem_bo *bo, unsigned flags)
{
if (PREFER_RENDER)
@@ -148,7 +148,7 @@ prefer_blt_ring(struct sna *sna, struct kgem_bo *bo, unsigned flags)
return can_switch_to_blt(sna, bo, flags);
}
-nonnull inline static bool
+__attribute__((nonnull)) inline static bool
prefer_render_ring(struct sna *sna, struct kgem_bo *bo)
{
if (sna->kgem.ring == KGEM_RENDER)
@@ -191,7 +191,7 @@ prefer_blt_composite(struct sna *sna, struct sna_composite_op *tmp)
return prefer_blt_bo(sna, tmp->src.bo, tmp->dst.bo);
}
-nonnull static inline bool
+__attribute__((nonnull)) static inline bool
prefer_blt_fill(struct sna *sna, struct kgem_bo *bo, unsigned flags)
{
if (PREFER_RENDER)
--
2.16.2

View File

@ -0,0 +1,49 @@
From a414d4e24461da1cb4cef8ee910bc57bab360ceb Mon Sep 17 00:00:00 2001
From: Adam Jackson <ajax@redhat.com>
Date: Tue, 6 Mar 2018 12:07:46 -0500
Subject: [PATCH] Fix build on i686
Presumably this only matters for i686 because amd64 implies sse2, but:
BUILDSTDERR: In file included from gen4_vertex.c:34:
BUILDSTDERR: gen4_vertex.c: In function 'emit_vertex':
BUILDSTDERR: sna_render_inline.h:40:26: error: inlining failed in call to always_inline 'vertex_emit_2s': target specific option mismatch
BUILDSTDERR: static force_inline void vertex_emit_2s(struct sna *sna, int16_t x, int16_t y)
BUILDSTDERR: ^~~~~~~~~~~~~~
BUILDSTDERR: gen4_vertex.c:308:25: note: called from here
BUILDSTDERR: #define OUT_VERTEX(x,y) vertex_emit_2s(sna, x,y) /* XXX assert(!too_large(x, y)); */
BUILDSTDERR: ^~~~~~~~~~~~~~~~~~~~~~~~
BUILDSTDERR: gen4_vertex.c:360:2: note: in expansion of macro 'OUT_VERTEX'
BUILDSTDERR: OUT_VERTEX(dstX, dstY);
BUILDSTDERR: ^~~~~~~~~~
The bug here appears to be that emit_vertex() is declared 'sse2' but
vertex_emit_2s is merely always_inline. gcc8 decides that since you said
always_inline you need to have explicitly cloned it for every
permutation of targets. Merely saying inline seems to do the job of
cloning vertex_emit_2s as much as necessary.
So to reiterate: if you say always-inline, it won't, but if you just say
maybe inline, it will. Thanks gcc, that's helpful.
- ajax
---
src/sna/compiler.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/sna/compiler.h b/src/sna/compiler.h
index 3c176a16..bc447c7a 100644
--- a/src/sna/compiler.h
+++ b/src/sna/compiler.h
@@ -32,7 +32,7 @@
#define likely(expr) (__builtin_expect (!!(expr), 1))
#define unlikely(expr) (__builtin_expect (!!(expr), 0))
#define noinline __attribute__((noinline))
-#define force_inline inline __attribute__((always_inline))
+#define force_inline inline
#define fastcall __attribute__((regparm(3)))
#define must_check __attribute__((warn_unused_result))
#define constant __attribute__((const))
--
2.16.2

View File

@ -0,0 +1,47 @@
