Gwenole Beauchesne [Fri, 4 Nov 2011 16:33:48 +0000 (17:33 +0100)]
doc: add placeholder for the Core API.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Austin Yuan [Wed, 16 Nov 2011 05:59:58 +0000 (00:59 -0500)]
va_tpi.h: corret a typo (Kbuf->KBuf)
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
Gwenole Beauchesne [Thu, 3 Nov 2011 09:41:59 +0000 (10:41 +0100)]
configure: add check for EGL.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Thu, 3 Nov 2011 09:28:00 +0000 (10:28 +0100)]
configure: simplify summary.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Thu, 3 Nov 2011 09:36:42 +0000 (10:36 +0100)]
configure: cosmetics (clean up configure options).
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Thu, 3 Nov 2011 09:16:13 +0000 (10:16 +0100)]
configure: cosmetics (sort AC_OUTPUT Makefiles).
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Wed, 2 Nov 2011 16:08:39 +0000 (17:08 +0100)]
doc: add Doxygen templates.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Wed, 2 Nov 2011 14:28:53 +0000 (15:28 +0100)]
libva: bump version for development.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Fri, 28 Oct 2011 13:40:28 +0000 (15:40 +0200)]
1.0.15.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Fri, 28 Oct 2011 13:34:26 +0000 (15:34 +0200)]
debian: add package for vainfo.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Fri, 28 Oct 2011 13:20:08 +0000 (15:20 +0200)]
debian: don't use simple-patchsys (upstream has no patches).
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Wed, 26 Oct 2011 11:45:50 +0000 (13:45 +0200)]
Add NEWS file.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Thu, 20 Oct 2011 12:05:38 +0000 (14:05 +0200)]
va: fix compiler warnings (drop unused variables).
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Fri, 14 Oct 2011 16:19:03 +0000 (18:19 +0200)]
Add auto-generated Debian packaging.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Tue, 18 Oct 2011 14:06:19 +0000 (16:06 +0200)]
libva: bump version for development.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Tue, 18 Oct 2011 14:05:50 +0000 (16:05 +0200)]
libva: allow pre-releases.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Sun, 16 Oct 2011 10:09:02 +0000 (12:09 +0200)]
libva: fix reported version as the one from VA-API.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Sun, 16 Oct 2011 10:09:10 +0000 (12:09 +0200)]
vainfo: report both VA-API and libva versions.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Sun, 16 Oct 2011 10:08:52 +0000 (12:08 +0200)]
API: fix versioning.
We have to consider three distinct versions at this time: the VA-API
version, the library package version number, and the DSO version.
* VA-API version:
- increment major for any ABI change (which shall not occur!)
- increment minor for any interface change (e.g. new or modified function)
- increment micro for any other change (e.g. new flag, new codec definitions)
- reset micro version to zero when minor version is incremented
- reset minor version to zero when major version is incremented
* libva package version number:
- major version is automatically generated from VA-API major version
- minor version is automatically generated from VA-API minor version
- increment micro for any library release
- reset micro version to zero when VA-API major or minor version is incremented
* DSO version:
The SONAME shall remain to libva.so.1 for VA-API 0.x.y as long as the ABI
is not changed. Thus, the library name is generated as libva.<x>.<y>.0 where
<x> = VA-API major version + 1
<y> = 100 * VA-API minor version + VA-API micro version
For example:
VA-API 0.32.0 generates libva.so.1.3200.0 (libva 1.0.14 as of today)
VA-API 0.34.1 generates libva.so.1.3401.0 (e.g. libva 1.2.1)
VA-API 1.2.13 generates libva.so.2.213.0 (e.g. libva 2.2.13)
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Austin Yuan [Thu, 20 Oct 2011 05:43:12 +0000 (01:43 -0400)]
test: fix build warnings on Android
Change-Id: Ibc4231395af8dfcdef403d44b98c644349e2ba04
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Gwenole Beauchesne [Tue, 18 Oct 2011 13:00:52 +0000 (15:00 +0200)]
android: fix <va/va_version.h> dependencies
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Tue, 18 Oct 2011 12:05:48 +0000 (14:05 +0200)]
android: fix generation of <va/va_version.h>
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Tue, 18 Oct 2011 12:25:52 +0000 (14:25 +0200)]
va: generate __vaDriverInit_*() function name at run-time.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Tue, 18 Oct 2011 12:04:02 +0000 (14:04 +0200)]
va: don't explicitly include "config.h"
sysdeps.h is the internal header file to include that will eventually
include config.h, if necessary. e.g. we don't use it on Android.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Tue, 18 Oct 2011 12:03:10 +0000 (14:03 +0200)]
build: drop generated files.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Mon, 17 Oct 2011 13:31:10 +0000 (15:31 +0200)]
vainfo: use system or user-defined DISPLAY to create the X display.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Mon, 17 Oct 2011 12:00:44 +0000 (14:00 +0200)]
dummy_drv_video: fix last commit.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Gwenole Beauchesne [Fri, 14 Oct 2011 16:13:54 +0000 (18:13 +0200)]
dummy_drv_video: only build if requested.
