| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
| |
Prior to this change the code used uninitialized memory to for the
majority of the config. The commit message which added this code
suggests that they might not have been using this function other
than to pass a test suite?
Either way, it can't be a bad change to get the old config as default
values rather than using uninitialized memory when calling set config.
Change-Id: I1ccec4a4f206f81af3127f442fdb8b0cfdcab1ad
|
|
|
|
| |
Change-Id: I07dfb4511c8fdd4f80569cf24cb8847963982170
|
|
|
|
| |
Change-Id: Idd32704c8eb2eb3f78f619c6dace0fb124b0db82
|
|
|
|
|
|
|
|
|
|
| |
When resolution changes, we should flush video video pipeline at
first, then free surface buffers.
Bug: 23160948
Change-Id: I84782d886d8a484c2131d0b3f529f82b9c6fd8c9
Signed-off-by: mahongpe <hongpeng.ma@intel.com>
|
|
|
|
|
|
|
|
|
| |
Bug: 22602467
fix 2 CTS issues: testOtherH264FlexMaxMax/testOtherH264FlexNearMaxMax
Change-Id: Iab33ce40d506a7c7b1773913cc8a07cb02040fbf
Signed-off-by: hding3 <haitao.ding@intel.com>
|
|
|
|
|
|
|
|
|
| |
Bug: 19197299
need align with gralloc change
Change-Id: Ief1c2e960b0e2d3c8566bf32032e7050ed0a6e5c
Signed-off-by: Gu, Wangyi <wangyi.gu@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
| |
this is to explicitly declare the capability to allow apps
to read/write the video buffer. gralloc module will manage
the cache flushing better when usage flags align between
alloc() and lock().
Bug: 22183821
Change-Id: I185b14a2f97598f68a1c278309ecbffaf6e5c113
Signed-off-by: Tianmi Chen <tianmi.chen@intel.com>
|
|
|
|
|
|
|
| |
Bug: 19197299
Change-Id: Iebe2487bb2b2f761b7412d6b802ff545961551b1
Signed-off-by: Gu, Wangyi <wangyi.gu@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
a. if out port frame width parsed from extractor is
not as same as parsed from SPS/PPS nalu decoding, align it to the
result from decoder.
b. when resolution changes in adaptive playback, make sure out port
frame height is 32-align again so that graphic buffer is allocated
as 32-align.
BUG: 22184303
Change-Id: Id070b2d51fcda899e70b5b12fc9ec841bb9fd44a
Signed-off-by: Yuanjun Huang <yuanjun.huang@intel.com>
|
|
|
|
|
|
|
| |
BUG: 21777583
Change-Id: I7f8ed7463bbf85418aa09a1f15a326e4a68f4a10
Signed-off-by: Tianmi Chen <tianmi.chen@intel.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Bug: 20165724
A more comprehensive codec level resource manager is
implemented.
This reverts commit f832fbe64172f3dcde2bf8d7f960375efd8a30d9.
Change-Id: Ia45655873c9b7e9d57bed6c8d5dad29e219a2e3c
|
|
|
|
|
|
|
| |
Bug: 19197299
Change-Id: I5ee078332ae04f847ef6d8858aaf4faff12de3b9
Signed-off-by: Gu, Wangyi <wangyi.gu@intel.com>
|
|
|
|
|
|
|
| |
BUG: 21777583
Change-Id: I919dc4f9d4bd35aa8ec717f3ca18fc6e2215c59f
(cherry picked from commit 70891a61fbe9d711b5786b4026753d0c74c8fe0a)
|
|
|
|
|
| |
Change-Id: Iac1e534cf6fd4f14c819caa1d6963dd9bd2c996d
Signed-off-by: Tianmi Chen <tianmi.chen@intel.com>
|
|\
| |
| |
| | |
codec." into mnc-dev
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
for each codec.
Initial implementation.
Implement returning OMX_ErrorInsufficientResources case.
Bug: 20165724
Change-Id: I88a7229f6342bbfb8cb36b7dc9629b81e2debf93
Signed-off-by: Yuanjun Huang <yuanjun.huang@intel.com>
|
|/
|
|
|
|
|
|
|
| |
Use meta data mode for AVC and VP8 adaptive playback.
