summaryrefslogtreecommitdiffstats
path: root/media/libstagefright/codecs/avc
Commit message (Collapse)AuthorAgeFilesLines
* am f143b608: am d1aeff55: Merge "Fix build with clang in C++11 mode"Lajos Molnar2014-12-091-1/+1
|\ | | | | | | | | * commit 'f143b60856d275a9dcda1f3084c768655b242022': Fix build with clang in C++11 mode
| * Fix build with clang in C++11 modeBernhard Rosenkränzer2014-11-271-1/+1
| | | | | | | | | | | | | | | | | | This removes the deprecated register keyword in many places, fixes one instance of passing '\0' where a char* is expected, and 2 switch type narrowing issues. Change-Id: Ie265ac26ff6a61267ad6406dfe237226d97ca9ac Signed-off-by: Bernhard Rosenkränzer <Bernhard.Rosenkranzer@linaro.org>
* | resolved conflicts for merge of e5ba80c5 to lmp-mr1-dev-plus-aospAndreas Gampe2014-11-273-5/+3
|\ \ | |/ | | | | Change-Id: Iea12c8a6cabf84584e4a89ad80e298c1f4ea3dd7
| * Merge "Stagefright: Fix unused variables, functions, values"Andreas Gampe2014-11-263-5/+3
| |\
| | * Stagefright: Fix unused variables, functions, valuesAndreas Gampe2014-11-253-5/+3
| | | | | | | | | | | | | | | | | | | | | For build-system CFLAGS clean-up, remove unused functions and variables. Change-Id: Ic3dee56b589ea9a693efa1d72ba394036efff168
* | | am 7fca411f: am b44760cf: Merge "Remove deprecated register keyword"Chih-hung Hsieh2014-11-251-1/+1
|\ \ \ | |/ / | | | | | | | | | * commit '7fca411f36f76493491fd45f91fac2e9bd70fd6d': Remove deprecated register keyword
| * | Remove deprecated register keywordBernhard Rosenkränzer2014-11-251-1/+1
| | | | | | | | | | | | | | | | | | | | | In C++11 mode, clang errors out on the register keyword. Change-Id: I034ba5815e37e3ff5e017551977e97b8516123e8 Signed-off-by: Bernhard Rosenkränzer <Bernhard.Rosenkranzer@linaro.org>
* | | am 112ba592: am 6c7d332f: Merge "Remove bogus forward declarations"Chih-hung Hsieh2014-11-251-2/+0
|\ \ \ | |/ / | | | | | | | | | * commit '112ba59269c4d7192cb4f590b05c6058f5daafb9': Remove bogus forward declarations
| * | Merge "Remove bogus forward declarations"Chih-hung Hsieh2014-11-241-2/+0
| |\ \ | | |/ | |/|
| | * Remove bogus forward declarationsBernhard Rosenkränzer2014-11-201-2/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | Remove forward declarations of MediaBuffer -- they're not needed because MediaBuffer.h is included earlier in the file anyway, and they incorrectly declared MediaBuffer as a struct instead of a class. Change-Id: Ia35ea8f938553e39244032b6fd6c061aa388859f Signed-off-by: Bernhard Rosenkränzer <Bernhard.Rosenkranzer@linaro.org>
* | | am 128b7850: am db1ce33e: Merge "Remove check for an impossible condition"Chih-hung Hsieh2014-11-251-4/+0
|\ \ \ | |/ / | | | | | | | | | * commit '128b785027033c8c74a858a3dac27cfcb2a9ae7f': Remove check for an impossible condition
| * | Merge "Remove check for an impossible condition"Chih-hung Hsieh2014-11-241-4/+0
