Age | Commit message (Collapse) | Author |
|
Recent changes to adbd have enabled nonblocking_ffs by default.
These have seeminly uncovered some bad memory corruption in the
kernel ffs gadget driver on every kernel I've tried so far.
Unfortunately I've not been able to diagnose what is going wrong,
so for now, lets disable the nonblocking_ffs.
We still need to root cause the issue as its a problem even with
the mainline kernel on HiKey, so its likely a generic kernel issue.
Change-Id: I7a0797cc9e8e5f263e533beea23642f768c0f017
Signed-off-by: John Stultz <john.stultz@linaro.org>
|
|
- Issue with returning a valid device descriptor
since BT SCO is not defined in IOProfile.
- With Android-Q changes, first attempt to retrieve device
descriptor from the avaiable HwModule may fail due to
address and encoded format conditions.
- Further device descriptor query is based on IOProfile
with devices defined in audio_policy_configuration.
- This change adds BT SCO to IOProfile.
Change-Id: I52225070e97dd702f965ca5dbb5db9d128df1faf
|
|
Bug: 72624793
Test: compilation
Change-Id: I78ef8705d9fd38db10a4d8f2bc0eac58bcf61d0a
|
|
Add CAP_SYS_BOOT capability for charger service so that it can
shutdown/reboot the system.
Change-Id: I17ccba74d5a3c38bd751c55bcf42fb3689d730a7
|
|
Make charger binary run as system instead of root. Since it
needs to handle input key events, add input as a supplemental
group in addition to system and graphics.
Change-Id: Id1dc46f335087147b7d8cf674664c26f0b511de1
|
|
charger binary is moved to /system/bin. Also the label got changed
to charger. Update it accordingly.
Change-Id: I78574238f9fc5ec028d10ab866676bfc0cd36370
|
|
Newer AOSP deprecates LOCAL_MODULE_TAGS := eng debug.
We now remove that and instead add the package to
PRODUCT_PACKAGES_ENG or PRODUCT_PACKAGES_DEBUG as applicable.
To ensure that we don't miss out on any package during this switch,
we add these explicitly, even though it might have been already
added under PRODUCT_PACKAGES at some other location.
Change-Id: I35bd67c3e7752be5bcd8f8682864062173e92724
|
|
* Disable fluence for speaker calls. This produces echo on the other side.
* Stock has a weird configuration of fluence. According to the properties
fluence is disabled at all but the log tells us that is probably enabled
in handset mode and that's why we have enabled it. Stock is using mixer
path voice-dmic-ef.
* However, fluence seems to be disabled in speaker mode: Stock uses id 11
as acdb_tx and voice-speaker-mic as mixer path which is the default for
disabled fluence.
Change-Id: I62dfcef5bfddfcc3eeb75d1c04b21275d0875efd
Signed-off-by: Arian <arian.kulmer@web.de>
|
|
Change-Id: Ia6f2ce370430e7329c5354cb16a84947e99e2542
|
|
Change-Id: Ifb8bb7161a04f17f93922d4df80fc50cb348d72c
|
|
11-11 17:09:09.364 435 435 E MediaCodecsXmlParser: Cannot find the role for a decoder of type video/divx
11-11 17:09:09.365 435 435 E MediaCodecsXmlParser: Cannot find the role for a decoder of type video/divx311
11-11 17:09:09.365 435 435 E MediaCodecsXmlParser: Cannot find the role for a decoder of type video/divx4
11-11 17:09:09.365 435 435 E MediaCodecsXmlParser: Cannot find the role for a decoder of type audio/alac
11-11 17:09:09.365 435 435 E MediaCodecsXmlParser: Cannot find the role for a decoder of type audio/dsd
Change-Id: I155b0632b961919a3d618e51d7554f45ba45f695
Signed-off-by: Arian <arian.kulmer@web.de>
|
|
Change-Id: Idd3354a94e7348fbdf2d51900e5780ceb20f98a2
|
|
This reverts commit e6dfb545d60213575c35870c9ad1e807f4489fe6.