Gwenole Beauchesne [Fri, 14 Oct 2011 15:31:24 +0000 (17:31 +0200)]
pkgconfig: move files to a specific directory.
Gwenole Beauchesne [Fri, 14 Oct 2011 15:11:56 +0000 (17:11 +0200)]
configure: drop check for intel-gen4asm (moved to intel-driver).
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Austin Yuan [Mon, 17 Oct 2011 02:29:03 +0000 (22:29 -0400)]
va_tpi: correct the first argument of vaCreateSurfaceFromKBuf
"VADisplay dpy" --> "VADriverContextP ctx"
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Gwenole Beauchesne [Tue, 11 Oct 2011 08:34:37 +0000 (10:34 +0200)]
va_fglrx: fix driver name detection.
VA_FGLRXGetClientDriverName() used to return success even if no
valid driver name was found, thus returning a NULL driver name.
Also fixed the X display name matching routine to filter out the
screen number, which may not be in the string, and make sure to
compare screen numbers from what ADL returns too.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
Austin Yuan [Thu, 22 Sep 2011 03:29:45 +0000 (23:29 -0400)]
vatrace/vafool: refine it
remove va_fool_getframe.c which looks strange
correct some coding style for file va.c
simplify va_fool.c for both decode and encode
refine the surface dump of va_trace.c
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
jiguoliang [Thu, 25 Aug 2011 15:39:53 +0000 (11:39 -0400)]
va_tpi: vaCreateSurfaceFromKBuf for streaming the video playback to remote monitor
For such use case, the pipeline is: decode->encode->send to network.
The local decode/playback and the encode process are seperated, thus it
needs the buffer sharing b/w the decode and encode in seperate process.
This API creates/wraps the decoded surface into an encode surface. Its
input parameter is the kernel buffer handle of decoded surface which is
only understandable by driver implementation.
Change-Id: I7953ecf3b7333c1cdb6af5a0f45b054c8840f8e2
Signed-off-by: jiguoliang <guoliangx.ji@intel.com>
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
ywan171 [Tue, 2 Aug 2011 14:38:05 +0000 (22:38 +0800)]
va.h: TopFieldOrderCnt/BottomFieldOrderCnt should be signed instead of unsigned
POC should be integer.
Change-Id: I75281aafa148c88f0de58981969e1c43ea8d2684
Signed-off-by: ywan171 <yi.a.wang@intel.com>
Xiang, Haihao [Mon, 19 Sep 2011 02:02:39 +0000 (10:02 +0800)]
va: remove i965 driver from libva repository
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Mon, 19 Sep 2011 01:12:33 +0000 (09:12 +0800)]
va/test: Fix DSO link issue
I didn't see this issue until I upgraded my system recently.
It's evident the default behavior of the linker is changed.
Explicitly adding the missing libraries to the link step can resolve
this issue.
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Tue, 2 Aug 2011 07:06:09 +0000 (15:06 +0800)]
va: update the check for intel-gen4asm
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Thu, 28 Jul 2011 00:38:03 +0000 (08:38 +0800)]
bump to 1.0.14
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Gwenole Beauchesne [Tue, 26 Jul 2011 11:48:25 +0000 (19:48 +0800)]
i965_drv_video: don't export internal driver functions.
