| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
| |
It builds mm-qcamera-app which we don't want and we don't need it anyways
Change-Id: I4088d642ca9de7eee41d1b92add0ff5d5949a5ec
Signed-off-by: mydongistiny <jaysonedson@gmail.com>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
When enabled video stabilization will affect the way
streams are mapped in the ISP and some of the supported
stream configurations will start failing.
Unless explicitly requested by client EIS should be
disabled by default.
Bug: 80227606
Test: Manual using application,
Camera CTS
Change-Id: If14281bda6b543949988c243dedfcb12f0e8e1af
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
- unistd.h added for close, write, etc.
- libsystem_headers, libhardware_headers added because global include
path isn't provided when building with vndk
- test modules are marked as LOCAL_PROPRIETARY_MODULE := true since they
are using other proprietary modules
Bug: 69526027
Test: mma -j under QCamera2
Change-Id: Id3b59c12a89120cc7e378e873bb663926d3bd45e
|
| |
|
|
|
|
|
|
|
|
|
| |
This is basically a cherry-pick of QCOM's change
73e915617b67d794a7c4b8b9c180229a2226ab3a from
hardware/qcom/camera/msm8998/QCamera2
Bug: 69526027
Test: mma with BOARD_VNDK_VERSION=current
Test: camera (still image, recording) works
Change-Id: I8e41ce2c4822bc4c84482bb27a7a8fecf8e79d12
|
| |
|
|
|
|
|
|
|
|
|
|
| |
When the vndk is enabled (BOARD_VNDK_VERSION=current),
vendor libraries must only link against vendor variants
in the build system. This also means that they receive
restricted sets of headers and that automatic global
headers are restricted.
Test: build mm-image-codec with BOARD_VNDK_VERSION=current
Bug: 69526027
Change-Id: I4ef73cd39c1731c85fb2653eb4d5c74037407f88
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| |
| | |
Populate the public AF scene change key as well as the vendor key.
Test: CTS
Bug: 64847201
Change-Id: I912f48566d4e1020188b0f2321277ccb743506ae
|
| |\ \ |
|
| | |/
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Advertise all session based parameters accordingly. Check the
incoming session settings during stream configuration and
initialize the respective streams.
Bug: 70560715
Test: Manual, Camera CTS
Change-Id: I2d7b1f5dc381609273a62d7c7fda9e4f40a3e9fd
|
| |/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Shading mode is required by spec to be present, but wasn't covered in
CTS. Default to FAST for everything but STILL_CAPTURE.
Hot pixel mode was also not covered, and needs to be HIGH_QUALITY if
that is supported, for STILL_CAPTURE.
Also clean up a little bit of oddness in static metadata with regards
to hot pixel mode.
Test: Shading mode and hot pixel mode appear in 'dumpsys media.camera
-t' output, new CTS tests pass
Bug: 70162745
Change-Id: I9087f0f37125eb2074788e983f7e68086b76f0b2
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
A dummy batch video stream is used internally in case clients are
working in constrained operating mode without an actual video stream.
The dummy stream is using exactly the same stream parameters as the
preview which includes the gralloc usage and this will eventually
override the dummy stream type from video to preview. Trying to link
two preview streams with batching enabled in camera backend will
result in failures.
Additionally fix a possible double free of the batched buffer
definitions inside the dummy batched stream. The stream will never
start and as a result stop will not be called as well. However
batched buffer definition will get allocated and en-queued inside the
free batched queue. If we avoid stopping, we also skip flushing any of
the free buffers inside the queue. They will remain in the queue until
the
stream object gets destroyed. The queue flush implementation will try to
release the queue node data which at this point is already freed by the
camera interface layer.
Bug: 70631003
Test: Camera CTS
Change-Id: I9e0c222397a42e51a1343ab17a7285dbc9e1b76d
|
| |
|
|
|
|
|
|
| |
This reverts commit 277412e7bd9e2ad75e931f1768a3ccacc6fcc809.
