Age | Commit message (Collapse) | Author | Files | Lines |
|
See https://github.com/01org/libva
Signed-off-by: Sean V Kelley <seanvk@posteo.de>
|
|
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99375
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
|
|
This lets any application using libva choose the best way to report
info and error messages to the user, for example graphical application
can open a popup on errors and write info messages in the toolbar.
Signed-off-by: Emmanuel Gil Peyrot <linkmauve@linkmauve.fr>
Reviewed-by: Sean V Kelley <seanvk@posteo.de>
|
|
The function strncpy() does not guarantee to nul terminate the
destination. In most cases, this cannot be triggered, but it is also
used to parse user environment variables. These are allowed to be longer
than 1023 characters, effectively resulting in an unterminated string.
I've adjusted other places as well, because it won't hurt.
https://bugs.freedesktop.org/show_bug.cgi?id=96677
|
|
Libva has a limited tracing function before. If more than one context is created
in one instance, then this tracing can't work and output wrong log messages.
This patch fixes this problem. Tracing is possible while multiple contexts are
being used.
Each log file name is suffixed by thread id. This file has all log message from
this thread. Each surface or codedbuf dump file name is suffixed by context id.
This file has all surface or codedbuf data of this context.
Signed-off-by: peng.chen <peng.c.chen@intel.com>
Reviewed-by: Sean V Kelley <sean.v.kelley@intel.com>
Tested-by: Daniel Charles <daniel.charles@intel.com>
|
|
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
|
|
va_GetDriverName fails to find the vdpau driver returning
something != SUCCESS and the driver name is not used at all.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=72822
Change-Id: I29ef398bf066badedc25de10873975ad0479dccf
Signed-off-by: Philippe Coval <rzr@gna.org>
|
|
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
|
|
Sometimes there exist multiple backend vaapi drivers on one system. In such case
user hopes to load the specific driver based on the request. Currently it can
be specified by using LIBVA_DRIVER_NAME environment. But it is not convenient.
So one API is added to allow the user to specify the loaded backend vaapi driver.
And it is called by using the following sequence:
>vaGetDisplay(
>vaSetDriverName(va_dpy, driver_name)
>vaInitialize
V1->V2: Use the reserved space to store override_driver_name based on Haihao's
suggestion.
Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
|
|
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
(cherry picked from commit 2fb37255e7c9ccb76e65ea74466f5ffafe9c68c0)
Conflicts:
configure.ac
|
|
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
|
|
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit 7f1b4be6623f5fe65063581252d2b68bd77fd076)
|
|
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit ce00603a8d58f2e25c7a6bb312f60fda0833e042)
|
|
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit e39ebdd95062995e4bdf082d50bd7d9005a3be6c)
|
|
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit 22925d9b7d1730468fc7435bee96d9d9c39f5987)
|
|
Add interfaces for low-level buffer exports to suport interop with
external APIs like EGL or OpenCL (OCL). Theory of operations:
- vaAcquireBufferHandle(): locks buffer for external API usage. This
is a synchronization point. Upon successful return, useful buffer
details are returned in a somewhat implementation-specific way to
support interop with external APIs.
- vaReleaseBufferHandle(): unlocks buffer after usage from external
API, and deallocates any resources that were needed.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
(cherry picked from commit 544d31d2fc096a838c21f4684cec05746cfdd7b6)
Conflicts:
va/va.h
|
|
Otherwise it will result in invalid reading in va_TraceEnd()
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
(cherry picked from commit d4988142a3f2256e38c5c5cdcdfc1b4f5f3c1ea9)
Tested-by: Daniel Charles <daniel.charles@intel.com> on ChromiumOS
|
|
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
|
|
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit e468728f9e49603f91ccbce3a232aa3bd98d1fd4)
[Haihao: Remove JPEG encode from this commit]
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
|
|
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
(cherry picked from commit 0b07495fc24646b47952b8df7d4e6d2df2a0f72c)
|
|
This is to fix Klockwork critical issue
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit 40e5e45edf97049137bca014575607b52d3bca78)
|
|
(cherry picked from commit 7b5c35273dc0e8e0e86cf40cf2a398119a766397)
|
|
Change-Id: If0388cd5e2e141184d88ff5de5612d15e51e3d20
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit 880c916092c6d68096c95e49405714d8a21f9322)
|
|
va_getDriverName is bypassed when LIBVA_DRIVER_NAME is set, but driver
needs DRM device fd which is only opened in va_getDriverName
Change-Id: I4fe11edf9d8c328806255d4c93313cf99655a91d
Signed-off-by: Fei Jiang <fei.jiang@intel.com>
(cherry picked from commit 04706eb13ae410b53b6b38150413d544b7fdb524)
|
|
Signed-off-by: Elaine Wang <elaine.wang@intel.com>
(cherry picked from commit 15a0fca64456d8c7de1f7aca2cb747c4929690cd)
|
|
env_value may be NULL which means the caller doens't
care the value of the environment variable, just check
if it is set.