| |\ \ | | |/ | |/|
| | * Remove check for an impossible conditionBernhard Rosenkraenzer2014-10-041-4/+0
| | | | | | | | | | | | | | | | | | | | | | | | offset_for_ref_frame is an array rather than a pointer - it can't be NULL (and the check causes a warning in clang). Change-Id: I9b96e1e723e3b8caebd41821137ed6fb669d8cd4 Signed-off-by: Bernhard Rosenkraenzer <Bernhard.Rosenkranzer@linaro.org>
| * | resolved conflicts for merge of 2edda09a to lmp-dev-plus-aospLajos Molnar2014-10-142-113/+19
| |\ \ | | | | | | | | | | | | Change-Id: I760286b72fe1ebda3e8b058ca2138cb97fbdb7e1
* | \ \ Merge "libstagefright: set actual stride info for SW encoder input port" ↵Lajos Molnar2014-10-281-0/+4
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | into lmp-mr1-dev automerge: 701b964 automerge: 294c7f3 * commit '294c7f36e6a92419017e7fe2a204fba1284cdc08': libstagefright: set actual stride info for SW encoder input port
| * | | | libstagefright: set actual stride info for SW encoder input portLajos Molnar2014-10-251-0/+4
| | |/ / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fixes encoding flexible YUV content using SW encoders. Also skip setting input color format if it is flexible YUV, as it has already been translated to a color format supported by the codec. Bug: 18124320 Change-Id: I423782936986f4d6cf65ea9ef89ae77a92e30140
* | | | resolved conflicts for merge of f4147ee9 to lmp-mr1-dev-plus-aospLajos Molnar2014-10-142-113/+19
|\ \ \ \ | |/ / / | | / / | |/ / |/| | Change-Id: I3b66673012989ab3707852b257844f2c3851f753
| * | stagefright: fix surface input handling of software encodersLajos Molnar2014-10-132-107/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | - added SoftVideoEncoder for common color conversion and extension handling logic - fix YUV420 SemiPlanar handling that should be NV12 not NV21 Bug: 17935149 Change-Id: I9b8d05678b1862dd37bf349ea83d67bdf1bb5560
* | | am 13006b9d: am 032ecb7f: Merge "Fix in SoftAVCEncoder plugin"Narayan Kamath2014-09-041-3/+4
|\ \ \ | | |/ | |/| | | | | | | * commit '13006b9d0586d17a0ea52e98124b54ebf6246feb': Fix in SoftAVCEncoder plugin
| * | Merge "Fix in SoftAVCEncoder plugin"Narayan Kamath2014-09-041-3/+4
| |\ \
| | * | Fix in SoftAVCEncoder pluginAshok Bhat2014-07-091-3/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | size of buffer_handle_t, a pointer type, was assumed to be 4 bytes. This patch uses sizeof(buffer_handle_t) instead. In addition, replaced magic number 4 with sizeof(OMX_U32) Change-Id: Id5a7ed805a7b8e9867d8e51655cd334a66ac2c2b Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
* | | | am e4478feb: Merge "Add adaptive playback support to VPX decoder." into lmp-devRonghua Wu2014-09-021-4/+0
|\ \ \ \ | | |_|/ | |/| | | | | | | | | | * commit 'e4478febd87f6a663bbdf2c22dd829af2a8166a6': Add adaptive playback support to VPX decoder.