Bug: 19197299
Change-Id: Ifd635d37b107502b8008eaf539bad1dab0de443b
Signed-off-by: Tianmi Chen <tianmi.chen@intel.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
a. Add support of OMX index - media codec priority config.
config type is OMX_IndexConfigPriority.
b. Add support of OMX index - media decoder operating rate.
config type is OMX_IndexConfigOperatingRate.
Bug: 20165525
Bug: 20165726
Change-Id: Ide80507f907b3ed78792f15fafdf19b0691ec786
Signed-off-by: Yuanjun Huang <yuanjun.huang@intel.com>
|
|
|
|
|
| |
Change-Id: I115389bff99e1e5230c102d47520deb44f3d3747
(cherry picked from commit ed3af66e496bf9ecbc0f26f1983333f27044e0dd)
|
|
|
|
|
|
|
| |
Bug: 19964594
Change-Id: Iebbdda9bda3cf7a541a286c4bed7cee5f80fe748
Signed-off-by: Rajesh Poornachandran <rajesh.poornachandran@intel.com>
|
|\
| |
| |
| |
| | |
* commit 'fe8ec911ccb70006170d72031e6d7c0aed867105':
Move init of variable after setjmp.
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The gcc 4.9 compiler might not store this variable on the stack, so
it needs to be initialized after the setjmp.
(cherry-picked commit 1b2eb1beaa8da59efe10f8557c3288c294a03bb7 from aosp)
Change-Id: I8e8e8b222259312985c4abd574489315ea2d5cb4
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The gcc 4.9 compiler might not store this variable on the stack, so
it needs to be initialized after the setjmp.
(cherry picked from commit fe8ec911ccb70006170d72031e6d7c0aed867105)
Change-Id: Ie9585c8c3c075b0327093fd0f338a26edb5bb3e7
|
|\|
| |
| |
| |
| |
| |
| | |
byte-buffer mode (raw data mode). The color format needs to be set to default as non-Intel format.
* commit '7ed7ddf7e571b6d490537f43e043ea067826fd72':
Fix defect when native buffer mode fallbacks to byte-buffer mode (raw data mode). The color format needs to be set to default as non-Intel format.
|
| |\
| | |
| | |
| | |
| | |
| | |
| | | |
mode (raw data mode). The color format needs to be set to default as non-Intel format.
* commit '15f60f1eccc69ecb05ec3ae6bd075e1ca0f59ff5':
Fix defect when native buffer mode fallbacks to byte-buffer mode (raw data mode). The color format needs to be set to default as non-Intel format.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
(raw data mode). The color format needs to be set to default as
non-Intel format.
Bug: 19317169
Change-Id: Ia818c667efb1aca450b30adf9be9b4fec15955cc
Signed-off-by: Yuanjun Huang <yuanjun.huang@intel.com>
|
|/ /
| |
| |
| |
| |
| |
| | |
OMX implementation.
Change-Id: I51013f784f36299c121ea266d8b0d492a9f56901
Signed-off-by: Yuanjun Huang <yuanjun.huang@intel.com>
|
|\|
| |
| |
| |
| |
| |
| | |
beginning, and make VP8 surface height uneccessary force to 32 bit align - keep it as same as original height.
* commit 'dd94f000f413fb14aaf4f61a2bb3f944033a38c8':
VP8 OMX component - Add flushing the decode pipeline at the beginning, and make VP8 surface height uneccessary force to 32 bit align - keep it as same as original height.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
and make VP8 surface height uneccessary force to 32 bit align - keep
it as same as original height.
This fix a CTS case failure:
android.media.cts.EncodeDecodeTest#testVP8EncodeDecodeVideoFromBufferToSurfaceQVGA
Bug: 19183778
Bug: 19184881
Change-Id: Ib07798b15eb25c7f81098ba812c3d9bec5220d4e
Signed-off-by: Yuanjun Huang <yuanjun.huang@intel.com>
|
|\|
| |
| |
| |
| |
| |
| | |
CTS cases.
* commit '237fe154e8c1a073f74685e7cba7d4feba4317b0':
AVC secure OMX component - Change the level to 4.2 to pass new CTS cases.
|
| |
| |
| |
| |
| |
| |
| | |
Bug: 19080217
Change-Id: I38452077288ccf9d370b730de839af11d5277725
Signed-off-by: Yuanjun Huang <yuanjun.huang@intel.com>
|
|\ \
| | |
| | |
| | | |
Change-Id: Id7d7b4429dcaefa20dd1f1d8546037febf0ee4ac
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
* OMXVideoDecoderVP9HWR.cpp, explicitly assigning user_priv to itself.