From 09bf64d69d97172f22fe08eb495bf50f920ae9ec Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 31 May 2016 11:11:21 +0300
Subject: [PATCH] sna: Avoid clobbering output physical size with
xf86OutputSetEDID()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
---
src/sna/sna_display.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/src/sna/sna_display.c b/src/sna/sna_display.c
index 10d33f2..57de410 100644
--- a/src/sna/sna_display.c
+++ b/src/sna/sna_display.c
@@ -3576,6 +3576,7 @@ sna_output_attach_edid(xf86OutputPtr output)
{
struct sna *sna = to_sna(output->scrn);
struct sna_output *sna_output = output->driver_private;
+ unsigned old_mm_width, old_mm_height;
struct drm_mode_get_blob blob;
void *old, *raw = NULL;
xf86MonPtr mon = NULL;
@@ -3678,12 +3679,17 @@ skip_read:
}
done:
+ output = sna_output->base;
+ old_mm_width = output->mm_width;
+ old_mm_height = output->mm_height;
sna_output_set_parsed_edid(output, mon);
if (raw) {
sna_output->edid_raw = raw;
sna_output->edid_len = blob.length;
sna_output->edid_blob_id = blob.blob_id;
}
+ output->mm_width = old_mm_width;
+ output->mm_height = old_mm_height;
}
static void
--
2.7.4

11
SOURCES/igt-stat.patch Normal file
View File

@ -0,0 +1,11 @@
diff -up intel-gpu-tools-1.9/tests/testdisplay_hotplug.c.jx intel-gpu-tools-1.9/tests/testdisplay_hotplug.c
--- intel-gpu-tools-1.9/tests/testdisplay_hotplug.c.jx 2014-12-12 11:08:14.000000000 -0500
+++ intel-gpu-tools-1.9/tests/testdisplay_hotplug.c 2015-06-03 15:43:55.127775543 -0400
@@ -24,6 +24,7 @@
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
+#include <sys/stat.h>
#include "testdisplay.h"
#ifdef HAVE_CONFIG_H

View File

@ -0,0 +1,12 @@
diff -up xf86-video-intel-20150520/src/sna/compiler.h.jx xf86-video-intel-20150520/src/sna/compiler.h
--- xf86-video-intel-20150520/src/sna/compiler.h.jx 2015-05-19 07:44:26.000000000 -0400
+++ xf86-video-intel-20150520/src/sna/compiler.h 2015-05-20 12:53:54.797918726 -0400
@@ -64,7 +64,7 @@
#endif
#if HAS_GCC(4, 6) && defined(__OPTIMIZE__)
-#define fast __attribute__((optimize("Ofast")))
+#define fast __attribute__((optimize("O3")))
#else
#define fast
#endif

22
SOURCES/make-git-snapshot.sh Executable file
View File

@ -0,0 +1,22 @@
#!/bin/sh
# Usage: ./make-git-snapshot.sh [COMMIT]
#
# to make a snapshot of the given tag/branch. Defaults to HEAD.
# Point env var REF to a local mesa repo to reduce clone time.
DIRNAME=xf86-video-intel-$( date +%Y%m%d )
echo REF ${REF:+--reference $REF}
echo DIRNAME $DIRNAME
echo HEAD ${1:-HEAD}
rm -rf $DIRNAME
git clone ${REF:+--reference $REF} \
git://git.freedesktop.org/git/xorg/driver/xf86-video-intel $DIRNAME
GIT_DIR=$DIRNAME/.git git archive --format=tar --prefix=$DIRNAME/ ${1:-HEAD} \
| bzip2 > $DIRNAME.tar.bz2
# rm -rf $DIRNAME

View File

