| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
| |
Change-Id: Idf62f2379bb7777161e71bd623dd598c4b75c04e
|
| |
|
|
|
|
|
| |
* Caf and there ancient code (-Wgnu-designator)
* Thanks to stackoverflow
Change-Id: Idae81fb1a7364f4216c310dd908a2983a0b415ba
|
| |
|
|
| |
Change-Id: Ia9071d620c3c18e052d3cc72a484a912c758fd4e
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When the firmware is still loading, mm-qcamera-daemon will
return EINVAL. Depending on firmware, loading might take a bit more
time than what's needed before the first request for camera information
occurs.
In Marshmallow, the first request for camera information comes from the
flashlight service which will trigger an EINVAL, exit the retry loop
and fail which means that enabling and disabling the torch using camera2
APIS will fail.
This patch also tolerates EINVAL as a retriable error, making the
flashlight work correctly and fixing the following CTS tests:
android.hardware.camera2.cts.FlashlightTest#testCameraDeviceOpenAfterTorchOn
android.hardware.camera2.cts.FlashlightTest#testSetTorchModeOnOff
android.hardware.camera2.cts.FlashlightTest#testTorchCallback
android.hardware.camera2.cts.FlashlightTest#testTorchModeExceptions
Change-Id: I26040dd9136a18679595a9d3bec5dcc4bbc3d541
|
| |
|
|
| |
Change-Id: I81105211e520b3a4a964acb018733b762ca1cf51
|
| |
|
|
|
|
| |
* Inspired on LineageOS/android_device_motorola_msm8916-common@10204d88d4b615221a022837c1c7c68fb0b4ed30
Change-Id: Ib5d3dce18ad88b620bdde0c6be919ff420d24da5
|
| |
|
|
| |
Change-Id: I9d3859530b96abba9f0c901f7a5cb4edd22d1e11
|
| |
|
|
|
|
|
|
|
|
| |
* error: field 'mLastAFScanTime' will be initialized after field 'mNumPreviewFaces' [-Werror,-Wreorder]
mLastAFScanTime(0),
* error: field 'mLastCaptureTime' will be initialized after field 'mNumPreviewFaces' [-Werror,-Wreorder]
mLastCaptureTime(0),
* Fix format error introduced in commit dev-harsh1998/android_device_lenovo_a6000@4fdbb02f4e9e4a06a66996fd6c31c28e15554607
Change-Id: I1779cf7bb135f8c8058d4480ba6866e91691adbe
|
| |
|
|
|
|
|
| |
Images should be padded to 64 bytes. This fixes the graphical artifacts in
images.
Change-Id: Iff57e1478d8f0ff84a67dac646d40bf2657cbff5
|
| |
|
|
|
|
|
|
| |
Offline post processing caused graphical artifacts to appear a while ago,
and it causes artifacts to appear 100% of the time when this camera HAL
is used on bacon, so disable it to be safe.
Change-Id: Ie3bd49f9475ab2270a816ffd33629b78720ab045
|
| |
|
|
|
|
| |
Better UX when video chatting.
Change-Id: I0215ff628f03bf4f6d7d6438a9e6a467dd65528f
|
| |
|
|
|
|
|
|
|
| |
Conflicts:
camera/QCamera2/HAL/QCamera2HWI.cpp
camera/QCamera2/HAL/QCamera2HWICallbacks.cpp
camera/QCamera2/stack/mm-camera-interface/src/mm_camera_stream.c
Change-Id: Ia7cc6acf0f1960bb9c99b7559f3d824df8e800a9
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Issue:
In some of the 3rd party Application's, Torch mode is set
in preview stopped state with default dimensions and
also recording hint is set to false.
so, we add a preview channel thinking it.s a flash case and
then App is setting the actual preview dimensions which we
don't honor since the preview channel is already there and
due to this mismatch in dimensions, garbled preview is seen.
Fix:
Delete the Preview Channel if it.s already there and add a
new preview channel with updated dimensions & format.
Change-Id: Ia1bd0d30a7939b8bdfd5d88fe100f696dfb4a7c6
|
| |
|
|
|
|
|
|
|
|
|
| |
Flash torch configuration shouldn't trigger
preview channel initialization/release once
we are in preview ready state. At this point
the client has already called 'startPreview'
and the flash configuration will get applied
once a valid preview window starts preview.