| * | | Add adaptive playback support to VPX decoder.Ronghua Wu2014-08-281-4/+0
| | | | | | | | | | | | | | | | | | | | Bug: 13842676 Change-Id: I9c054ea489fd3a71b3b2394f15a85b84d42edb5a
* | | | am 24a5a7c3: am b7a88e31: Merge "h264_enc: Use free() to release memory ↵Narayan Kamath2014-07-141-2/+2
|\ \ \ \ | |/ / / |/| / / | |/ / | | | | | | | | | allocated with malloc()" * commit '24a5a7c30ad041971ca511f4c7f784936090dfc6': h264_enc: Use free() to release memory allocated with malloc()
| * | h264_enc: Use free() to release memory allocated with malloc()Ashok Bhat2014-06-251-2/+2
| |/ | | | | | | | | | | | | | | | | mInputFrameData and mSliceGroup are allocated memory with malloc() but released using delete. Changed this to use free() instead. Change-Id: Iecf6684d97fabb2229d9abd94279a24faf9ac2d7 Signed-off-by: Ashok Bhat <ashok.bhat@arm.com>
| * media: memset sizeof argument issuesMark Salyzyn2014-04-111-1/+1
| | | | | | | | Change-Id: I37ef154eea0f77fb7f22faf6891f3025f6ff346e
* | Merge commit '52a23bd3' into manualmergeGlenn Kasten2014-04-031-2/+2
|\ \ | |/ | | | | Change-Id: I7a6105f467a2fa3631c62be3d306845e9cf9b365
| * 64-bit: Correction to OMX_U32 and OMX_S32Glenn Kasten2014-04-021-2/+2
| | | | | | | | Change-Id: I9c41bb34c35595ac19a7ab8faaacde8cb4fa9a15
* | Merge commit 'c250980f' into manualmergeGlenn Kasten2014-03-191-2/+0
|\ \ | |/ | | | | Change-Id: I254d456e8cb6c580dd77d602b391bed09110454e
| * libstagefright is no longer 32 bit only.Narayan Kamath2014-03-191-2/+0
| | | | | | | | Change-Id: I88d5fcfc005a2c2acd0246cdd9c08b6c00b6c39b Signed-off-by: Glenn Kasten <gkasten@android.com>
* | resolved conflicts for merge of 3c780188 to masterGlenn Kasten2014-03-141-0/+2
|\ \ | |/ | | | | Change-Id: Ic579d346c27ff05ea6444faaa60fa6caaec86fbf
| * mediaserver and associated services are 32-bit onlyGlenn Kasten2014-03-121-0/+2
| | | | | | | | | | | | also 32-bit only command-line apps Change-Id: I9ac557a8d02bbf6986a9b5c3cdce23d400b306a3
* | Even more warnings (now errors) fixed to unbreak the sdk-x86 build(s).Andreas Huber2014-02-121-1/+1
| | | | | | | | Change-Id: Ib4a5ab1a9e50fe9e7747a907f8b4a8c207099fdb
* | am 1c178d3e: am 7eacea11: Merge "avcenc: Only do startcode escaping if the ↵Lajos Molnar2014-02-111-6/+9
|\ \ | |/ | | | | | | | | | | next byte requires it" * commit '1c178d3e7b8d037ac8d348922ba841b28eee662f': avcenc: Only do startcode escaping if the next byte requires it
| * avcenc: Only do startcode escaping if the next byte requires itMartin Storsjo2014-02-111-6/+9
| | | | | | | | | | | | | | | | | | | | | | | | Section 7.4.1 in the H.264 standard says that the only valid bytes to follow a sequence that starts with 0x000003 are 0x00, 0x01, 0x02 or 0x03. This makes EncodeDecodeTest pass properly when decoding using OMX.google.h264.decoder, which is strict about the forbidden escape sequences. Change-Id: Ice113d9b934015003ea9cb10d0b21cee4d18d774
* | warnings be gone.Andreas Huber2014-02-113-4/+17
|/ | | | Change-Id: Ie3bae3f037730e316d7fca12e7a3527973f752ef
* avcenc: Update video port parameters in the base classMartin Storsjo2013-11-211-0/+11
| | | | | | | | | | This makes sure that the right parameters are returned on both input and output ports if queried after setting. This also makes sure that the output MediaFormat from the MediaCodec class contains the right video size. Change-Id: I0667b3b4c6bb90331ad0ae7d51388e1bca3d1bbd
* avcenc: Prepend startcodes to output buffersMartin Storsjo2013-09-251-0/+6