Bug: 67821724
Test: Camera CTS
Change-Id: I0da586a671c74247c1974cf8b6da1d8d44cd291e
|
| |\
| |
| |
| |
| |
| |
| |
| | |
set"" into oc-mr1-dev
am: 6a37587a71
Change-Id: Id228e12c1032fe4a274b4dee5af607a1843296bb
|
| | |
| |
| |
| |
| |
| |
| |
| | |
This reverts commit 29d643327b94a5f6884b312246927f7afe835724.
Bug: 67821724
Test: Camera CTS
Change-Id: Iabec9e0afdf31a1988d8c34810219d14e74bda3e
|
| |\|
| |
| |
| |
| |
| |
| |
| | |
oc-mr1-dev
am: 6bb70203f7
Change-Id: Idff91afeff0cfde0809f835876c6b613114374a2
|
| | |\
| | |
| | |
| | | |
oc-mr1-dev
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
All vendor supported tags should be advertised in the respective
available lists.
Remove initial capability optimization. The vendor tag ops are not
initialized at this point which will lead to errors when trying to
append vendor tags.
Bug: 63817523
Test: Camera CTS
Change-Id: I417963660815b8cfe5678e8e017981f462037f0d
|
| |\| |
| | |
| | |
| | |
| | |
| | | |
am: 29d643327b
Change-Id: If916ebf52a1bc17d9b32562999c30d896f9ca002
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Extend current EIS logic such that even if there is no video stream
configured, link EIS module if the VIDEO_STABILIZATION control is on.
This fixes the API1 shim video recording issue where there is a sudden FOV
change at the start of recording when video stream is created.
Test: Use TestingCamera
Bug: 67001070
Change-Id: If0348569866729310f2d01665d8892bfb32ebfd3
|
| |\| |
| | |
| | |
| | |
| | |
| | | |
am: 6f2972c45f
Change-Id: Ic59cc208bda8a2a22c3e160a3694e6ea6b86f574
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The video stream stop sequence can encounter a race conidtion
in case pending buffers arrive during the same time frame. Any
incoming stream callbacks that manage to execute during the
window between 'mDataQ' flush and 'flushFreeBatchBufQ' will modify
the 'mBufsStaged' counter without the knowledge of the top
'QCamera3HardwareInterface' instance. This makes subsequent calls
'aggregateBufToBatch' prone to batch overflow errors.
To resolve this, call 'flushFreeBatchBufQ' before flushing 'mDataQ'.
This way calls to 'aggregateBufToBatch' from failing stream
callbacks should not be successfull in modifiying 'mBufsStaged'.
Bug: 65549208
Test: Camera CTS
Change-Id: Idd2ec4992d6a979d4308dbd2175897107bdffbe0
|
| |\| |
| | |
| | |
| | |
| | |
| | | |
am: a5a4b4d961
Change-Id: I152a1c0904c703b659732d6387ab052b5745dcce
|
| | | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
If partial result of a particular frame number is dropped, HAL needs to
trigger
notify(RESULT_ERROR) per API contract.
This means that for each dropped HAL metadata, there will be 2
RESULT_ERROR.
