- b1bf6ce The second part of a fix for the AAC decoder to recover from corrupted audio frame. by James Dong · 15 years ago
- 2ad8941 Don't invalidate the memory pointed by the arm argument pointer, by James Dong · 15 years ago
- 325eeae am 2fb88637: am 08fad4c7: Don\'t handle input buffer in the OMX_StateLoaded state too. The input buffers have been returned while doing the state transition from OMX_StateExecuting to OMX_StateIdle. This is a similar fix for the one corresponding to output buffer (I78 by James Dong · 15 years ago
- 2fb8863 am 08fad4c7: Don\'t handle input buffer in the OMX_StateLoaded state too. The input buffers have been returned while doing the state transition from OMX_StateExecuting to OMX_StateIdle. This is a similar fix for the one corresponding to output buffer (I7887c7b9) by James Dong · 15 years ago
- 08fad4c Don't handle input buffer in the OMX_StateLoaded state too. by James Dong · 15 years ago
- 91c98e0 am fa500457: am a7b22907: Fix a bug where the conditional wait is not correct for state transition from OMX_StateIdle to OMX_StateLoaded In addition, since we have already returned all buffers to the client after completing the state transition from OMX_StateExecuting by James Dong · 15 years ago
- fa50045 am a7b22907: Fix a bug where the conditional wait is not correct for state transition from OMX_StateIdle to OMX_StateLoaded In addition, since we have already returned all buffers to the client after completing the state transition from OMX_StateExecuting to OMX_State by James Dong · 15 years ago
- a7b2290 Fix a bug where the conditional wait is not correct for state transition from OMX_StateIdle to OMX_StateLoaded by James Dong · 15 years ago
- f9e4eb4 am 7eb36421: am 5e41d631: Fix another EOS lost bug by James Dong · 15 years ago
- 13d8f58 am e11d2a35: am 2e35c709: am bf099c10: Fix a bug related to use a mutex that is meant for cross-processes sharing but is not set its attribute to enforce it. by James Dong · 15 years ago
- 0662c8f am 29ca6266: am c4a7d89c: Fix potential memory leak from video mio WM is still using the old video MIO to do rendering Over the test, I did not see it leaks, but it does not hurt to add a check. by James Dong · 15 years ago
- 7eb3642 am 5e41d631: Fix another EOS lost bug by James Dong · 15 years ago
- e11d2a3 am 2e35c709: am bf099c10: Fix a bug related to use a mutex that is meant for cross-processes sharing but is not set its attribute to enforce it. by James Dong · 15 years ago
- 29ca626 am c4a7d89c: Fix potential memory leak from video mio WM is still using the old video MIO to do rendering Over the test, I did not see it leaks, but it does not hurt to add a check. by James Dong · 15 years ago
- 5e41d63 Fix another EOS lost bug by James Dong · 15 years ago
- 2e35c70 am bf099c10: Fix a bug related to use a mutex that is meant for cross-processes sharing but is not set its attribute to enforce it. by James Dong · 15 years ago
- bf099c1 Fix a bug related to use a mutex that is meant for cross-processes sharing but is not set its attribute to enforce it. by James Dong · 15 years ago
- c4a7d89 Fix potential memory leak from video mio by James Dong · 15 years ago
- 24ef98f The input and output buffers may not be returned after the state by James Dong · 15 years ago
- d07a616 Overlay sometimes does a flush, and thus needs to be restarted. by James Dong · 15 years ago
- b11e4b8 Fix missing & bug - 2484098 by James Dong · 15 years ago
- 0db3b90 Return all buffers when the AAC encoder finishes state transition from by James Dong · 15 years ago
- 43a7253 Fix comparisons: unsigned values < 0 by James Dong · 15 years ago
- 5807caa Boost thread priorities of the mp3, aac and wbamr decoder threads. These priorities are then inherited by the DSP messaging thread as well. by Andreas Huber · 15 years ago
- 08cf2e8 Fix a EOS problem which can potentially lead to hang during video playback by James Dong · 15 years ago
- 595d195 Merge "Remove unnecessary assignment for mNumberOfFramesToHold The base class switch to use 1 already for mNumberOfFramesToHold" by James Dong · 15 years ago
- bd2ff03 Add an empty CleanSpec.mk by Jean-Baptiste Queru · 15 years ago
