| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Leverage BOOTLOADER OFFSET to avoid the bootloader's wipe command
that is written when entering recovery via the key combo.
Using key-combo to enter recovery causes the bootloader to write in
"recovery.--wipe_data" which starts at offset 64 and ends at offset 83.
Rather than start at a "strange" offset like 84, round up to 128.
NOTE: ROM and recovery BOTH need to have the same offset, or you'll
probably get stuck if you try to do an OTA.
Change-Id: I5b5c5b07eada187fca0f5e77e59d33ccb62c4c6b
|
| |
|
|
| |
Change-Id: I6a5cb09801de83416105c90d9af1bcc6a0dbd3b6
|
| |
|
|
| |
Change-Id: Ib647c226ac744af3aa2e6bfadd72edd914e862f1
|
| |
|
|
|
|
|
| |
* This is only available on Nexus/Pixel devices, and we have our own
implementation in livedisplay.
Change-Id: Ic7f44d7e74fc3ce8efa6ff38bc3f03749d0d5314
|
| |
|
|
|
|
| |
(reverted from commit 8b03b18cf158ca1d6cee0f4e3bd3c211cbbc79ba)
Change-Id: Ib2d80cc22eec68408eedbf43763bf99b214ef8ac
|
| |
|
|
| |
Change-Id: I39e1d15163a6b25bd3bfad992f0dd7b4a332fe25
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Rewrite sepolicy
Based mostly on: https://github.com/LineageOS/android_device_lge_g5-common/commit/a45e48b746b80a2b3ecccfddef6a61954f69420a
* In set_baseband.te, big change:
allow set_baseband toolbox_exec:file { getattr };
- changed to -
allow set_baseband toolbox_exec:file { execute_no_trans open read getattr execute };
* Change baseband script to not require getprop
* Change baseband script to check for common variants first (speed)
* (REVERTED in PS5) ///Slight tweak to when the script is run.///
PS2:
* Fix overbroad { dac_override dac_read_search } by running service as
correct user 'system'
PS3:
* Fix PS2. The shell script is owned by root:shell. If we use user
'system' we fix the overbroad permissions. Using group 'shell' we
regain access to the shell script.
PS5:
* Revert script timing, as this may have been causing intermittent
function.
* Output baseband script tests to /dev/null so there is no shell output
Change-Id: I2e39de86e6f172596718d427aad7376f734ce883
|
| |
|
|
|
|
|
| |
Fixes denial:
avc: denied { connectto } for pid=6605 comm=omm.timeservice path=0074696D655F67656E6F6666 scontext=u:r:platform_app:s0:c512,c768 tcontext=u:r:time_daemon:s0 tclass=unix_stream_socket permissive=0
Change-Id: I6bbe409f1dcd502e54cb6e8daf028dc6d5ddc7d0
|
| |
|
|
|
|
| |
avc: denied { write } for name="tracing_on" dev="debugfs" ino=3203 scontext=u:r:init:s0 tcontext=u:object_r:debugfs:s0 tclass=file permissive=0
Change-Id: Ic4150d4a932e33124d760fce360c1ac042ed1387
|
| |
|
|
|
|
|
|
| |
Addresses several gpu related denials:
[ 146.731900 / 01-26 18:33:24.645] type=1400 audit(1485473604.645:29): avc: denied { read } for pid=6725 comm=474C546872656164203634 name=gpuclk dev=sysfs ino=12173 scontext=u:r:untrusted_app:s0:c512,c768 tcontext=u:object_r:sysfs:s0 tclass=file permissive=0
Change-Id: Iec05367ec27968891c732e051a3b83353db7827e
See: https://android.googlesource.com/device/lge/hammerhead/+/android-n-preview-5/sepolicy/file_contexts
|
| |
|
|
| |
Change-Id: I8ab0cb497d6c747156759fb3d8b1d28aa6ac283b
|
| |
|
|
|
|
| |
This is needed to get full disk encryption working.