Test: Camera CTS
Bug: 65245828
Change-Id: Ieab9449af4f1c49e546f1431edfb7809df9db4cf
|
| |\| |
| | |
| | |
| | |
| | |
| | | |
am: 8c77ac200d
Change-Id: I04cbe5f29ca3710358110cc6e27313ad41277852
|
| | |/
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Test: compare with RAW10 output
Bug: 65552901
Change-Id: I5104571fd18ec6df2ecebb0433db4aaaec6e4be8
QCamera2: HAL3: Fix wrong conversion for first 2 pixels
Test: Camera CTS
Bug: 65552901
Change-Id: I3ee00beb9ebef811e74e95134ead886e3da6d476
|
| | |\
| | |
| | |
| | |
| | |
| | | |
am: 2634d2f9fc
Change-Id: Idd55d20e2503c562138cd94d066eee2e017b4e44
|
| |\ \ \
| | | |
| | | |
| | | |
| | | |
| | | | |
am: 0e05e0e2e1
Change-Id: Ie59c8939613eb1e9f988f7a4c9f3b8e9052ee092
|
| | |\ \ \
| | |_|/
| |/| |
| | | |
| | | |
| | | | |
am: 83de65b605
Change-Id: I2cdd5c6a8ee180c901095af4214b963a67c7f0ce
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Fix the following warning:
device/google/marlin/camera/QCamera2/HAL3/QCamera3PostProc.cpp:2365:9:
warning: Potential leak of memory pointed to by 'exif'
[clang-analyzer-cplusplus.NewDeleteLeaks]
Bug: b/27101951
Test:Warning is gone.
Change-Id: Ifc7f5f538ed303ccb1e309d8664b23c49e916b0d
|
| |\| | |
| |_|/
|/| |
| | |
| | |
| | | |
am: 0be03dfda5
Change-Id: I302e047883a55b5deb2b1ada1be42a29a94889c3
|
| | |\ \
| | | |
| | | |
| | | |
| | | |
| | | | |
am: 78ad69467e
Change-Id: I826c3de488bdde0949f80836266746f670dbe8cc
|
| | | |/
| | |
| | |
| | |
| | |
| | |
| | | |
Bug: 36716469
CRs-Fixed: 2053027
Change-Id: I6699c3682eec419fb6efb0583e9909e5739476a0
(cherry picked from commit aa3b55c1b278c52d8e93853c9f89e8a475d59026)
|
| |\ \ \ |
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
We no longer have legacy powerhal module but IPower instance.
Bug: 63039461
Test: Got powerhint from powerhal service
Change-Id: I5bc76646662efd3bcffc1ce3e374cbec68b90287
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
property_get_bool only returns true/false. If the property is set to 2,
property_get_bool returns 0.
Test: Manually test GoogleCamera with 3A debug turned on
Bug: 64614329
Change-Id: Ic5597cd9683f4d0fc5383bf4963e186822d7e0eb
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Test: Manually run camera
Bug: 64614329
Change-Id: I82188719e26d596601fbfb200b8bfa3bb99adb66
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
ImageReader buffers are invalidated when calling lock when buffer is
consumed, and it's not necessary to unconditionally invalidating them
when they return to HAL.
For HAL internal buffers, if they are consumed by reprocessing, they are
invalidated with postproc.
Test: Camera CTS
Bug: 64605640
Change-Id: I80e4a992e29a314dae444db958f8307c8a54fdb6
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Cache clean (write-back) is only necessary if actual CPU-write is
performed to generate RAW16 format.
Test: Camera CTS
Bug: 64605640
Change-Id: I6b98851fc4abdfe8f53195e4928b3e528bbfd020
|
| |/ / /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Store the flags recevied from v4l2 buffers and pass them
to reprocess channel through param.
CRs-Fixed: 1035387
Change-Id: Ic52824c036ba5f849f1feedbe0424238575ee502
mm-jpeg-interface: Jpeg cache ops cmd interface to HAL
Add Jpeg cache ops command interface to HAL.
CRs-Fixed: 1035387
Change-Id: Iebf5920675583db32da01e64142093a7400a52d9
QCamera2: Cache Op changes for stream buffers
Do not invalidate/clean invalidate by default on the stream
buffers. Assume buffers coming from kernel are clean
and do required cache invalidate, clean if HAL touches
the buffers in CPU.
Test: Camera CTS
Bug: 64605640
CRs-Fixed: 1035387
Change-Id: Ic39a73f1a759ed726002aa413df202387c1ee370
|
| |\ \ \
| | | |
| | | |
| | | |
| | | |
| | | | |
* changes:
QCamera2: HAL3: Initialize variables in PendingRequestInfo
QCamera2: HAL3: Support AF_STATE in both partial result and final result
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This makes sure the AF_STATE variables have sane values.