Change-Id: Ib44af3af593daabc7f1b901cd0b74eee773d20ad
|
| |
|
|
|
|
| |
* According to logs its breaking video recording on apps like IMO & WeChat
Change-Id: Ia943e46aff3b2a6476dbcabcd5c934642c33b868
|
| |
|
|
|
|
| |
* Backported to LA.BR.1.1.2_rb1.15 HAL by @dev_harsh1998 originally in LA.BR.1.2.9_rb1.27 commit 68ab81a24bdd35d7ec98a2addabb1223cce3de17
Change-Id: I9d2f2944a846aaeb9d5a3b4ae041b5ec63e13465
|
| |
|
|
| |
Change-Id: Ib7471ba104ab9e685b764dd8c8b3be581eeca3f5
|
| |
|
|
|
|
|
|
| |
* When the exposure time was set to a value greater than 2 seconds, the ROUND macro would round the value down to 0, causing a division by 0 and an incorrect exposure time reported in the Exif data.
* Fix the division by 0 to fix the reported exposure time for exposures greater than 2 seconds.
* Ported from SultanXDA's cam hal
Change-Id: Ib2619b1058825a77995d3c378d4700bbf9df946b
|
| |
|
|
| |
Change-Id: I1d7ed215c8c9316fbf58da70c021dc081fbb5b50
|
| |
|
|
|
|
| |
* Also add passthrough for legacy camera & graphic blobs
Change-Id: Icc7619d7dcaf60c7a36401e32139171487edbe77
|
| |
|
|
| |
Change-Id: I0d1b4f6170f3ddf8829604faf102bcba99fa0c76
|
| |
|
|
| |
Change-Id: I6563159bba4a1837e0f72ad9358384b03b8a0005
|
| |
|
|
|
|
| |
https://review.lineageos.org/#/c/188451/
Change-Id: I064cde4c4c5351f5702ff3e595261b01283c3088
|
| |
|
|
| |
Change-Id: If7b63c6a2d3e705228a2230d4a8b7e6623abf05e
|
| |
|
|
| |
Change-Id: I92b59ec7369276cd6e0022db38e6edff211a6a21
|
| |
|
|
|
|
|
|
| |
@dev-harsh1998: adapt for our older Lenovo A6000's cam hal
Test: HIDL HAL1 preview up and running
Bug: 30985004
Change-Id: I085f163de6c1c6552925c8241e744c723697d544
|
| |
|
|
| |
Change-Id: Ie62b6bfb02a35c1955ace1a93a8542974f03514b
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Due to commit I9b346b4053ec12c8a78a143a4dc0e708c44888a2
"Support EGL_KHR_partial_update without EGL_EXT_buffer_age"
in libhwui, EGL_EXT_buffer_age extension was forcibly enabled
if EGL_KHR_partial_update was available, even if the EGL driver
was not providing EGL_EXT_buffer_age support
* On older Adreno drivers, renderthread sources could use bufferAge
specific swap behaviours and would result in partially missing
display redraws upon animations and rotated screens,
for instance flashing action bars, blinking loading bars
and half black displays in landscape rotations
Change-Id: I16fe13f0726792522e382716f8f24eccf2a27701
Signed-off-by: Adrian DC <radian.dc@gmail.com>
|
| |
|
|
| |
Change-Id: Ie94f8fe6c2bbe97e034070c81a48abea8da0f820
|
| |
|
|
|
|
| |
* Fixes build errors on O
Change-Id: I9f89c3eb98158a11e8bd8552f41cb3cf8d94fdba
|
| |
|
|
|
|
| |
* causes bluetooth to hang and crash repeatedly
Change-Id: I35387618965538f76a0b0beaadbbc8a27127cb1e
|
| |
|
|
|
|
|
| |
* I personally never wanted to add shim but now device is old and our hal requires it to work on oreo
* Kanged & Adapted from luca stefni's xiaomi msm8996 repo
Change-Id: I4949536417c70aa50cc106ad2304438b922a0b3f
|
| |
|
|
| |