Change-Id: I452c413d5f267da624e90cdcc81e46fcc9a15f32
|
| |
|
|
| |
Change-Id: I7f3d10afdea185a29e626446d1e59938b34cde18
|
| |
|
|
|
|
| |
ops, need to test it.
Change-Id: I36b9304cdc072caa5cab46f7f8f906f9117c5f8f
|
| |
|
|
|
|
|
|
| |
Thanks to:
Andreas Blaesius
<skate4life@gmx.de>
Change-Id: I29e659882551646b226ba9a43d89d8d5ca10322c
|
| |
|
|
| |
Change-Id: I774d852008f7dcfff23e830204c25d26068ee5b6
|
| |
|
|
| |
Change-Id: I013d1cee30a70381671c5753cfe06ed5a3486ec9
|
| |
|
|
| |
Change-Id: I2cfa5ce6df72bcf894bc07339d4edbf723745d1b
|
| |
|
|
|
|
|
| |
Allow mpdecision to read /proc entries associated with other UIDs.
Bug: 23310674
Change-Id: I952d364172d779c015bb543b80afa983f7ffee5a
|
| |
|
|
|
| |
Bug: 25864142
Change-Id: I97628282a23cb7fe0936c69f3a67a81c2afed60c
|
| |
|
|
| |
Change-Id: I6eb753bff7612249a8e5da9bce0635317777ae50
|
| |
|
|
| |
Change-Id: I485bb7532f66a6f105c3a8879881bb7887a9ee24
|
| |
|
|
| |
Change-Id: Icb739ee189dfbefe0441bd8e5ed49e0db3a3acb3
|
| |
|
|
|
| |
Change-Id: I921487b842237f481e04aecbcafaa2c4c4f8748a
(cherry picked from commit a31deac1b22fc77a9fec011d04f051fbf46939dc)
|
| |
|
|
|
|
|
| |
* Use SecureSettingSwitchPreference for Ambient Display
* Use the XML preference dependency attribute
Change-Id: I3480af71e334110aed834a53f49b33a853f16316
|
| |
|
|
|
|
|
|
| |
* We have 4 CPUs, let's utilize them.
* Use lz4 compression for zram.
Change-Id: I5195d6f59541129eb6ff55228743ae879c2f9d3f
(cherry picked from commit 7287b272a06acc0e40f8d411e08d86a2fc2bc76e)
|
| |
|
|
|
|
|
|
|
| |
ZRAM is a clear win for app switching (15% shorter ave start time
when cycling through various sets of apps). Currently using the
same size as N9 (~500M) -- may want to shrink later.
Change-Id: I61e723935503ae68594f279ebea65f13f8b94374
(cherry picked from commit f4c21cb5cbad75142660094085a73d2769ed550f)
|
| |
|
|
|
|
|
| |
* I can't believe we've been using an incorrect tunable for this long.
Change-Id: I6c4627534399047df46da0e02f802c4c4858b231
(cherry picked from commit c91cc86432ace2bff5e6c48d338a650de4b01a19)
|
| |
|
|
|
| |
Change-Id: Ice6e2da014834249772f62c2cfa2b42fe8712e2a
(cherry picked from commit c6a11cc077faf2761f1f81a1a176a4d8a6a997a2)
|
| |
|
|
|
|
|
|
| |
* Move some settings out of init to allow them to read properly
* Remove usbmode stuff, this is nonexistent
Change-Id: Ia55a70dbb6116b8786bb947dfe205c678bfb8952
(cherry picked from commit db6dba09be4e508c0b377ad284780175427b29a2)
|
| |
|
|
|
|
|
|
| |
* Adds adb entry to charge_only
* Updates product IDs
* Additional cleanups and enhancements
Change-Id: Ie1c049e553d08131e1cfdebf855fa0e5522141c2
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When atime tracking is enabled on a filesystem, simple read-only
operations result in touching the atime value, resulting in writes.
This is bad for both performance and flash lifetime.
Many historical Android devices have shipped with 'noatime' enabled,
so we're just bringing marlin/sailfish in alignment with them.