- 4aeb9f6 Add some logging for potential reset problem in the decoder by James Dong · 15 years ago
- 0ab8f38 Remove unnecessary assignment for mNumberOfFramesToHold by James Dong · 15 years ago
- 46cb881 Merge "Shift the bridge ioctls around to account for changes necessary for the 32 kernel" by Rebecca Schultz Zavin · 15 years ago
- 689f12a Do not use DTX VAD mode for NBAMR encoder for now and also re-enable hardware NBAMR encoder by James Dong · 15 years ago
- 3f5870a Fix a memory leak from OpenCore for authoring by James Dong · 15 years ago
- e0eb7d4 First attempt to remove redundant consecutive calls to VIDDEC_Return -rebased by James Dong · 15 years ago
- 7b285fd Restore the makefile changes by James Dong · 15 years ago
- 0ca152d Shift the bridge ioctls around to account for changes necessary for by Rebecca Schultz Zavin · 15 years ago
- a7d8550 Fix buffer ownership and buffer count when write fails by James Dong · 15 years ago
- eea0f5b Merge "Fixed a problem where a variable is not set before its value is used." by James Dong · 15 years ago
- cb134ac Merge "Return OMX_ErrorNone from OMX_GetState when OMX_GetState timed out Plus, tell the caller that the omx component is in a invalid state." by James Dong · 15 years ago
- 1535a78 Return OMX_ErrorNone from OMX_GetState when OMX_GetState timed out by James Dong · 15 years ago
- ae9edf2 Fixed a problem where a variable is not set before its value is used. by James Dong · 15 years ago
- 08303a9 Improve the quality of recorded AVC videos by enabling deblock filtering and 4x4 intra-prediction for key frames by James Dong · 15 years ago
- 17665c1 Merge "Prevent video decoder component to overwrite client's set parameter request This patch was originally from MOT/TI" by James Dong · 15 years ago
- f11ed81 Merge "Fix timestamp issue related to AAC and AMR" by James Dong · 15 years ago
- ea412a3 Prevent video decoder component to overwrite client's set parameter request by James Dong · 15 years ago
- b04e7d5 Fix timestamp issue related to AAC and AMR by James Dong · 15 years ago
- ea4296e don't use unsupported constants. by Mathias Agopian · 15 years ago
- 81080e8 Fixed three issues by James Dong · 15 years ago
- a5d6932 Merge "Fix the 0-timestamp issue Don't return an output buffer back to the omx client if the input buffer(s) processed are only meant for setting the configuration" by James Dong · 15 years ago
- 411bc95 Merge "Fix the problem where the first h264 frame in the recorded video is not marked as sync frame The root cause was that the socket node was not sending the info for SLICE bitstream mode Changes include: 1. Change I-Sync frames to be IDR frames for h264 2. Inform encoder of framerate changes to ensure correct bitrate output" by James Dong · 15 years ago
- 6b13c05 Fix the 0-timestamp issue by James Dong · 15 years ago
- 6bf5fad Fix the problem where the first h264 frame in the recorded video is not marked as sync frame by James Dong · 15 years ago
- 1380bb7 Merge "Do not assert if overlay's dequeue_buffer fails, instead silently fail." by Andreas Huber · 15 years ago
- fcd443a Do not assert if overlay's dequeue_buffer fails, instead silently fail. by Andreas Huber · 15 years ago
- de0dc1d We disable the TI AMR hardware encoder as it started emitting comfort noise in a recent update and we don't support this during playback. by Andreas Huber · 15 years ago
- ec79ebc Only return buffers back to OMX client after DSP processed them; don't return them prematurely in flush call. by James Dong · 15 years ago
- dc0a5b6 split libsurfaceflinger_client and libcamera_client out of libui by Mathias Agopian · 15 years ago
- 5c64d74 Support for both OMX_COLOR_FormatCbYCrY and OMX_COLOR_FormatYUV420Planar in the overlay rendering code. by Andreas Huber · 15 years ago
- 696f1c1 Merge "Remove (1000 ms-long) unnecessary waiting time" by James Dong · 15 years ago
- 9c3bf94 Remove (1000 ms-long) unnecessary waiting time by James Dong · 15 years ago
- a3a9de2 am a3a9c69b: am 87d1452b: reconcile main tree with open-source eclair by The Android Open Source Project · 15 years ago
- a3a9c69 am 87d1452b: reconcile main tree with open-source eclair by The Android Open Source Project · 15 years ago