Make sure to use our internal functions. In particular, we override
some DRM functions and they have to be used. e.g. this fixes VA/GLX.
Austin Yuan [Tue, 26 Jul 2011 06:11:08 +0000 (14:11 +0800)]
va.h: VA_DECODE_ERROR_TYPE changed to VADecodeErrorType for unified codying style (CamelCase)
VA_DECODE_SLICE_MISSING --> VADecodeSliceMissing
VA_DECODE_MB_ERROR --> VADecodeMBError
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Gwenole Beauchesne [Wed, 20 Jul 2011 15:44:29 +0000 (17:44 +0200)]
dri2: ignore BadDrawable errors.
Hi,
This patch fixes the following error.
X Error of failed request: BadDrawable (invalid Pixmap or Window parameter)
Major opcode of failed request: 137 (DRI2)
Minor opcode of failed request: 4 (DRI2DestroyDrawable)
Resource id in failed request: 0x4800001
Serial number of failed request: 30
Current serial number in output stream: 32
The X drawable can die before we notice it in free_drawable_hashtable(),
called through vaTerminate(). This patch is ported from Mesa (GLX/DRI2).
Regards,
Gwenole.
Gwenole Beauchesne [Wed, 20 Jul 2011 14:48:51 +0000 (16:48 +0200)]
Fix make dist.
Gwenole Beauchesne [Wed, 20 Jul 2011 14:19:52 +0000 (16:19 +0200)]
Fix out-of-source builds.
Gwenole Beauchesne [Wed, 20 Jul 2011 11:10:08 +0000 (13:10 +0200)]
i965_drv_video: cosmetics (cleanup Makefile).
Gwenole Beauchesne [Wed, 20 Jul 2011 11:01:28 +0000 (13:01 +0200)]
cosmetics: clean up Makefiles.
Gwenole Beauchesne [Wed, 20 Jul 2011 08:56:24 +0000 (10:56 +0200)]
VA/X11: drop useless casts.
Gwenole Beauchesne [Wed, 20 Jul 2011 08:52:29 +0000 (10:52 +0200)]
VA/GLX: drop useless casts.
Austin Yuan [Tue, 26 Jul 2011 05:57:42 +0000 (13:57 +0800)]
vatrace: trace JPEG encode parameters
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Austin Yuan [Thu, 14 Jul 2011 03:18:44 +0000 (11:18 +0800)]
Update on vatrace/test and va.h
vatrace:print H263 decode paramters
test/putsurfac: create drawable with a geometry from commandline
va.h: add BGRA fourcc
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Xiang, Haihao [Tue, 12 Jul 2011 07:28:29 +0000 (15:28 +0800)]
i965_drv_video: update post processing interface
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Fri, 8 Jul 2011 02:54:31 +0000 (10:54 +0800)]
i965_drv_video: fixes assertion failure
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Thu, 7 Jul 2011 07:55:58 +0000 (15:55 +0800)]
i965_drv_video: fix next slice vertical position for field picture
This fixes https://bugs.freedesktop.org/show_bug.cgi?id=38628
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Austin Yuan [Thu, 30 Jun 2011 09:23:29 +0000 (17:23 +0800)]
vatrace: print MPEG4/H263 encode parameters information
Change-Id: Ib39b436865c2b665dc49c916d5f364f03e8c79b5
igned-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Gwenole Beauchesne [Tue, 28 Jun 2011 01:30:21 +0000 (18:30 -0700)]
i965_drv_video: add support for VA_SUBPICTURE_DESTINATION_IS_SCREEN_COORD.
Gwenole Beauchesne [Tue, 28 Jun 2011 04:19:51 +0000 (21:19 -0700)]
i965_drv_video: fix subpicture scale factor for Y axis.
Gwenole Beauchesne [Tue, 28 Jun 2011 00:02:58 +0000 (17:02 -0700)]
i965_drv_video: simplify put_surface() and put_subpicture() args.