Change-Id: Ibd74587646f505cfd590e7b12cfc972800238a7e
|
|
Change-Id: Id6860755f6e11496f1fa191d45809e6cae1b2f48
|
|
* We changed 43 to 13 and 41 to 6 previously which made the device too loud during calls
* Not overwriting the acdb id's restores the expected behaviour.
Change-Id: I6a9ac391e58a8fdcd1ea36a97d1b162244b1db66
|
|
Change-Id: Icdfdac046bc0bfc0ced4a2c1846579322f606153
|
|
Change-Id: I16199a0f3219fc468ccdb4a385f4efcd9077e68a
|
|
Change-Id: I09d42c051f400840db25dcb558dc26e51bff554c
|
|
Change-Id: I0355db4289511f3008a29cc03c772cb94d50f2dd
|
|
* With the stock blobs this doesn't work correct
Change-Id: I5ddec461c72e7e5bc3c43d0281c23f24aa307f8d
|
|
Disable backlight when blanked out.
Ref: CRACKLING-626
Change-Id: Ifc698c1470b086b0e1f083c3c234f272988d108c
|
|
* Otherwise adb will try to start and crash the device
while offmode charging at a computer.
Change-Id: I4752de69b765c4528cedf0ce58092a882bc584a3
|
|
Change-Id: I95d5f00b45c456f76b2adb3079144e8118b3ce69
|
|
|
|
Change-Id: I9db1bd8795e62f5f7cc60ea1f801b903e824d078
|
|
Change-Id: I1b2ac64116c056000bf482a613b0a4904f73b1ae
|
|
Change-Id: I73afaf78ed14b492a471ee1e4ecedf3707cdbcda
|
|
Change-Id: I64c9118fed18f250099c1bb9cffc7881c4fd5342
|
|
Change-Id: I3d5ace9ad0661e89b00741e92190a88dd04272b4
|
|
Change-Id: I755d490db1f15e827301f17a71aef8fcce784bb0
|
|
Change-Id: Ic774191714283f5b0568cbccd8e37e94f3627f61
|
|
Change-Id: I18ca325c809a5567af51e6201f8bbd0136a581bb
|
|
Change-Id: I3c24076930fd7e4698ab21607e9ca92c08160d26
|
|
Change-Id: Ied174a171ede1dfb42fefc9c386e18fc921b0cb7
|
|
device support for extended length NFC successfully tested with German
ID card and "AusweisApp2" (horribly unreliable, but working) on
cheeseburger (OnePlus 5).
Change-Id: I8db6a7eb13e2daf0fa7e7db3723db2b6b367281a
|
|
Change-Id: Ib234f09778166848d66de6ba0f1d5237d839de0c
|
|
* NFC stack in Android M has native support for Mifare Classic,
but default protocol number is wrong. This property will correct
the number for PN547C2.
* Mostly copied from sample configuration, with
NCI_DISCOVERY_TYPE_POLL_KOVIO changed to 0x77. Any other value will
cause errorneous commands sent to NFCC, and NFC stack will fall into
infinite loop.
* PSE and Android Beam is working.
Change-Id: I84b32efd27316292aab753c308df62357677d543
|
|
Change-Id: Icb7e64684d7722bd4ec350409a5e9a0be247f59d
|
|
Change-Id: Ie1d0acaab00761904a7bda1ce55a34cbcb3f5f1f
|
|
* https://github.com/LineageOS/android_system_nfc/tree/lineage-15.1/halimpl/pn54x/common
* support for pn54x has been dropped in lineage-16.0 and above
Change-Id: I19f2c3a6e3066bf9aaaaf4ddcd7d7ad867de0ed8
|
|
* firmware has been moved to /vendor/firmware
* remove code which is commented-out
Change-Id: If38da425cc2db94fb53821d3d73ba225c3169fd0
|
|
* data and system is shinano-common, rhine has a bigger data and smaller system partition
Change-Id: I193637500c9aad637d164ad0f410290ef92397f3
|
|
* BOARD_HAVE_BLUETOOTH is automatically true when BOARD_HAVE_BLUETOOTH_BCM is true
Change-Id: I9304013a1981b1475792021688d82278f6b6f118
|
|
Change-Id: Ie3fe0d32000e03c56cd6a5998c81ce8cef84af10
|
|
Change-Id: Ie8f6db203696481b4a5ee304b5f579a58293e52f
|
|
Change-Id: I0a6b116400cf542588ef885e09b25864901b6565
|
|
Change-Id: I85db3526042ec91ba51ef82be5a666631a002457
|
|
Solves the following denial.