@ -0,0 +1,22 @@
#!/bin/sh
# Usage: ./make-git-snapshot.sh [COMMIT]
#
# to make a snapshot of the given tag/branch. Defaults to HEAD.
# Point env var REF to a local mesa repo to reduce clone time.
DIRNAME=intel-gpu-tools-$( date +%Y%m%d )
echo REF ${REF:+--reference $REF}
echo DIRNAME $DIRNAME
echo HEAD ${1:-HEAD}
rm -rf $DIRNAME
git clone --depth 1 ${REF:+--reference $REF} \
git://git.freedesktop.org/git/xorg/app/intel-gpu-tools $DIRNAME
GIT_DIR=$DIRNAME/.git git archive --format=tar --prefix=$DIRNAME/ ${1:-HEAD} \
| bzip2 > $DIRNAME.tar.bz2
# rm -rf $DIRNAME

View File

@ -0,0 +1,541 @@
%define moduledir %(pkg-config xorg-server --variable=moduledir )
%define driverdir %{moduledir}/drivers
%define gputoolsver 1.9
%define gitdate 20180618
%define gitrev .%{gitdate}
%undefine _hardened_build
Summary: Xorg X11 Intel video driver
Name: xorg-x11-drv-intel
Version: 2.99.917
Release: 38%{?gitrev}%{?dist}
URL: http://www.x.org
License: MIT
Group: User Interface/X Hardware Support
%if 0%{?gitdate}
Source0: xf86-video-intel-%{gitdate}.tar.bz2
%else
Source0: https://xorg.freedesktop.org/archive/individual/driver/xf86-video-intel-%{version}.tar.bz2
%endif
Source1: make-intel-gpu-tools-snapshot.sh
Source3: https://xorg.freedesktop.org/archive/individual/app/intel-gpu-tools-%{gputoolsver}.tar.bz2
Source4: make-git-snapshot.sh
Patch0: intel-gcc-pr65873.patch
Patch1: igt-stat.patch
# https://bugs.freedesktop.org/show_bug.cgi?id=96255#c11
Patch4: 0001-sna-Avoid-clobbering-output-physical-size-with-xf86O.patch
Patch5: 0001-Fix-build-on-F28-and-later.patch
Patch6: 0001-Fix-build-on-i686.patch
ExclusiveArch: %{ix86} x86_64 ia64
BuildRequires: autoconf automake libtool
BuildRequires: flex bison
BuildRequires: xorg-x11-server-devel >= 1.10.99.902
BuildRequires: libX11-devel
BuildRequires: libXcursor-devel
BuildRequires: libXdamage-devel
BuildRequires: libXext-devel
BuildRequires: libXfixes-devel
BuildRequires: libXinerama-devel
BuildRequires: libXrandr-devel
BuildRequires: libXrender-devel
BuildRequires: libXtst-devel
BuildRequires: libXvMC-devel
BuildRequires: libXfont2-devel
BuildRequires: mesa-libGL-devel >= 6.5-9
BuildRequires: libdrm-devel >= 2.4.25
BuildRequires: kernel-headers >= 2.6.32.3
BuildRequires: libudev-devel
BuildRequires: libxcb-devel >= 1.5
BuildRequires: xcb-util-devel
BuildRequires: cairo-devel
BuildRequires: python3
BuildRequires: libXScrnSaver-devel
BuildRequires: libXext-devel
BuildRequires: pixman-devel
Requires: Xorg %(xserver-sdk-abi-requires ansic)
Requires: Xorg %(xserver-sdk-abi-requires videodrv)
Requires: polkit
%description
X.Org X11 Intel video driver.
%package devel
Summary: Xorg X11 Intel video driver development package
Group: Development/System
Requires: %{name} = %{version}-%{release}
Provides: xorg-x11-drv-intel-devel = %{version}-%{release}
%description devel
X.Org X11 Intel video driver development package.
%package -n intel-gpu-tools
Summary: Debugging tools for Intel graphics chips
Group: Development/Tools
%description -n intel-gpu-tools
Debugging tools for Intel graphics chips
%if 0%{?gitdate}
%define dirsuffix %{gitdate}
%else
%define dirsuffix %{version}
%endif
%prep
%setup -q -n xf86-video-intel-%{?gitdate:%{gitdate}}%{!?gitdate:%{dirsuffix}} -b3