Xiang, Haihao [Fri, 24 Jun 2011 01:20:32 +0000 (09:20 +0800)]
va/test/putsurface_x11: avoid warning 'warning: ‘win’ is used uninitialized in this function'
This fixes broken putsurface on X11
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Fri, 24 Jun 2011 01:14:10 +0000 (09:14 +0800)]
va/test: return a value to silence compiler waring
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Fri, 24 Jun 2011 01:10:47 +0000 (09:10 +0800)]
va: fix warning about memset.
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Austin Yuan [Mon, 20 Jun 2011 08:03:41 +0000 (01:03 -0700)]
test/putsurface: more cleanup, remove #ifdef ANDROID from putsurface_common.c
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Austin Yuan [Mon, 20 Jun 2011 03:15:23 +0000 (20:15 -0700)]
test/putsurface: add UV element into the auto-generated image
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Gwenole Beauchesne [Thu, 16 Jun 2011 13:03:46 +0000 (21:03 +0800)]
Fix vaTerminate() with non DRI-based drivers.
Austin Yuan [Thu, 16 Jun 2011 05:58:11 +0000 (13:58 +0800)]
va-android: same as X11, also always create a new VA display on android
Previously a VA display will be shared if a native display is shared by multiple threads,
it will casue some thread safety issue in a multi-threaded program.
Remove the global VADisplayContextP pDisplayContexts which is not thread safe
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Xiang, Haihao [Wed, 15 Jun 2011 01:11:26 +0000 (09:11 +0800)]
i965_drv_video: fix assertion failure
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Tue, 14 Jun 2011 01:49:44 +0000 (09:49 +0800)]
i965_drv_video: fix GPU hang issue when decoding field coded MPEG2 picture.
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Edgar Hucek [Mon, 13 Jun 2011 05:46:06 +0000 (13:46 +0800)]
libva: fix memeory leaks in libva
Signed-off-by: Edgar Hucek <gimli@dark-green.com>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Edgar Hucek [Mon, 13 Jun 2011 05:43:37 +0000 (13:43 +0800)]
i965_drv_video: fix memory leaks in i965 driver
Signed-off-by: Edgar Hucek <gimli@dark-green.com>
Xiang, Haihao [Mon, 13 Jun 2011 04:46:16 +0000 (12:46 +0800)]
i965_drv_video: track the internal format of a surface
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Austin Yuan [Fri, 10 Jun 2011 06:00:11 +0000 (14:00 +0800)]
Revert the change in VA_DRI2SwapBuffers: _XReply -->_XSend
We should use _XReplay. _XSend will introduce more CPU usage
Signed-off-by: Edgar Hucek <gimli@dark-green.com>
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Austin Yuan [Fri, 10 Jun 2011 05:42:56 +0000 (13:42 +0800)]
Merge branch 'master' of git+ssh://git.freedesktop.org/git/libva
Xiang, Haihao [Fri, 10 Jun 2011 02:53:29 +0000 (10:53 +0800)]
test/encode/avcenc: don't include <pciaccess.h>
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Fri, 10 Jun 2011 02:48:16 +0000 (10:48 +0800)]
i965_drv_video: encode on Ivybridge
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Thu, 9 Jun 2011 08:22:03 +0000 (16:22 +0800)]
i965_drv_video: fix VME shaders
1. The response length for inter type on Ivybridge is 6.
2. fix register region
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Thu, 9 Jun 2011 05:13:24 +0000 (13:13 +0800)]
i965_drv_video: new shaders for VME on Ivybridge
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Alexander Osin [Tue, 7 Jun 2011 18:37:11 +0000 (22:37 +0400)]
Corrected trace prints (updated, change format, etc), replaced \t by spaces in source code
Xiang, Haihao [Thu, 9 Jun 2011 01:56:16 +0000 (09:56 +0800)]
i965_drv_video: set surface base address in VME
It is easy to fill the binding table without relocation and make sure
all offsets in binding table only uses bits[15:0]
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Thu, 26 May 2011 02:33:47 +0000 (10:33 +0800)]
i965_drv_video: clean up
Don't emit PIPE_CONTROL directly, instead call intel_batchbuffer_emit_mi_flush.