denied { write } for pid=548 comm="init" name="storage" dev="rootfs" ino=15996 scontext=u:r:vendor_init:s0 tcontext=u:object_r:rootfs:s0 tclass=dir permissive=0
Change-Id: If246f497bd8384f0fee81469d8d29dcb528d4d4e
|
|
* The core SELinux policies for vendor_init are being restricted as to
comply with Treble. Simply adding a local rule to allow creating
folders under rootfs would be enough, but at the same time defining
the extra root folders and making use of AID/fs configuration file
to set perms is more neat and clean.
* Furthermore, with System As Root / is a read-only filesystem
so we can't create our mountpoints from init script anymore
Change-Id: Idabd7ae72e4c4fd9daac5ba3be3e6236f79f251b
|
|
Change-Id: If3761cc54f925ce7646eee95f65b0d0aa8017554
|
|
Change-Id: I5056d8f1d3d33d72abeb0129ddfe3929e18163c2
|
|
* cutils/log.h is deprecated.
Change-Id: Id29be75e58dab602ed3abcaa64571b4d565fabb3
Signed-off-by: Adrian DC <radian.dc@gmail.com>
|
|
* PRODUCT_SHIPPING_API_LEVEL is set there
Change-Id: I8adecfbaafb538277f6451a5dbc8abc279bf0c15
|
|
Do not allow access to sensors services from clients group id 3011.
Change-Id: Ib40ad5347065eb16ce90ef4212464ecd0a9ba26c
|
|
|
|
* reorder alphabetically
* remove Filesystem management tools since the are in /build now
* drop charger_res_images which is included by default now
|
|
* reorganize alphabetically
* use COMMON_PATH
* Commonize some kernel configs
* drop MALLOC_SVELTE which is deprecated
* drop TARGET_DTB_EXTRA_FLAGS which is deprecated
* drop doubled BOARD_USES_ALSA_AUDIO
* drop doubled USE_LEGACY_LOCAL_AUDIO_HAL
* enable AUDIO_FEATURE_ENABLED_HFP
* drop BOARD_HAVE_ALTERNATE_FM which is deprecated
* drop BOARD_DISABLE_FMRADIO_LIBJNI which is deprecated
* drop TARGET_PROVIDES_GPS_LOC_API which is deprecated
Change-Id: I72cd88b52134dd68c7756635471feeb73c717afe
|
|
|
|
This reverts commit 398f6cc16d50754a0d5b6bc1a7235c5766c076b6.
|
|
* Glove mode should be set from the touch HAL
* Move TFA9890 AMP and Double tap to wake to init.qcom.rc
|
|
|
|
This reverts commit 7c6eeb92c503a0da01d48ce3b61fa689e16ba390.
|
|
* Move qcamerasvr to main to it restarts on decryption
* Also trigger a qcamerasvr restart when media or all services restart,
since it needs to be properly restarted with cameraservice
Change-Id: I2361c5658359ed9f2fe937a0bd3329ca9c00e66e
|
|
* Remove entries created by init.rc
* Remove obsolete & unused entries
* Remove useless chmod when proper perms are set by mkdir
Change-Id: Ifb1a2d8c915de13438393e5a0cac7706d7decd7a
|
|
To support end-to-end DRM content playback with SG List feature,
android.hardware.drm@1.0-service needs to have access permission
to /dev/qce to map & unmap ION buffer virtual address with HW Crypto
Engine. So, set /dev/qce device permission 0660 and let it be
accessible by user "system" and group "drmrpc".