Comment out the unused parameter.
* OMXVideoDecoderVP9Hybrid.h hides overloaded virtual function FillRenderBuffer.
(This needs more changes in all definitions and uses of FillRenderBuffer).
Change-Id: Ia973b37a6be4eb5847c393f13c206f9d1d4d4781
|
|\| |
| | |
| | |
| | |
| | | |
* commit 'af3c298ce2c1ab2ed8a3ae40fdce68e6440fc860':
Suppress some clang compiler warnings.
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
* Replace xyz = xyz; with (void)xyz;
* Ignore clang warnings about partially initialized structures.
Change-Id: I039a2493cf9fcb340ba0a71e5511d535a53bdd4d
|
|\ \ \
| | |/
| |/|
| | |
| | | |
* commit 'd7a24953f38460360b0cbfc14145b4067f9ed02b':
vp9 hybrid omx
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
if eos frame is non-shown frame, return a NULL frame with eos flag
Bug: 18771255
Change-Id: Idc1312280e0aa4a89ec475ddcb26b08867d70953
Signed-off-by: ywan171 <yi.a.wang@intel.com>
|
|\| |
| | |
| | |
| | |
| | | |
* commit '77e0e77a812f416a90713150431db498ad7559e5':
pass buffer width to libDecoderVp9Hybrid
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Bug: 18156738
BZ: 227732
Signed-off-by: ywan171 <yi.a.wang@intel.com>
Change-Id: Iaac77a4fca810bbd9adb9e4886fa05530f64d2ac
|
|\| |
| | |
| | |
| | |
| | |
| | |
| | | |
change, retain the input buffer\'s len
* commit '3b7c847ac11fe6d547346fc47b932ae0534425cc':
vp9 hybrid omx: if it's eos frame and encounter resolution change, retain the input buffer's len
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
the input buffer's len
Bug: 18635096
Signed-off-by: ywan171 <yi.a.wang@intel.com>
Change-Id: I5b433c7c94f463777652bba94c443d979cba0f0f
|
|\ \ \
| | |/
| |/|
| | | |
Change-Id: Id283ead449ea2779971c9caa07d0045798de30c1
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
For build-system CFLAGS clean-up, remove unused variables.
Bug: 18632512
Change-Id: I7644f8bc2355c27b199ec69097774b579e836048
|
|\ \ \
| | |/
| |/|
| | |
| | | |
* commit 'd5673bef309abf657748a91cbad83562de186803':
To support standard omx api for passing GTS test
|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Bug: 18665040
Change-Id: I1329040409263fa001216f6be1b85b3d28e4da09
Signed-off-by: hding3 <haitao.ding@intel.com>
|
|\ \ \
| |/ /
|/| /
| |/
| | |
* commit '8a050ed9dc33830ebab505940e05e36a1a999080':
Libvpx: Turn off unused-function warning
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
For build-system CFLAGS clean-up, disable unused-function warning
locally.
There are way too many static functions declared in headers that
are not used in all files the header is included in. Turn off the
warning for now.
Bug: 18632512
Change-Id: Ib61d235f38342816cd76c1ae27607193e7535c12
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
GRALLOC_USAGE_SW_WRITE_OFTEN
Original author: "Wang, Yi A" <yi.a.wang@intel.com>
Bug: 18647694
Bug: 18470801
BZ: 229103
Change-Id: I2b560fbd56eca7f29390f73592dff7dc165b9a5a
Signed-off-by: ywan171 <yi.a.wang@intel.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
change TARGET_HAS_VPP to TARGET_HAS_ISV
BZ: 229811
Bug: 17383204
Change-Id: I9a7212f1e51ed686f734635c5db581e2b448ef70
Signed-off-by: Jason Hu <jason.hu@intel.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
available buffer
Bug: 17729532
BZ: 225243
Change-Id: I779ce0febbd65f2a436cddc671f4156b2d1c3d69
Signed-off-by: ywan171 <yi.a.wang@intel.com>
|