| | | | | | | | | | | | | | | | | This makes the encoder output data that is similar to what actual HW encoders output. This form is also what is expected by virtually every other component - the only exception seems to be stagefright's MPEG4Writer which happens to work with this format. This also makes the encoder's output work in other decoders when passing through directly. If feeding the encoded data straight through to a decoder (such as in the EncodeDecodeTest CTS test), the output from OMX.google.h264.encoder cannot be decoded by e.g. OMX.Nvidia.h264.decode on a Nexus 7 or OMX.SEC.avc.dec on a Galaxy S3. Change-Id: I6e16433a4a66936e3080723c8233681f53bd55e5
* Add liblogYing Wang2013-04-091-0/+1
| | | | | Bug: 8580410 Change-Id: If493d87d60d71be664ad75b140c62acadb75b0d0
* am a923dd4e: am 1b0ae9e4: Merge "avcenc: Set OMX_BUFFERFLAG_EOS on the last ↵Jean-Baptiste Queru2012-09-121-0/+3
|\ | | | | | | | | | | | | output frame" * commit 'a923dd4e6c14f5cba2fc5f6f152619ec6bc70c5d': avcenc: Set OMX_BUFFERFLAG_EOS on the last output frame
| * Merge "avcenc: Set OMX_BUFFERFLAG_EOS on the last output frame"Jean-Baptiste Queru2012-09-121-0/+3
| |\
| | * avcenc: Set OMX_BUFFERFLAG_EOS on the last output frameMartin Storsjo2012-08-291-0/+3
| | | | | | | | | | | | | | | | | | Without this, the caller might hang waiting for more output buffers. Change-Id: Ief6be00b7600be8498a953e8e88f0582ea8fce85
* | | am bf30787f: am 735d1b92: Merge "avcenc: Don\'t erase the InputBufferInfo ↵Jean-Baptiste Queru2012-08-291-1/+1
|\ \ \ | |/ / | | | | | | | | | | | | | | | before reading info from it" * commit 'bf30787f2694057e775f3b527b649a9f8074237b': avcenc: Don't erase the InputBufferInfo before reading info from it
| * | avcenc: Don't erase the InputBufferInfo before reading info from itMartin Storsjo2012-08-291-1/+1
| |/ | | | | | | | | | | | | | | | | | | Previously, the timestamp and flags were read from the InputBufferInfo struct after it had been erased. This fixes cases where the output timestamp from the avc encoder could spuriously be garbage every now and then. Change-Id: I21acf23cfac93bcc201a603feace6499ec6d8e76
* | am 8d4d590d: am 1e75e972: Merge "avcenc: Update a comment about the inline ↵Jean-Baptiste Queru2012-08-213-6/+3
|\ \ | |/ | | | | | | | | | | assembly" * commit '8d4d590dcbbd2b178976d0bd024625e9d6216365': avcenc: Update a comment about the inline assembly
| * avcenc: Update a comment about the inline assemblyMartin Storsjo2012-08-213-6/+3
| | | | | | | | | | | | | | Since the inline assembly was fixed in commit 3fdb405597f the comment was no longer accurate. Change-Id: Id4f8a111d1fa6e0c2e97f7db64262e8d1604c519
* | am 41ee747f: am 5ebce52c: Merge "avcenc: Fix the gcc inline assembly ↵Jean-Baptiste Queru2012-08-213-17/+17
|\ \ | |/ | | | | | | | | | | constraints" * commit '41ee747fc019eb5856454c698adf23f55073c69e': avcenc: Fix the gcc inline assembly constraints
| * avcenc: Fix the gcc inline assembly constraintsMartin Storsjo2012-08-203-17/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use +r for registers whose value is used as input to the block. Also switch input registers to output registers with +r, for registers that are modified by the assembly block (SUB_SAD in sad_inline.h). Without this, the inline assembly fails on modern gcc versions (tested on gcc 4.6), where the intended input value never is loaded into the registers, causing segfaults at runtime. Currently this assembly isn't used, but this change makes it work properly if enabled. Change-Id: I37c247cd5c8aff213445e31ab3bd33d8f0f37e05
* | am 894a7e45: am 9e13e927: Merge "avcenc: Properly indent assembly blocks"Jean-Baptiste Queru2012-08-213-18/+108
|\ \ | |/ | | | | | | * commit '894a7e450704b0c7858a2187a5432b479ac69c1d': avcenc: Properly indent assembly blocks
| * avcenc: Properly indent assembly blocksMartin Storsjo2012-08-203-18/+108
| | | | | | | | | | | | | | Also line break multiline assembly blocks - previously they were virtually unreadable. Change-Id: Icb269909b78746e26b28ab7dcb6979c4655a0b0c
generated by cgit v1.1 at 2024-07-04 01:12:56 +0000