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Alexander Osin [Tue, 7 Jun 2011 18:35:33 +0000 (22:35 +0400)]
i965_drv_video: Added check of obj_surface->bo field inside i965_media_h264_surface_state()
Austin Yuan [Wed, 8 Jun 2011 01:48:19 +0000 (18:48 -0700)]
test/basic testing programs do not need to be installed
in /usr/bin; one can run make valgrind here as a check,
or run them under make check too.
From gimli at dark-green.com
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Zhou Chang [Wed, 1 Jun 2011 08:01:14 +0000 (16:01 +0800)]
i965_drv_video: improved MV quality for VME
Austin Yuan [Tue, 31 May 2011 02:10:12 +0000 (19:10 -0700)]
Bump to 1.0.13
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Austin Yuan [Tue, 31 May 2011 02:02:25 +0000 (19:02 -0700)]
libva: several changes
1) add free_drawable
2) clean the license of Android.mk
3) define VA_ROTATION_270 to 3 instead of 4
Signed-off-by: Zhaohan Ren <zhaohan.ren@intel.com>
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Xiang, Haihao [Wed, 25 May 2011 06:14:42 +0000 (14:14 +0800)]
i965_drv_video: clean up codes
Check and allocate surface BO in a same function
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Wed, 25 May 2011 04:55:46 +0000 (12:55 +0800)]
i965_drv_video: rendering for Ivybridge
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Wed, 25 May 2011 01:32:50 +0000 (09:32 +0800)]
i965_drv_video: new shaders for rendering on Ivybridge
SEND on Ivybridge uses GRFs instead of MRFs
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Wed, 25 May 2011 01:27:07 +0000 (09:27 +0800)]
i965_drv_video: VC1 decoding on Ivybridge
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Wed, 25 May 2011 01:24:08 +0000 (09:24 +0800)]
i965_drv_video: H.264 & MPEG2 decoding on Ivybridge
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Tue, 17 May 2011 09:00:22 +0000 (17:00 +0800)]
i965_drv_video: Ivybridge PCI IDs
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Mon, 16 May 2011 01:49:01 +0000 (09:49 +0800)]
i965_drv_video: thread safety for object allocation
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Fri, 13 May 2011 07:45:22 +0000 (15:45 +0800)]
i965_drv_vidoe: thread safety for rendering
Xiang, Haihao [Mon, 16 May 2011 03:42:51 +0000 (11:42 +0800)]
va: always create a new VA display on X11
Previously a VA display will be shared if a native display is shared by
multiple threads, it will casue some thread safety issue in a multi-threaded
program.
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Mon, 16 May 2011 00:48:01 +0000 (08:48 +0800)]
test/putsurface/putsurface: fix multi-thread issue on X11
On X11, a VA display is dependent on the display returned by XOpenDisplay,
vaPutSurface maybe accesses this display. So the same access mechanism is needed
for vaPutSurface and other Xlib funtion calls. Another way is to call XInitThreads
before XOpenDisplay.
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Fri, 13 May 2011 07:43:51 +0000 (15:43 +0800)]
test/putsurface/putsurface: pass a valid DRAWABLE to vaPutSurface
Xiang, Haihao [Fri, 13 May 2011 05:17:58 +0000 (13:17 +0800)]
i965_drv_video: move batchbuffer to context
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Thu, 12 May 2011 08:00:30 +0000 (16:00 +0800)]
i965_drv_video: clean up batchbuffer interface
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Wed, 11 May 2011 03:20:29 +0000 (11:20 +0800)]
update .gitignore
Xiang, Haihao [Tue, 10 May 2011 05:06:22 +0000 (13:06 +0800)]
i965_drv_video: store post process parameters in context
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
Xiang, Haihao [Tue, 10 May 2011 04:52:05 +0000 (12:52 +0800)]
i965_drv_video: store kernel info in the corresponding context
Xiang, Haihao [Tue, 10 May 2011 04:45:56 +0000 (12:45 +0800)]
i965_drv_video: use the same structure for all kernels