%patch0 -p1 -b .gcc
%patch4 -p1
%patch5 -p1
%patch6 -p1
pushd ../intel-gpu-tools-%{gputoolsver}
%patch1 -p1 -b .stat
popd
%build
autoreconf -f -i -v
%configure --enable-kms-only --with-default-dri=3 --enable-tools
make %{?_smp_mflags} V=1
pushd ../intel-gpu-tools-%{gputoolsver}
touch lib/check-ndebug.h
mkdir -p m4
autoreconf -f -i -v
# --disable-dumper: quick_dump is both not recommended for packaging yet,
# and requires python3 to build; i'd like to keep this spec valid for rhel6
# for at least a bit longer
%configure --disable-dumper
# some of the sources are in utf-8 and pre-preprocessed by python
export LANG=en_US.UTF-8
make %{?_smp_mflags}
popd
%install
%make_install
pushd ../intel-gpu-tools-%{gputoolsver}
make install DESTDIR=$RPM_BUILD_ROOT
rm -f $RPM_BUILD_ROOT%{_bindir}/eudb
popd
find $RPM_BUILD_ROOT -regex ".*\.la$" | xargs rm -f --
# libXvMC opens the versioned file name, these are useless
rm -f $RPM_BUILD_ROOT%{_libdir}/libI*XvMC.so
%ldconfig_post
%ldconfig_postun
%files
%doc COPYING
%{driverdir}/intel_drv.so
%{_libdir}/libIntelXvMC.so.1*
%{_libexecdir}/xf86-video-intel-backlight-helper
%{_datadir}/polkit-1/actions/org.x.xf86-video-intel.backlight-helper.policy
%{_mandir}/man4/i*
%files devel
%{_bindir}/intel-gen4asm
%{_bindir}/intel-gen4disasm
%{_libdir}/pkgconfig/intel-gen4asm.pc
%files -n intel-gpu-tools
%doc COPYING
%{_bindir}/gem_userptr_benchmark
%{_bindir}/intel*
%exclude %{_bindir}/intel-gen4asm
%exclude %{_bindir}/intel-gen4disasm
%{_datadir}/gtk-doc
%{_mandir}/man1/intel_*.1*
%changelog
* Fri Jun 29 2018 Adam Jackson <ajax@redhat.com> - 2.99.917-38.20180618
- Use ldconfig scriptlet macros
* Mon Jun 18 2018 Adam Jackson <ajax@redhat.com> - 2.99.917-37
- Today's git snapshot (commit 3d395062)
- Clean up some conditionals
- Build with python3
* Wed May 30 2018 Adam Jackson <ajax@redhat.com> - 2.99.917-37
* Mon Apr 02 2018 Adam Jackson <ajax@redhat.com> - 2.99.917-36.20171025
- Rebuild for xserver 1.20
* Sun Mar 18 2018 Iryna Shcherbina <ishcherb@redhat.com> - 2.99.917-35.20171025
- Update Python 2 dependency declarations to new packaging standards
(See https://fedoraproject.org/wiki/FinalizingFedoraSwitchtoPython3)
* Mon Mar 05 2018 Adam Jackson <ajax@redhat.com> - 2.99.917-34
- Fix build on F28+
* Fri Feb 09 2018 Fedora Release Engineering <releng@fedoraproject.org> - 2.99.917-33.20171025
- Rebuilt for https://fedoraproject.org/wiki/Fedora_28_Mass_Rebuild
* Mon Nov 06 2017 Merlin Mathesius <mmathesi@redhat.com> - 2.99.917-32.20171025
- Cleanup spec file conditionals
* Wed Oct 25 2017 Adam Jackson <ajax@redhat.com> - 2.99.917-31.20171025
- New git snapshot
* Thu Aug 03 2017 Fedora Release Engineering <releng@fedoraproject.org> - 2.99.917-30.20160929
- Rebuilt for https://fedoraproject.org/wiki/Fedora_27_Binutils_Mass_Rebuild
* Thu Jul 27 2017 Fedora Release Engineering <releng@fedoraproject.org> - 2.99.917-29.20160929
- Rebuilt for https://fedoraproject.org/wiki/Fedora_27_Mass_Rebuild
* Sat Feb 11 2017 Fedora Release Engineering <releng@fedoraproject.org> - 2.99.917-28.20160929
- Rebuilt for https://fedoraproject.org/wiki/Fedora_26_Mass_Rebuild
* Tue Jan 10 2017 Hans de Goede <hdegoede@redhat.com> - 2.99.917-27
