summaryrefslogtreecommitdiff
path: root/libopencorehw
AgeCommit message (Collapse)Author
2011-07-12get rid of overlay HALandroid-sdk-adt_r20android-sdk-adt_r16.0.1android-sdk-4.0.3_r1android-sdk-4.0.3-tools_r1android-cts-verifier-4.0_r1android-cts-verifier-4.0.3_r1android-cts-4.0_r1android-cts-4.0.3_r2android-cts-4.0.3_r1android-4.0.4_r2.1android-4.0.4_r2android-4.0.4_r1.2android-4.0.4_r1.1android-4.0.4_r1android-4.0.3_r1.1android-4.0.3_r1android-4.0.2_r1android-4.0.1_r1.2android-4.0.1_r1.1android-4.0.1_r1tools_r20ics-plus-aospics-mr1-releaseics-mr1ics-mr0-releaseics-mr0ics-factoryrom-2-releaseMathias Agopian
Change-Id: I495a6b46ef2745406578108558b6b29400fd522f
2011-03-14hardware/ti/omap3: remove LOCAL_PRELINK_MODULEIliyan Malchev
Change-Id: Ic3f88a52f673d84d4436890a75ba1745f59e7c24 Signed-off-by: Iliyan Malchev <malchev@google.com>
2010-03-26Fix potential memory leak from video mioJames Dong
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. bug - 2385066 Change-Id: I7a32cbd5db7a36461d4ed576902f15ef91e77b5d
2010-03-04Remove unnecessary assignment for mNumberOfFramesToHoldJames Dong
The base class switch to use 1 already for mNumberOfFramesToHold bug - 2478420
2010-01-21The overlay has a new orientation parameter (not used here).Chih-Chung Chang
2010-01-06am b536cdc2: For overlay based platform, holding more than one frames in the ↵James Dong
video MIO can sometimes starve the video decoder. Thus, we set the number of video frames hold in the video MIO to be 1. Merge commit 'b536cdc2f73d9cb7148068d1164b2ad91c6c88cf' * commit 'b536cdc2f73d9cb7148068d1164b2ad91c6c88cf': For overlay based platform, holding more than one frames in the video MIO
2010-01-06For overlay based platform, holding more than one frames in the video MIOandroid-2.1_r2.1sandroid-2.1_r2.1p2android-2.1_r2.1pandroid-2.1_r2eclair-sholes-release2eclair-releaseeclair-passion-releaseJames Dong
can sometimes starve the video decoder. Thus, we set the number of video frames hold in the video MIO to be 1.
2009-12-04Make it possible to build the TI device without opencore, fix a few abuses ↵Andreas Huber
of dlerror()
2009-10-22Fix for overlay creation crashDan Vacura
When the overlay has not fully shutdown in one instance and a second instance is created the call to create the overlay will fail. Upon failure, a bug in the error handling was exposed. This patch fixes the error handling and retries the overlay creation.
2009-10-05Add support for 8 overlay buffersDan Vacura
2009-09-17Add support for software codecs in video MIO. This change allows theDave Sparks
software codecs to be used with the video overlay hardware. When the YUV420 planar format is detected, the MIO will not allocate buffers, but allow the decoder to allocate its own buffers. This change detects the YUV420 planar format, and then does a software color plane conversion to YUV422-UYVY format. This allows us to use accelerate YUV->RGB color conversion and scaling in the DSP. The primary goal is to enable fallback to software codecs when necessary.
2009-09-17Change in VMIO to use the YUV format sent from player engine;Jianhong Jiang
Don't use overlay buffer allocator as decoding buffer if YUV format is NOT the same as display YUV format.
2009-09-16Suppress buffer queue error logs during stream_offDan Vacura
Clients of the overlay abstraction do not know if streaming has been started, so they will continue to queue/dequeue buffers. This will result in extra errors logs that are not really errors. Originally from: https://partner.source.android.com/g/#change,1060
2009-09-09Fixes video playback crash. Unregister buffer should not be called because ↵Benny Wong
we never register buffers anyway. Originally from: https://partner.source.android.com/g/#change,999 We may have software fallback issue which needs to be further studied.
2009-09-08Make hi-res video playback smoother by increasing the overlay buffersBenny Wong
from 4 to 6. 6 is the number of buffers actually being supported already in the kernel. Originally from: https://partner.source.android.com/g/#change,928 Removed all changes to the LOGE so that we keep the failure log messages for now.
2009-09-01This is to ensure that the MIO is synced up with the overlay when a stream_offBenny Wong
happens as all the buffers are flushed when stream_off is called Bug 2087567 Originally from: https://partner.source.android.com/g/#change,903 Fixed a typo "=" => "==" on line 183, android_surface_output_omap34xx.cpp
2009-08-24Modified overlay_getBufferAddress function to return the overlay buffer file ↵Benny Wong
descriptor, offset and size
2009-08-19Wait for first frame to finish queuing to avoid delaysBenny Wong
2009-08-07An attempt at fixing the build.Andreas Huber
2009-07-30Modify libopencorehw to support ti dsp release 25.9Rebecca Schultz Zavin
Signed-off-by: Rebecca Schultz Zavin <rebecca@android.com>
2009-07-17Adding omap3 libopencorehw halRebecca Schultz Zavin
Signed-off-by: Rebecca Schultz Zavin <rebecca@android.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-22 20:38:30 +0000