Skip to content

Conversation

@MxkulSharma
Copy link

Since we have switched to jemalloc (bionic).

Change-Id: Ie89f9a7be4e965abf346402084a2a2d032771d3e

TheCrazyLex and others added 18 commits March 7, 2024 14:35
Sometimes we need to override ro.* properties by using our vendor init
extension.

Previously there was a security check which was blocking that.
To resolve the issue, we need to weaken the security check during the
execution of our vendor init extension.

This is safe because the vendor init extension gets executed as part of init
construction and it is considered a trusted system component.

Change-Id: I6198b453745cb92c65d3e3d49e3262354cddd2a2
Signed-off-by: Alex Naidis <alex.naidis@linux.com>
Signed-off-by: Park Ju Hyung <qkrwngud825@gmail.com>
Doing this in the userspace allows more properties to be spoofed
and eliminate the needs for a hack in the kernel.

Former-commit-id: e036a461c7dd4d97e1df77979c85f3c198e1e784
Change-Id: I76f6e210247a032b764dea2f5a23a184745f59a0

Former-commit-id: 74af52f30476991814fff83b850af1883e3944ee
Former-commit-id: 0709c6917bbd814ae88fc04ff82425c2892ef4ad
Former-commit-id: 00e6bc4f619eaad5b7fcd0524c99cf31bcc5b5e1
aswinas@pixysos: add some more props from magisk hide to userspace hack by arter97

Former-commit-id: fc79269db601c9cd0dad3781d4e6ee8f209c55fc
Change-Id: I8a88862674ca5a9eb8df5050e04344a2acb0a79f

Former-commit-id: bd20ecc029c4d82226b40f7c56185abdd59955fc
Former-commit-id: 10d57a483aca29386196d4284085c35f84d7cff9
Former-commit-id: af9890ace1d4df6eb040be59a7fe5dcbed2749f9
Change-Id: I740afaa27de82bec1e6d58b58d431141ca6b4e3f

Former-commit-id: ca62a22d017f44c9f63553a44f5017eb5b8e9095
Change-Id: Icea7076c6c0ffc2ab3d66899335a5a477ccc519a

Former-commit-id: 688821317d5bed7701362875143dc43c4b152630
Former-commit-id: ba856e490845683ea08fc22183e12ee419f064a2
Former-commit-id: 2eb3d3e12d17668c17f1f2ccb9a5149bda5ea479
Change-Id: Ic07539b4a7a97316720defd000425d1b6d15fd67

Former-commit-id: 9d4ca9403943feecd6f902e69d581aad3ee84839
Change-Id: Ic34d95c23afd8caf95c7b2a2517650dbf116fdde

Former-commit-id: 1b99c0d0cc89f113d35eb065e435f61b51408b12
Former-commit-id: 6a32cd3778d9aca584751d7acd092be95cb85985
Former-commit-id: 6bcf9412a4f3def6c2bb49507045e685df2e55a4
Also reformat code

Change-Id: I98aafcc2c1d8dae1448ecf3c18981fb7945599ba

Former-commit-id: 72351d5aebdd5ed340429bb7228ec0ce5fc55318
Former-commit-id: acefb8c963d303ba91fbe861728860b51c7c5917
Former-commit-id: 0ccf4976491b9e85a1293ccd1f062a79656cbf90
The real prop values must be retained in recovery/fastbootd in
order for fastbootd to allow/deny flashing correctly based on the
bootloader lock state. This is accomplished by checking androidboot keys
in the kernel cmdline and bootconfig (necessary on Pixel 6), and not
spoofing anything if the boot isn't a normal full-blown Android boot.

@jhenrique09 - Adapt to PE

Change-Id: I66d23fd91d82906b00d5eb020668f01ae83ec31f

Former-commit-id: 33d4578679733fb2d6fd0fd9b7baba8fd5f0be57
Former-commit-id: b929a87f6ee30e21f795442d52c6a9ece2822c66
Signed-off-by: Jarl-Penguin <jarlpenguin@outlook.com>
Change-Id: Ib6d3808c3b8f3e0cffab685a24d3cdd436b0fe9b

Former-commit-id: 739111e6414984ad9dc0d30358e2370bfb4edc29

Former-commit-id: 8bd32066ec64aa48270ba7e11042b4cca19a35e8
Former-commit-id: 8686645611965f468ee635c1e81f6f743385fb91
OnePlus SM8250 sets this as a value while rebooting the OS for some reason.
This causes the checks to fail and finally SafetyNet to fail after reboot.

Signed-off-by: zlewchan <x@zlew.tf>
Change-Id: Idc8cbd084c86b83815616be17f2a0828aa16f3af
This reverts commit 10a8c2c9c499571b217af9552623dade9d7e3e4a.

Change-Id: Ie085871215d08ab022da3b410a9f44f5344ba8ff
This reverts commit 6fb5a48277ff81fd9a72a8b0dc278582c91a17e5.

Change-Id: Ic7ed201729c7fb1d860fb687658c27826a0a855d
Checking androidboot.mode properties will never work on devices where this
property is always absent, primarily non-Pixel devices.

Use existing IsRecoveryMode() check instead which is ugly, but works for this
very purpose.

Change-Id: Idc79fb2bf45f0416b242a1e1aa12bdb07bcf56b9
Signed-off-by: Albert I <kras@raphielgang.org>
Signed-off-by: Alexander Winkowski <dereference23@outlook.com>
Critical threads go to top-app, less critical threads
currently go to system-background, but this starves
SF and leads to performance regressions. Moving
SF to the foreground cpuset still puts it below
the most critical threads, but allows it to use
some of the big cores.

Change-Id: I99050859d936d0d334877e7fa1e803b4f79bd8ec
PixelExperience/android-issues#1008
Change-Id: I4d611ee8b8490832b310fca48f2576fc9790d483
Change-Id: Icef14fa8b2ea353e3ba7cd2c6d1573dbc82474a1
Signed-off-by: xyyx <xyyx@mail.ru>
To relieve memory pressure on Android, we should start writing
dirty data sooner than later.

The Linux kernel's default is way too conservative on Android.
Android doesn't have a good reason to delay writing dirty data
anyways.

This change will make the kernel start writing dirty data when
it reaches 50M and switch to synchronous on 200M, which is far
more sensible values than 1.2G and 2.4G respectively on a device
with 12G of RAM.

Change-Id: I22f9ec9010dd028710a1a5c2e3d26d8444a4c914
Signed-off-by: Juhyung Park <qkrwngud825@gmail.com>
Since we have switched to jemalloc (bionic).

Change-Id: Ie89f9a7be4e965abf346402084a2a2d032771d3e
Co-authored-by: Gustavo Mendes <gusttavo.me@outlook.com>
Signed-off-by: MxkulSharma <sharmamukul06175@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.