- Drop the patch to fail probe on skylake and newer, the xserver now
defaults to the modesetting driver for all recent GPUs
* Thu Sep 29 2016 Hans de Goede <hdegoede@redhat.com> - 2.99.917-26
- Update to latest git master for use with xserver-1.19
- Rebuild against xserver-1.19
* Wed Jul 20 2016 Bastien Nocera <bnocera@redhat.com> - 2.99.917-25
- Avoid clobbering output physical size, should fix hidpi on Surface 3
* Tue Jul 12 2016 Hans de Goede <hdegoede@redhat.com> - 2.99.917-24.20160712
- Git snapshot du jour, bringing in a bunch of bugfixes
- Fix dpi issues on Surface 3 (also needs some kernel fixes)
- Hopefully fix rhbz#1354124
* Thu May 12 2016 Hans de Goede <hdegoede@redhat.com> - 2.99.917-23.20160512
- Update to 20160512 snapshot
- This fixes laptops with switchable graphics hanging after a dpms off of
the lcd screen (rhbz#1334581)
- Fix fd-leak when falling back to mode-setting on skylake, this fixes
Xorg exiting with a "drmSetMaster failed" error when not using server
managed fds (e.g. using a different login manager then gdm)
- Fix duplicate binaries in -devel / intel-gpu-tools subpackages (rhbz#1323641)
* Mon Mar 07 2016 Hans de Goede <hdegoede@redhat.com> - 2.99.917-22.20160119
- xorg-x11-drv-intel hardly has any accel on skylake and newer, so make
Xorg fallback to modesetting + glamor by returning FALSE from probe
- Using glamor also gives us proper Xvideo support on skylake (rhbz#1305369)
* Fri Feb 05 2016 Fedora Release Engineering <releng@fedoraproject.org> - 2.99.917-21.20160119
- Rebuilt for https://fedoraproject.org/wiki/Fedora_24_Mass_Rebuild
* Tue Jan 19 2016 Kevin Fenzi <kevin@scrye.com> - 2.99.917-20-20160119
- Update to 20160119 snapshot
* Sun Dec 06 2015 Adel Gadllah <adel.gadllah@gmail.com> - 2.99.917-19.20151109
- Update to 20151206 snapshot
* Tue Nov 17 2015 Adel Gadllah <adel.gadllah@gmail.com> - 2.99.917-18.20151109
- Reenable DRI3 - we ship xserver 1.18 now
* Mon Nov 09 2015 Kevin Fenzi <kevin@scrye.com> - 2.99.917-17.20151109
- Update to 20151109 snapshot
* Wed Sep 16 2015 Dave Airlie <airlied@redhat.com> - 2.99.917-16.20150729
- 1.18 ABI rebuild
* Wed Jul 29 2015 Dave Airlie <airlied@redhat.com> 2.99.917-15.20150729
- update to upstream git snapshot for ABI
* Wed Jul 29 2015 Dave Airlie <airlied@redhat.com> - 2.99.917-14.20150615
- 1.15 ABI rebuild
* Fri Jun 19 2015 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 2.99.917-13.20150615
- Rebuilt for https://fedoraproject.org/wiki/Fedora_23_Mass_Rebuild
* Mon Jun 15 2015 Dave Airlie <airlied@redhat.com> 2.99.917-12
- you guessed it, git snap of the day
- this should bring some PRIME corruption fixes.
* Wed Jun 03 2015 Adam Jackson <ajax@redhat.com> 2.99.917-11
- Today's git snap
* Tue May 26 2015 Dave Airlie <airlied@redhat.com> 2.99.917-10
- update git snap
- fixes uninitialised properties crash
* Wed May 20 2015 Adam Jackson <ajax@redhat.com> 2.99.917-9
- Today's git snap
- Don't force the default to DRI3, use upstream's preference
- Fix build failure due to GCC PR65873
* Mon Mar 02 2015 Dave Airlie <airlied@redhat.com> 2.99.917-8
- this time for sure, now less hardended.
* Mon Mar 02 2015 Dave Airlie <airlied@redhat.com> 2.99.917-7
- remove cement, X.org drivers aren't hard enough.