Change-Id: If9c0ed70acc2bb063344692374f51441fe84eff6
|
|
Change permissions of AMR-NB device node and change owner
and group of AMR-WB node to fix failure in opening of these
nodes from OMX component.
CRs-Fixed: 2072938
Change-Id: I5508529726ce0a3b763ac1364960def32887c6a2
|
|
rmt_storage should not be automatically killed by Android during
shutdown since it is needed to actually process the peripheral
shutdown
Change-Id: I58a29fe86955f0abad944727bb4208dad89f495d
|
|
* Remove them all.
Change-Id: Iaabd720bbdd0d7120a1ac7c05ecb13e389bf74e4
|
|
|
|
* Match the new changes introduced in O
* Important sidenote: if the kernel doesn't support one of the capabilities,
that's enough for the whole capability set to fail for that particular
service. Previous to adding the full support for these capabilities into
the kernel, the following was printed in the dmesg:
init: init.variant.rc: 5: capability 'BLOCK_SUSPEND' not supported by the kernel
While checking rild's capabilities (`cat /proc/$$/status`, where $$ is the PID),
it could be seen that those were totally messed up:
Name: rild
...
Uid: 1001 1001 1001 1001
Gid: 1001 1001 1001 1001
...
Groups: 1005 1007 2001 2901 3003 3009 3010 9998
...
CapInh: 0000000000000000
CapPrm: 0000000000000000
CapEff: 0000000000000000
CapBnd: ffffffffffffffff
CapAmb: 0000000000000000
...
After adding CAP_BLOCK_SUSPEND support to the kernel, it changed to:
Name: rild
...
Uid: 1001 1001 1001 1001
Gid: 1001 1001 1001 1001
...
Groups: 1005 1007 2001 2901 3003 3009 3010 9998
...
CapInh: 0000001000003000
CapPrm: 0000001000003000
CapEff: 0000001000003000
CapBnd: fffffff000003000
CapAmb: 0000001000003000
...
In this particular case, messed up capabilities resulted in broken
mobile data.
Change-Id: I8e5888ebeb4bcf26eed85a2c4dca320faad8f7e0
|
|
The /vendor symlink is available as soon as /system
is mounted, just use /vendor.
Change-Id: I2e92ed2abe225f1366cac418b13763bb5617f704
|
|
Change-Id: Iab2699bf26ca4402b835d1af8d4493f9b28595d9
|
|
The reason these files are moved to /vendor/etc/init/hw and
not /vendor/etc/init is because 'init' scans the /vendor/etc/init folder
for vendor service .rc fragments to import. That would either result
in duplicate imports or out-of-order imports of these hardware
specific .rc files and that *may* have unintended consequences.
In order to preserve the current import order for MTP, we move all
hardware specific .rc files to /vendor/etc/init/hw which is not
autoscanned by init BUT leave the init.${ro.hardware}.rc (i.e.
init.qcom.rc in this case) in root for 'init' to find.
Change-Id: I14531da05e2a3a256f734f8de96e1f1969c753f9
|
|
Change-Id: I460888519e447da8ab3800d62d60c5600b902da1
|
|
* Allows mpdecision to work properly on boot to restore selected power profile
Change-Id: Ib92f971f28ed875481170b056ac16aad05aadcfc
Signed-off-by: Arian <arian.kulmer@web.de>
|
|
* revert e80a06f8c2632df3036818685ac2bf9fc06556b2
* don't enable adb from here
|
|
|
|
|
|
|
|
|
|
* Move common TARGET_LD_SHIM_LIBS to msm8974-common
* Drop libsonycamera, simply replace the permission
sed -i "s|com.sonyericsson.permission.CAMERA_EXTENDED|android.permission.CAMERA\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0\x0|g" \
lib/hw/camera.vendor.qcom.so
|
|
messages during build:
device/sony/shinano-common/configs/nfcee_access_debug.xml:4: namespace error : Namespace prefix android for signature on signer is not defined
Change-Id: I7f3a1588b9cf05cb6aff11ccc59d14e75cc6e98e
|
|
* Radio blobs taken from SW_binaries_for_Xperia_AOSP_M_MR1_3.10_v13_legacy.zip
* NFC HAL from 8.1.0
Change-Id: Ic73bc81077cbd53cc1398b91d1a278de7e505988
|
|
|
|
* Also update DRM and TA related blobs for camera functionality
Change-Id: I9e2303dd5f0fcf176702c8cb9bc25ac40850ad5f
|
|
* Excluding camera and related DRM, as they were done in a seperate
commit
Change-Id: I8493c0afa0489fec23670b6fa3c2dcff1a6b3352
|
|
* Fix build.