- 87d1452 reconcile main tree with open-source eclair by The Android Open Source Project · 15 years ago
- b65f253 Revert "Revert "Omx component change due to TI's socket node update from 25.9 to 25.14"" by James Dong · 15 years ago
- eaaadaf Revert "Omx component change due to TI's socket node update from 25.9 to 25.14" by James Dong · 15 years ago
- 36733d6 Omx component change due to TI's socket node update from 25.9 to 25.14 by James Dong · 15 years ago
- 6d7931d reconcile android-2.1_r1 snapshot by The Android Open Source Project · 15 years ago
- 835f42e am 4f50ecc1: merge from open-source master by The Android Open Source Project · 15 years ago
- 952e6f5 Apparently the overlay needs to be created with the decoded size rather than the display size. by Andreas Huber · 15 years ago
- ab7d368 Allow set parameter call even when port is disabled With minor comment fixes by Benny Wong · 15 years ago
- 58ed709 The overlay has a new orientation parameter (not used here). by Chih-Chung Chang · 15 years ago
- 4f50ecc merge from open-source master by The Android Open Source Project · 15 years ago
- 3f188ff am e7a8c510: Increasing buffering in video encoder by Benny.Wong@motorola.com · 15 years ago
- ac4fead am 2eb87248: Re-enable TI AAC MP3 with MP3 CRCerror and EOS fix by Benny.Wong@motorola.com · 15 years ago
- 32298cf android-2.1_r1 snapshot by The Android Open Source Project · 15 years ago
- 42c043e empty initial commit by The Android Open Source Project · 15 years ago
- ef5a44a am b536cdc2: For overlay based platform, holding more than one frames in the video MIO can sometimes starve the video decoder. Thus, we set the number of video frames hold in the video MIO to be 1. by James Dong · 15 years ago
- b536cdc For overlay based platform, holding more than one frames in the video MIO by James Dong · 15 years ago
- e7a8c51 Increasing buffering in video encoder by · 15 years ago
- 2eb8724 Re-enable TI AAC MP3 with MP3 CRCerror and EOS fix by · 15 years ago
- 6239f8a Implement new getRoles API for this platform. by Andreas Huber · 15 years ago
- 20d52a6 Jpeg encoder: larger thumbnail resolution by Dan Vacura · 15 years ago
- 0bb7cc2 More clean up, mainly in UseBuffer() and AllocateBuffer() by James Dong · 15 years ago
- 604fb06 Fix bugs in getting custom extension index for JPEG encoder and decoder by James Dong · 15 years ago
- bcf0630 Fix a small bug ComponentRoleEnum() in OMX video encoder where a "goto" statement was prematurely called. by James Dong · 15 years ago
- cbd96ff Fix a memory leak under error condition by James Dong · 15 years ago
- e074e61 WMV Video decoder byte alignment fix - rebased ... by Benny Wong · 15 years ago
- c5b21bf WMV/WMA seek fix by Benny Wong · 15 years ago
- bfecd49 Improve avc encoder quality by Dan Vacura · 15 years ago
- 0f47807 Fix a small mistake where the initializer for an array actually contains more elements than it claims. by James Dong · 15 years ago
- 5579158 Merge change I29c93f80 into eclair-mr2 by Android (Google) Code Review · 15 years ago
- 2d589f6 Merge change If4081058 into eclair-mr2 by Android (Google) Code Review · 15 years ago
- f408105 Support for properly freeing an OMX node instance, now that the API for it exists. by Andreas Huber · 15 years ago
- 9b6fe75 Merge change I17e995d3 into eclair-mr2 by Android (Google) Code Review · 15 years ago
- 29c93f8 Don't enable instrumentation for production code by default by James Dong · 15 years ago
- 17e995d Make it possible to build the TI device without opencore, fix a few abuses of dlerror() by Andreas Huber · 15 years ago
- 68b4c96 FreeBuffer() for OMX AAC encoder was over complicated by James Dong · 15 years ago
- a0a60cc A certain GetParameter call on both the AMR and WBAMR codecs attempted to illegally write to the parameter block. by Andreas Huber · 15 years ago
- 3000a01 Fix race conditions where modification of some shared counters between threads was not atomic. by James Dong · 15 years ago
- 7565496 Clean up OMX_EmptyBufferDone() and OMX_FillBufferDone() by James Dong · 15 years ago
- c854c12 Address assignment was incorrect for finding the supported profile and level. by James Dong · 15 years ago