* Thu Feb 26 2015 Hans de Goede <hdegoede@redhat.com> - 2.99.917-6
- Really really build intel-virtual-output (rhbz#1195962)
* Thu Feb 26 2015 Hans de Goede <hdegoede@redhat.com> - 2.99.917-5
- Add more missing BuildRequires so that intel-virtual-output really gets
build (rhbz#1195962)
* Thu Feb 26 2015 Hans de Goede <hdegoede@redhat.com> - 2.99.917-4
- Add missing BuildRequires libXext-devel so that intel-virtual-output gets
build (rhbz#1195962)
* Wed Feb 11 2015 Hans de Goede <hdegoede@redhat.com> - 2.99.917-3
- Git snapshot of the day to bring in various DPMS and other fixes
- Update gpu-tools to 1.9
- Xserver 1.17 ABI rebuild
* Wed Dec 24 2014 Adel Gadllah <adel.gadllah@gmail.com> 2.99.917-2
- Enable DRI3
* Mon Dec 22 2014 Kevin Fenzi <kevin@scrye.com> 2.99.917-1
- Update to 2.99.917
* Mon Nov 17 2014 Adam Jackson <ajax@redhat.com> 2.99.916-3
- Today's git snapshot
* Thu Sep 11 2014 Dave Airlie <airlied@redhat.com> 2.99.916-2
- backport some SNA and MST fixes.
* Wed Sep 10 2014 Dave Airlie <airlied@redhat.com> 2.99.916-1
- Rebase to 2.99.916
* Wed Sep 03 2014 Dave Airlie <airlied@redhat.com> 2.99.914-4
- Add UXA MST support as a fallback
* Tue Sep 02 2014 Adel Gadllah <adel.gadllah@gmail.com> - 2.99.914-3
- Backport fix for sna to fix broken shadow rendering in gtk
* Mon Aug 18 2014 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 2.99.914-2
- Rebuilt for https://fedoraproject.org/wiki/Fedora_21_22_Mass_Rebuild
* Mon Jul 28 2014 Hans de Goede <hdegoede@redhat.com> - 2.99.914-1
- Rebase to 2.99.914
* Tue Jul 22 2014 Adel Gadllah <adel.gadllah@gmail.com> - 2.99.912-6
- Apply fix for sna render corruption due to missing fencing, FDO #81551
* Fri Jul 11 2014 Hans de Goede <hdegoede@redhat.com> - 2.99.912-5
- Fix a security issue in the backlight helper (CVE-2014-4910)
* Tue Jul 1 2014 Hans de Goede <hdegoede@redhat.com> - 2.99.912-4
- Re-enable DRI3 support (the latest mesa fixes the gnome-shell hang)
* Wed Jun 18 2014 Hans de Goede <hdegoede@redhat.com> - 2.99.912-3
- xserver 1.15.99.903 ABI rebuild
* Thu Jun 12 2014 Hans de Goede <hdegoede@redhat.com> - 2.99.912-2
- DRI3 support causes gnome-shell to hang, disable for now
* Wed Jun 11 2014 Hans de Goede <hdegoede@redhat.com> - 2.99.912-1
- Rebase to 2.99.912
- Rebuild for xserver 1.15.99.903
- Update intel-gpu-tools to 1.7 release
* Sun Jun 08 2014 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 2.99.911-3
- Rebuilt for https://fedoraproject.org/wiki/Fedora_21_Mass_Rebuild
* Mon Apr 28 2014 Hans de Goede <hdegoede@redhat.com> - 2.99.911-2
- xserver 1.15.99-20140428 git snapshot ABI rebuild
- Add 2 patches from upstream to not close server-fds of udl devices
* Thu Apr 17 2014 Hans de Goede <hdegoede@redhat.com> - 2.99.911-1
- Rebase to 2.99.911
- Rebuild for xserver 1.15.99.902
* Mon Jan 13 2014 Adam Jackson <ajax@redhat.com> - 2.21.15-13
- 1.15 ABI rebuild
* Sat Dec 21 2013 Ville Skyttä <ville.skytta@iki.fi> - 2.21.15-12
- Call ldconfig in %%post* scriptlets.