Test: Camera CTS
Bug: 63766738
Change-Id: I6bda97cbac0c865b140c9a59035742aada0a2345
|
| | |/ /
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Currently AF_STATE during AF_TRIGGER is sent 3 frames early
(apply_delay 2, and report_delay 1). Without AF_TRIGGER, the AF_STATE
is reported at the right time: after AF stats for the current readout
frame are analyzed.
In the case where AF_TRIGGER results in a state transition, send the new
state in partial result, and use last know AF state for all preceding
results. This shift applies to both TRIGGER_START and TRIGGER_CANCEL.
Also move af_mode back to partial result.
Test: Camera CTS, and AF_STATE during HDR+ burst.
Bug: 63766738
Change-Id: I968ca06d6909996a0b77d3b247ed0722ab92bb6a
|
| |/ /
| |
| |
| |
| |
| |
| |
| |
| |
| | |
"mOfflineMetaBuffers" and "mOfflineBuffers" can be modified
by two threads at the same leading to crashes. To avoid this
serialize the access by using an additional mutex.
Bug: 64194071
Test: Camera CTS
Change-Id: I358a1412e944084d4a54f3cdd9aa4a7eb08e8a95
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
"QCamera3YUVChannel::streamCbRoutine" will acquire "mOfflinePpLock"
before calling the channel error callback. The current callback
implementation will also try to acquire "mMutex". This could trigger
a deadlock in case a request for the YUV channel is also currently
executing. The request will acquire "mMutex" first and then try
to lock "mOfflinePpLock".
Bug: 63002361
Test: Complete Camera/Camera2 CTS.
Change-Id: I2877aebf94c5a15b6742548bcbd1d1c4adc5e9d2
|
| |\|
| |
| |
| |
| |
| | |
am: 9d4d50f881
Change-Id: Ia235a598326de8b7ea8511b29f4b2ce028adfdff
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently reprocess input buffers are unmapped upon JPEG encode
completion. This is problematic because after HAL returns input buffer
after CPP processing, it doesn't own the buffer any more.
The sequence of events leading to the issue:
1) Reprocess request 1 enters, input buffer FD is N
2) HAL caches the input buffer
3) CPP runs on input buffer, returns buffer to framework.
Cache entry remains. JPEG encoding starts
4) Framework frees buffer; FD N is now reusable
5) Reprocess request 2 enters, new buffer also gets FD N
6) HAL finds buffer in cache
7) CPP tries to run, but uses stale cache entry, crashes
8) (if no crash) JPEG encoding completes, cache cleaned up
Bug: 62721870
Test: Camera CTS
Change-Id: I22a06fad9ead03af917e547a9ee4d49d8db2e2d0
|
| |\ \
| |/
|/|
| | |
android.hardware.camera.common@1.0-helper" into oc-dr1-dev
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
libcamera_client is not a VNDK library and thus not available to
vendors. These vendor libs were using the library due to CameraMetadata
and CameraParamters classes. The classes are now available via
.common@1.0-helper library which is available to vendors. Thus switching
to the lib.
Bug: 62523241
Test: 2016 pixel devices builds and boots. Camera functionalities work.
Change-Id: Idc25ce56ee5175aea4243fbce770f240ff0b62ca
|
| |/
|
|
|
|
|
|
|
| |
Sometimes, the stream buffers come in disorder,
treat the disordered frames as error frames.
Test: Camera CTS
Bug: 37657585
Change-Id: I910153e9216ed7a83648663176c58d3696b27107
|
| |
|
|
|
|
|
|
| |
This reverts commit 900ac10969a9aff966f7c9cf06178b4ed2b84320.
Bug: 62188950
Test: Complete Camera/Camera2 CTS
Change-Id: Ia207dc26b6450185d49f828fcdabc4ea49e01471
|