Test: booted, benchmarked before/after
Bug: 32144070
Change-Id: I333e67ed63d1804acb2a1bfaf981fa1224f0f079
(cherry picked from commit 5e2840e4db4387d8f84b402daf88fdd059707554)
|
| |
|
|
|
|
| |
* It is the default now
Change-Id: Ia0f8feec9ab378e795d1d61d7036e364fb7aec54
|
| |
|
|
| |
Change-Id: Ifa37af76b9ddfaa4e47a2a0bd8e5d244093bf80e
|
| |
|
|
|
|
|
|
|
| |
When atime tracking is enabled on a filesystem, simple read-only
operations result in touching the atime value, resulting in writes.
This is bad for both performance and flash lifetime.
Change-Id: Iaf4690cdb75c750d6fcb731fd63a3de4ed92a4ac
|
| |
|
|
|
|
| |
* Saves enough space to build the recovery image
Change-Id: I6a6e0375d0228190cb3f9f7648c5343e8132cee4
|
| |
|
|
|
|
|
|
| |
* Translations moved to the new project
android_packages_resources_devicesettings
Change-Id: Ifa634a5f08127e0747a4cff688d43fa4a0d43efc
Signed-off-by: Adrian DC <radian.dc@gmail.com>
|
| |
|
|
|
|
|
| |
The old overlays have now been removed in favour
of a single unified and extensible overlay.
Change-Id: I3113bd0d5ba880a2320209961644c0f8b435991c
|
| |\
| |
| |
| | |
into n7.1
|
| | |
| |
| |
| |
| |
| |
| |
| | |
This breaks f2fs
This reverts commit 5e354a23e803cb924a2b122cf2345e7d47c5d9c2.
Change-Id: I791ca7c4336c330428c8118f286be71ddb1646b9
|
| | |
| |
| |
| | |
Change-Id: I55a64740b42fb07f2a67d6b0c7ed25d859f57e9e
|
| |\ \
| |/
|/| |
|
| | |
| |
| |
| |
| |
| | |
This reverts commit 45d6cd1602a31ef53b2bf14ab18e5390b1907a8b.
Change-Id: I940687dd78eb9decd81ca6f3bda7569951a760ce
|
| | |
| |
| |
| |
| |
| |
| | |
* In the latest merge, flush merge is enabled by default. Disable it as
it causes unnecessary overhead with little benefit.
Change-Id: I51609b1b7ed84a3f05d78ebcdfa51bb1be5861dc
|
| | |
| |
| |
| |
| |
| | |
* Not sure why I didn't get these in previous commit.
Change-Id: Ia5ad39c6d6f448a5572bcfdf92e615dedd76ff55
|
| | |
| |
| |
| | |
Change-Id: I8d89b4773e3f5400e8244fec09508ce7cfbc2475
|
| | |
| |
| |
| |
| |
| | |
* Fix recovery fstab flag type while we're here
Change-Id: I33f114919b8d636c485a73fe23108249e83c048c
|
| | |
| |
| |
| |
| |
| | |
This reverts commit 4e18bf4b07d5353cf2c82b2b7502915bf2cbde34.
Change-Id: I44f76aa8b4d13fd5930eebbd77f5e22466f72f24
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Currently, the preference titles are very vague and confusing,
with no way for the user to understand what the preferences
do at a glance (e.g. in Settings search).
Make the purpose of the preferences more explicit through the titles.
Also, the current drawable is colored white, when it should be
colored green. This results in a near-invisible icon in the
Settings search results.
Color the drawable the appropriate teal instead.
Change-Id: I2dac9a73e8689f14d676b41922e0eba364c6500f
|
| | |
| |
| |
| |
| |
| |
| |
| | |
Bring this up to speed on the Settings changes in N:
* Use support libs for preferences
* Hook up to Settings drawer through SettingsDrawerActivity
Change-Id: I9365b3ebd1bbfed2936302e30da50e3f9af06665
|