* Tue Dec 17 2013 Adam Jackson <ajax@redhat.com> - 2.21.15-11
- 1.15RC4 ABI rebuild
* Wed Nov 20 2013 Adam Jackson <ajax@redhat.com> - 2.21.15-10
- 1.15RC2 ABI rebuild
* Wed Nov 06 2013 Adam Jackson <ajax@redhat.com> - 2.21.15-9
- 1.15RC1 ABI rebuild
* Mon Oct 28 2013 Adam Jackson <ajax@redhat.com> - 2.21.15-8
- Don't patch in xwayland in RHEL
* Fri Oct 25 2013 Adam Jackson <ajax@redhat.com> - 2.21.15-7
- ABI rebuild
* Thu Oct 24 2013 Adam Jackson <ajax@redhat.com> 2.21.15-6
- Disable UMS support in F21+
* Thu Oct 24 2013 Adam Jackson <ajax@redhat.com> 2.21.15-5
- xserver 1.15 API compat
* Wed Oct 02 2013 Adam Jackson <ajax@redhat.com> 2.21.15-4
- Default to uxa again
* Mon Sep 23 2013 Adam Jackson <ajax@redhat.com> 2.21.15-2
- Change xwayland requires to be explicitly versioned
* Mon Sep 23 2013 Adam Jackson <ajax@redhat.com> 2.21.15-1
- intel 2.21.15
- xwayland support
* Tue Aug 06 2013 Dave Airlie <airlied@redhat.com> 2.21.14-1
- intel 2.21.24
- add fix to make build - re-enable autoreconf
* Sun Aug 04 2013 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 2.21.12-2
- Rebuilt for https://fedoraproject.org/wiki/Fedora_20_Mass_Rebuild
* Tue Jul 23 2013 Adam Jackson <ajax@redhat.com> 2.21.12-1
- intel 2.21.12
* Tue Jun 11 2013 Adam Jackson <ajax@redhat.com> 2.21.9-1
- intel 2.21.9
- New i-g-t snapshot
- Drop useless symlinks from -devel
- Repurpose -devel for intel-gen4{,dis}asm
- Default to SNA in F20+
* Tue May 28 2013 Adam Jackson <ajax@redhat.com> 2.21.8-1
- intel 2.21.8
* Fri Apr 12 2013 Dave Airlie <airlied@redhat.com> 2.21.6-1
- intel 2.21.6
* Thu Mar 21 2013 Adam Jackson <ajax@redhat.com> 2.21.5-1
- intel 2.21.5
* Mon Mar 11 2013 Adam Jackson <ajax@redhat.com> 2.21.4-1
- intel 2.21.4
* Thu Mar 07 2013 Adam Jackson <ajax@redhat.com> 2.21.3-1
- intel 2.21.3
* Thu Mar 07 2013 Peter Hutterer <peter.hutterer@redhat.com> - 2.21.2-4
- ABI rebuild
* Fri Feb 15 2013 Peter Hutterer <peter.hutterer@redhat.com> - 2.21.2-3
- ABI rebuild
* Fri Feb 15 2013 Peter Hutterer <peter.hutterer@redhat.com> - 2.21.2-2
- ABI rebuild
* Tue Feb 12 2013 Adam Jackson <ajax@redhat.com> 2.21.2-1
- intel 2.21.2
- New i-g-t snapshot
- Pre-F16 changelog trim
* Wed Jan 16 2013 Adam Jackson <ajax@redhat.com> 2.20.18-2
- Compensate for rawhide's aclocal breaking in a newly stupid way
* Wed Jan 16 2013 Adam Jackson <ajax@redhat.com> 2.20.18-1
- intel 2.20.18
* Tue Jan 08 2013 Dave Airlie <airlied@redhat.com> 2.20.17-2
- Fix damage issue for reverse prime work
* Fri Jan 04 2013 Adam Jackson <ajax@redhat.com> 2.20.17-1
- intel 2.20.17
* Wed Jan 02 2013 Dave Airlie <airlied@redhat.com> 2.20.16-2
- Fix uxa bug that trips up ilk on 3.7 kernels
* Mon Dec 17 2012 Adam Jackson <ajax@redhat.com> 2.20.16-1
- intel 2.20.16
* Wed Nov 28 2012 Adam Jackson <ajax@redhat.com> 2.20.14-1
- intel 2.20.14
* Mon Oct 22 2012 Adam Jackson <ajax@redhat.com> 2.20.12-1
- intel 2.20.12
* Fri Oct 19 2012 Adam Jackson <ajax@redhat.com> 2.20.10-2
- Today's i-g-t
- Don't bother building the nouveau bits of i-g-t on OSes without an X
server with prime support.