Change-Id: I7cc498f51a13f9b594921f5116d82c5602842d14
|
| |
|
|
| |
Change-Id: Ib581c492323676a742f3e4cf3a11a46d761f0ee1
|
| |
|
|
| |
Change-Id: I3eaed8d7961c6c5700ff42bd31580607ff034f02
|
| |
|
|
|
|
|
|
|
|
|
| |
-> @dev-harsh1998: remove algo sensor data & calibration files
-> @dev-harsh1998: Kanged a lot of goodies from oppo 8939 common
-> @dev-harsh1998: Bring sensor stack bit closer to lettuce
-> @dev-harsh1998: Bringback, update compass & gyro sensor for Lenovo K30-'X' series
-> @dev-harsh1998: cleanup while we are at this
-> @dev-harsh1998: Updated the sensors to the latest avilable CAF tag as of now
Change-Id: I278e6cce18faa7a98a52572a822e148fe27b5d3f
|
| |
|
|
| |
Change-Id: I302e2cbeae78c8a462b25ca078914fc3fb9f7cc1
|
| |
|
|
|
|
| |
* remove is-vendor-board-platform QCOM condition
Change-Id: Idb738791739103bd82e2c14c7487d618d4693fac
|
| |
|
|
| |
Change-Id: I2cea547518d625128a8086c2f950ccf88354554a
|
| |
|
|
|
|
| |
* Kill old init
Change-Id: I1a2f8c872c1d1a93cb9f9a6ff4987570bdb714b2
|
| |
|
|
|
|
| |
* These profiles are no longer supported on android O
Change-Id: I394f6a40cbdbfd76c4dc6ff2aae251cc5e737c37
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The current logging macro always uses LOGE, which is
confusing to external developer looking at our logs.
Also changed LOC_LOGx definition to follow the same
syntax as that of LOC_LOGX.
Bug: 29499503
CRs-Fixed: 1113702
Test: marlin
Change-Id: I803233a9d0b241bf9aeb2ee0d4bd2e7cc52ed75b
Merged-In: I803233a9d0b241bf9aeb2ee0d4bd2e7cc52ed75b
Signed-off-by: Iliyan Malchev <malchev@google.com>
|
| |
|
|
|
|
|
|
| |
Bug: http://b/31289077
Bug: http://b/27171986
Test: remove liblog from libcutils and recompile
Change-Id: Icd2f1ae735e263296a5c6705c7b528b00aa1a94a
(cherry picked from commit 532f3378c214c3134afcb5bcc7c19ce047ec02ac)
|
| |
|
|
|
|
| |
* find gps/ -type f -exec sed -i 's/qcom_diag/diag/g' {} \; :D Enjoy
Change-Id: Ib638e7a649cbd4e4653c1e243f4118503e4cc3ba
|
| |
|
|
|
|
|
|
| |
PRODUCT_DEFAULT_PROPERTY_OVERRIDES
* find product/ -type f -exec sed -i 's/ADDITIONAL_DEFAULT_PROPERTIES/PRODUCT_DEFAULT_PROPERTY_OVERRIDES/g' {} \;
Change-Id: I17993ed3baea11b470df4f3477873cb9a929dfa0
|
| |
|
|
|
|
| |
* During the first start of camera the buffer available to the "bufferCnt" isnt enough for starting HDR Snapshot and requires a spin between both front and back camera to get it working this commit addresses this issue and HDR works without a spin
Change-Id: I0259e7f1970e34ffe5117b1ad1760cc26325c2b9
|
| |
|
|
|
|
| |
Latest blobs from bullhead's (Google Nexus 5x) Fixes this weird issue of data not starting after some random boot or wifi switch
Change-Id: I3ec1253eacbe232d39cd2c37df7ce862f36e048c
|
| |
|
|
| |
Change-Id: I7c4269fceba12ac0dfcf246a21c35e2393ddb48b
|
| |
|
|
|
|
| |
* This should smoothen performance on bit of battery cost
Change-Id: I92a16130e0a5f0943c86be8485276b7dd8121f97
|
| |
|
|
| |
Change-Id: I611c012eeabfc1d955f2fd05002d4676d72eed55
|