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit 2e11d2273b2974a7d1959cbcaf8db5b8e9aedd9e)
|
|
vaEndPicture will kick HW decode/encode, and at this time use CPU
to dump surface content can keep GPU/CPU in parallel.
Also for encode trace, don't need to wait current frame completions
but for decode, we must wait decode till it is completed
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
(cherry picked from commit 0af365de4c03c12fc20183093a944f7513a1c7be)
|
|
Fix NULL pointer dereference risk issues and uninitialized variables
Signed-off-by: Elaine Wang <elaine.wang@intel.com>
Change-Id: I839a9def7b08c41c1f075e4cd1cc06e8bb0c8c9e
|
|
Add an initial implementation for vaQuerySurfaceAttributes() in terms
of vaGetSurfaceAttributes(). This is an inefficient approach because
it dynamically allocates the whole set of surface attributes we guessed
are available.
This code is to be removed once the VA drivers are updated.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
vaQuerySurfaceAttributes() is the recommended way to query the driver for
all surface attributes that are supported by the underlying hardware. This
is more useful than guessing the attribute values and testing them through
vaGetSurfaceAttributes().
The older vaGetSurfaceAttributes() function is hereby removed. However, the
relevant VA driver hook is kept for now, and will be removed/replaced for a
future update.
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
Signed-off-by: Austin Yuan <shengquan.yuan@intel.com>
|
|
|
|
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
|
|
Log functions available are now Android compatible and log statements
will show on user space android logging system.
Defined a LOG_TAG and make it available for all versions of Android.
Signed-off-by: Daniel Charles <daniel.charles@intel.com>
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
Signed-off-by: Dmitry Ermilov <dmitry.ermilov@intel.com>
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
Signed-off-by: Xiang, Haihao <haihao.xiang@intel.com>
|
|
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
Signed-off-by: Gwenole Beauchesne <gwenole.beauchesne@intel.com>
|
|
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>
|
|
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>
|
|
|
|
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
|
|
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
|
|
device when the WM is not aware of it
2. add interface VARenderMode for APP to swith between overlay and GPU post-processing
3. add VADisplayAttribOverlayColorKey/VADisplayAttribOverlayAutoPaintColorKey which is simliar to Xv for overlay post-processing
4. fix the build issue on Android 2.3
Signed-off-by: Austin Yuan <shengquan.yuan@gmail.com>
Signed-off-by: Jason Hu <jason.hu@intel.com>
Signed-off-by: Zhaohan Ren <zhaohan.ren@intel.com>
|
|
|
|
|
|
Move test/vainfo into test/vainfo/*
Add prototype of vaCreateSurfacesForUserPtr for V4L2 user pointer support
Fix klocwork critical issues. Use strnlen and snprintf instead of strlen and sprintf.
New interface vaQuerySrufaceError for decode error report and concealment
Encode data structure change (long term reference picture interface)
Add h264 constrained baseline profile support.
Signed-off-by: Elaine Wang <elaine.wang@intel.com>
Signed-off-by: Bignlin Chen <binglin.chen.wang@intel.com>
Signed-off-by: Yan Zhuang <yanx.zhunag@intel.com>
Signed-off-by: Guoliang Ji <guoliangx.ji@intel.com>
|