* Revert GraphicBuffer method to Marshmallow (for 23.5.A.1.291 blobs).
* Apply module loading changes.
* TODO: Remove unneeded sources.
Change-Id: Ifce6a23840bd8c6170e0b8a714233d7b5e99acb4
|
|
* Copy from Pie framework sources.
Change-Id: I903eb0c3c80a752d4a8da44dec47e348c7dd23fc
Signed-off-by: Nikhil Punathil <nikhilpe@gmail.com>
|
|
we need updated sources that break rhine compatibility for working cam
Change-Id: I274c91efc797b0304a2074baf4b908766d321356
Signed-off-by: Nikhil Punathil <nikhilpe@gmail.com>
|
|
|
|
|
|
|
|
Change-Id: I45c5a123c86f3a9ec1ac72011c018a937d4fc70f
|
|
- Remove the define "HCILP_INCLUDED false" to allow Android's bt stack
emitting BT_WAKE calls again.
- Switch back from using the IOCTL interface to the ProcFS interface to
make the wake asserts for BT. Since we now use the brcm ldisc driver and
the interface functionally equivalent, we do not need to implement a
special IOCTL handling in the ldics driver's code.
Change-Id: I9b8fa941619aac657c4c6d6dc3053e7c1cf1d480
Signed-off-by: Alexander Diewald <Diewi@diewald-net.com>
|
|
Use ro.product.device property and naming BT dynamically.
Signed-off-by: Humberto Borba <humberos@gmail.com>
Change-Id: Iff4f4dd51a2ad5588e35c14dbf70b63e1d69a940
|
|
New name is config_dozeAfterScreenOffByDefault
Bug: 73178121
Change-Id: Ie3d1ea5a2cc480655592bf6e49b37f5ea07f3f58
Signed-off-by: Arian <arian.kulmer@web.de>
|
|
02-20 17:19:50.305 463 463 E MediaCodecsXmlParser: Cannot find the role for a decoder of type video/divx
02-20 17:19:50.305 463 463 E MediaCodecsXmlParser: Cannot find the role for a decoder of type video/divx311
02-20 17:19:50.305 463 463 E MediaCodecsXmlParser: Cannot find the role for a decoder of type video/divx4
02-20 17:19:50.305 463 463 E MediaCodecsXmlParser: Cannot find the role for a decoder of type video/x-ms-wmv
Change-Id: I21e6634343e106aee0a9ded8effc9461c3a45acf
|
|
Fixes double tap to wake on enforcing
Change-Id: Iaef8822a283e18ad40656bf679acea7de6075662
|
|
This addresses an underlying problem since /dev/hw_random is used. The
issue is that EntropyMixer, which runs in system_server, needs to read
from /dev/hw_random, but thanks to our qualcomm ueventd, EntropyMixer
can never accress this node, resulting in warnings like this:
W EntropyMixer: Failed to add HW RNG output to entropy pool
W EntropyMixer: java.io.FileNotFoundException: /dev/hw_random (Permission denied)
AOSP ueventd handles this well, so just remove this line to get things
back to stage. This issue was never exposed because Google derps the
logging here, which is already fixed on Pie.
https://android.googlesource.com/platform/frameworks/base/+/c78a463e875d91eae2cf256042fd99d130217c38
Change-Id: I42c9aaa8f57306536e51c0a17febad57a5a77581
|
|
Change-Id: I00c5cf1b55c4937e2665c84be2130deb3545423d
|