Signed-off-by: Dhina17 <dhinalogu@gmail.com>

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

GuaiYiHu authored 5 years ago and Dhina17 committed 4 years ago

Change-Id: I58e2a50bd1e74ea11e523744ff8df285212b4d4d

Bruno Martins authored 5 years ago and Dhina17 committed 4 years ago

 * VNDK runtime isolation is kept disabled for now because that's also how Xiaomi is shipping P on their devices.

Change-Id: I43140442ab33f3e9633bb4f347e2bee9be97b447

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

-From onc-user-9-PKQ1.181021.001-V10.3.3.0.PFFINXM-release-keys

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

Signed-off-by: Dhina17 <dhinalogu@gmail.com>

* 512M disk is just too small for us and gets overfilled pretty quickly.

This reverts commit f5c43922.

Change-Id: I4bddc3fb35f81f2dce441cb269417c60c2b03956

* Our current HEAD of bootable/recovery seems to hate
  sdcards and I can't figure out how to fix it as gdb
  doesn't want to cooperate with me.
* By adding a separate recovery fstab that doesn't have
  any voldmanaged devices we can avoid crashing entire
  recovery and getting anyone stuck in it whenever they
  try to boot it.

Change-Id: Ie5d2dfedc5627f239c9bdf95ee32a96c54aa9934

This reverts commit 7e079391.

Change-Id: Idd3784f87123368435b1186d449eaae4da99928f

This reverts commit cec9091c.

Change-Id: Id44e00a5e188bc1fa072623d6010552aca8174ae

* No matter how hard QC tried to improve these features, they still suffer
  from stability issues, mainly due to the unpredictable nature of vmpressure
  they rely on.

Change-Id: Icd14c79298a3c268abffa06ed17a79dececf423a

* There's no reason in limiting this to 4 cores on msm8953.

Change-Id: I7b67c3853f75a22facb169a8e954bff37164e0d4

This partially reverts commit 77cc799c.

mido's modem is too old to be able to use qcrild in stable condition.
Keep some changes around.

Signed-off-by: Albert I <krascgq@outlook.co.id>
Change-Id: Ief5a0c4a8ad3d71195df47b8778478165fb1581f

Change-Id: I364001efbf02e8780ef7c3512fbb3452a6c3f672

Change-Id: I470b1c2ff430f12bdbfbd69341c7cfdfe972b7dd

* Due to the way newer TWRP versions are behaving, our relatively small
  persist partition gets overfilled with recovery logs very quickly.
* Let's just remove the recovery cache from persist on every boot to be sure
  that we always have enough space for anything that writes to persist.

Change-Id: Id1502dfb5e504504601a21efb7812b84a7155e58

* It should be persist.vendor.usb.config.extra, not persist.vendor.sys.usb.config.extra

Change-Id: Icc5531c5563f1975314be38f5f7567e1c0f5eb8d

* Historically, triple buffering was used to workaround missed VSYNC
  in GPU composition usecases on devices with high resolution panels.
  As a downside, an additional frame buffer increased input latency.
  Nowadays, most GPUs are fast enough to fit the VSYNC period, rendering
  triple framebuffers useless.
* Custom VSYNC offsets are intended to reduce input latency at cost of
  increased sync error counts. Our values were never calculated properly
  and were just taken from another device instead. That led to jank in
  some usecases.

Change-Id: I57ac4362ca28a9c128cc45c3f6e3734a24654955

* For some reason, booting with headphones plugged in results
  in a broken speaker output if FDE secure boot is enabled.
* Restarting audio HAL after decryption seems to resolve the issue.

Change-Id: Ie14b89841bf811f0fb09edb0a04fd28aafecde87

* We don't support NFC by hardware so these permissions are
  technically useless

Change-Id: I35ca7307208cb2d19f81260cd84da0bc0a79c576

* They are not really common, as different targets need
  different values.

Change-Id: I28c08f9b60873ad09c5ff948c53633d22312a9d2

* Use fsync_mode=nobarrier to improve random write and fsync performance
* Enable runtime discard to take advantage of asynchronous TRIM
* Remove formattable flag to be sure we always have ext4 by default

Change-Id: Ic7f6fbad178ef6251924e4414c660f4dbe1a8c44

Change-Id: Ided3922c5d74c336c31664ba39d678d391c9142c

 * CAF's init.qcom.early_boot.sh checks /persist/alarm/powerOffAlarmSet
   before setting ro.alarm_boot, but on Oreo nothing writes to that
   file, thus code is never executed and the prop is never set.
 * Without ro.alarm_boot=true, the system will boot normally instead of
   boot to min framework even on rtc, and alarm will ring after boot
   completed.

Change-Id: Ie4d30358314337c229a66a16195f675198ef7d52

 * Needed by telephony-common

Change-Id: I18c058d295ed2eaf23e8ca5d1f3933b65806ed48

 * Matches rild.legacy.rc from Pie's RIL

Change-Id: Icdb0a3e4859bc0a4011e34fdd483cef2ca3b8a7f

Bug: 66875961
Test: atest VulkanFeaturesTest
Change-Id: I58f904984320dee174637ba0ddfecdd343fab7e6

 * No need to explicitly build the packages anymore,
   since they're now display HAL dependencies.

Change-Id: Ied3be4c1e5fc7e01e0426268fb219c61a4be86e9

 * These shared libs are built when TARGET_USES_QTI_CAMERA_DEVICE
   flag is enabled.

Change-Id: Ifb39b7c2c8de2cbb1f59c6dc492a3763d32f1b7a

Change-Id: I3bfbe3e224cd7d90973e63adee4eb3d3e6bd27dc

Change-Id: I26da013dfa359ec8f184a20bc881558be0144b4a

* camera.device@3.2-impl is a dependency of camera provider
  therefore there's no need to add it to PRODUCT_PACKAGES.

Change-Id: I3ae211965a31b5f85af3dd0fa1c7811b802586c3