* This caused numerous issues. The biggest of which being that /cache's
block device was labeled incorrectly as it was latemounted. Due to
this, Lineage Updater would fail to write to /cache directories for
updates, causing them to fail.
* Lesson learned, several optimizations Google included on the Pixel's
won't work/aren't feasible on non-A/B devices.
This reverts commit ed5a9106b1dfcfa1f1790d68e49bfc1d17e285f6.
Change-Id: I21f277c6888e06453866b8fd99004f38dce07ebb
If the noatime flag isn't set the atime information is updated on
every read access. This is bad for performance and flash lifetime.
Change-Id: I9d1a64e1a876dd239fc4593c2d7ffa75790f3a6a
This allows us to get rid of the dac_override SELinux rules, by applying
proper permissions to the files managed by the daemon. While at it, also
Treble-ized the location of said files.
Change-Id: I344b4c80412af0b22dc35da234ef3c9f88983761
Signed-off-by: Daniel Jarai <jaraidaniel@gmail.com>
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
The entire /data/misc/location is gone with Pie, and our GPS HALs
(specifically, libloc_eng) need to write in there.
Change-Id: Ic5462b21b1098604f29de0baeabdfac9abaa58fa
Signed-off-by: D. Andrei Măceș <andrei@unlegacy-android.org>
Signed-off-by: Daniel Jarai <jaraidaniel@gmail.com>
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
* Just bind mount gps.conf to gps_debug.conf
We can't directly build gps_debug.conf because it's a common target
and we're not allowed to override targets anymore.
Change-Id: Ia3c96665b50b0fe478e44186d1c65be723acbf5a
Google added a check for this prop in R, and if it's running
it will show a notification about it that it's impacting performance.
Why it's not disabled by default I don't know, so just force disable
it here.
Change-Id: Ia06ef15388e7e701ed526664e7748f133ad01ee2
The owner of /sys/class/leds/lcd-backlight/brightness is system.
Since thermald runs as root, and this node has 644 perms, writing it
generates a dac_override denial. Changing the perms to 664 fixes the
issue.
Change-Id: Ic7a3bc583c2e52c5799ef536c4b9ea07a921e54b
Signed-off-by: Daniel Jarai <jaraidaniel@gmail.com>
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
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
Makes this consistent with other targets and maken sure
any changes in healthd will be transparent to targets as long
as the symlink is maintained.
Update the seclabel accordingly too.
Test: Tested charge-only mode
Change-Id: Ic71e32f52c39bc45eae8d2552259695934592fc7
Signed-off-by: Sandeep Patil <sspatil@google.com>
There is an efs sync problem because of class type of rmt_storage service.
When encrypting phone, modem crash may occur due to not being able to sync
properly between AP and CP. So, rmt_storage class should be a core class.
Bug: 11325693
Change-Id: Ic51c3ab363b2de664c268b2ba0216b81e76189dc
Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Based on dmesg:
init: /init.duma.rc: 267: Unable to decode GID for 'qcom_oncrpc': getpwnam failed: No such file or directory
init: /init.duma.rc: 384: Unable to decode GID for 'qcom_diag': getpwnam failed: No such file or directory
* The permissions were not set to 222 instead of 220,
thus introducing regular logs from libcutils:
"cutils-trace: Error opening trace file: Permission denied (13)"
* Refers to frameworks/native/cmds/atrace/atrace.rc
Change-Id: I236ea25d2f2068f9f86a43492807f5f05e92da96
* Adapt the service from configstore-hal-1-0 service
to vendor.configstore-hal for AOSP P
* Also update gralloc-2-0 to vendor.gralloc-2-0
* Resolves surfaceflinger first launch crash
due to configstore not being running earlier enough
Change-Id: I5aae6d8357ce71933e7c6673e9e4d37adb55a6af
The kernel does not support aio with ffs.
Set the property so adb/mtp know to use the compatible
functions, and the same while in recovery.
Bug: 37916658
Test: Use adb / use adb in recovery
Change-Id: Id35a05fcb9952dcf7239bd15a61cc620537419bc
Framework will now add interface to supplicant (via HIDL) when it
needs to control an interface, so don't specify them in the startup
params.
Bug: 69426063
Test: Device boots up and able to connect to wifi.
Change-Id: I817d28093f8b982e3806c212babd888c0f5eb7c0
Wpa_supplicant's random pool is not necessary on Android. Randomness
is already provided by the entropymixer service which ensures
sufficient entropy is maintained across reboots. Commit b410eb1913
'Initialize /dev/urandom earlier in boot' seeds /dev/urandom with
that entropy before either wpa_supplicant or hostapd are run.
Bug: 34980020
Test: Use wifi and wifi tethering on flo
Change-Id: Iaa2a8a42e4e3fd941c5be0f8fa691245cdb429e9
Meh. Even with the relatively good kernel impl this doesn't work
really good for us. New zram driver doesn't correlate very well
with our ancient code in mm. This leads to a noticeable performance
degradation when the kernel starts to swap the pages into zram disk and device
becomes a laggy mess after several days of uptime even without
very active usage.
Change-Id: Ib885bcfe3961bb76ec72985a5b51f86234a6a081
* Split our fstab into two stages: early and late mount.
* Start bootanimation and graphics before /data and /cache mounting
Change-Id: I42f8bb460db80e86616c92b11752242c086b4aec
Some QCOM devices require sysfs to trigger boot/init which are blocking
the init process.
[ 7.453205] init: Command 'write /sys/kernel/boot_adsp/boot 1' action=post-fs-data (/init.angler.rc:166) returned 0 took 271.936ms.
This CL is to put those slow to start devices in a
separate service and wait for the service to be done later on.
Bug: 32712851
Test: On device
Change-Id: Idd4e965f122cbc8421b443a41573d363112dfa50
followmsi
shivatejapeddi
Nolen Johnson
elektroschmock
Arne Coucheron
Daniel Jarai
D. Andrei Măceș
Alessandro Astone
Jimmy Chen
D. Andrei Măceș
voron00
Subbaraman Narayanamurthy
Sandeep Patil
Devin Kim
Den Kondakov
dianlujitao
Adrian DC
Jerry Zhang
Roshan Pius
Jeff Vander Stoep
Artem Borisov
Wei Wang