* Mon Oct 15 2012 Dave Airlie <airlied@redhat.com> 2.20.10-1
- intel 2.20.10
* Fri Oct 05 2012 Adam Jackson <ajax@redhat.com> 2.20.9-1
- intel 2.20.9
- Today's intel-gpu-tools snapshot
* Fri Sep 21 2012 Adam Jackson <ajax@redhat.com> 2.20.8-1
- intel 2.20.8
* Mon Sep 10 2012 Adam Jackson <ajax@redhat.com> 2.20.7-1
- intel 2.20.7
* Fri Sep 07 2012 Dave Airlie <airlied@redhat.com> 2.20.6-2
- latest upstream git snapshot with prime + fixes
* Tue Sep 04 2012 Adam Jackson <ajax@redhat.com> 2.20.6-2
- Only bother to build UMS (read: i810) support on 32-bit. If you've
managed to build a machine with an i810 GPU but a 64-bit CPU, please
don't have done that.
* Tue Sep 04 2012 Adam Jackson <ajax@redhat.com> 2.20.6-1
- intel 2.20.6 (#853783)
* Thu Aug 30 2012 Adam Jackson <ajax@redhat.com> 2.20.5-2
- Don't package I810XvMC when not building legacy i810
* Mon Aug 27 2012 Adam Jackson <ajax@redhat.com> 2.20.5-1
- intel 2.20.5
* Mon Aug 20 2012 Adam Jackson <ajax@redhat.com> 2.20.4-3
- Rebuild for new xcb-util soname
* Mon Aug 20 2012 Adam Jackson <ajax@redhat.com> 2.20.4-2
- Backport some patches to avoid binding to non-i915.ko-driven Intel GPUs,
like Cedarview and friends (#849475)
* Mon Aug 20 2012 Adam Jackson <ajax@redhat.com> 2.20.4-1
- intel 2.20.4
* Thu Aug 16 2012 Dave Airlie <airlied@redhat.com> 2.20.3-3
- fix vmap flush to correct upstream version in prime patch
* Thu Aug 16 2012 Dave Airlie <airlied@redhat.com> 2.20.3-2
- snapshot upstream + add prime support for now
* Wed Aug 15 2012 Adam Jackson <ajax@redhat.com> 2.20.3-1
- intel 2.20.3
* Wed Aug 01 2012 Adam Jackson <ajax@redhat.com> 2.20.2-1
- intel 2.20.2
- Only disable UMS in RHEL7, since i810 exists in RHEL6
* Mon Jul 23 2012 Adam Jackson <ajax@redhat.com> 2.20.1-1
- intel 2.20.1
* Sun Jul 22 2012 Fedora Release Engineering <rel-eng@lists.fedoraproject.org> - 2.20.0-2.20120718
- Rebuilt for https://fedoraproject.org/wiki/Fedora_18_Mass_Rebuild
* Wed Jul 18 2012 Dave Airlie <airlied@redhat.com> 2.20.0-1.20120718
- todays git snapshot
* Tue Jun 12 2012 Dave Airlie <airlied@redhat.com> 2.19.0-5.20120612
- today's git snapshot
- resurrect copy-fb
* Tue May 29 2012 Adam Jackson <ajax@redhat.com> 2.19.0-4.20120529
- Today's git snapshot
- Enable SNA (default is still UXA, use Option "AccelMethod" to switch)
- build-fix.patch: Fix build with Fedora's default cflags
* Tue May 29 2012 Adam Jackson <ajax@redhat.com> 2.19.0-3
- Don't autoreconf the driver, fixes build on F16.
* Mon May 21 2012 Adam Jackson <ajax@redhat.com> 2.19.0-2
- Disable UMS support in RHEL.
- Trim some Requires that haven't been needed since F15.
* Thu May 03 2012 Adam Jackson <ajax@redhat.com> 2.19.0-1
- intel 2.19.0