ro.vendor.graphics.memory and vendor.debug.rs.* are used by Render
script (one of same-process HALs).
So they should be public-readable because Render script can be loaded
from almost everywhere.
Bug: 109653662
Test: succeeded building and tested with taimen
Change-Id: I5c6d6dd2f2406feaec60c965a763215c4a064f52
vendor.debug.egl.changepixelformat and vendor.debug.prerotation.disable
are used by libEGL (one of same-process HALs).
So they should be public-readable because libEGL can be loaded from
almost everywhere.
Bug: 80135368
Test: succeeded building and tested with taimen
Change-Id: I2e9c0809a4868329ab76a94800a144283f523579
Merged-In: I2e9c0809a4868329ab76a94800a144283f523579
(cherry picked from commit 52ca941f7a)
Since NFC has a common vendor library, adding dontaudit rules
for properties which are not used by this product.
type=1400 audit(0.0:35): avc: denied { read } for comm="nfc@1.1-service"
name="u:object_r:default_prop:s0" dev="tmpfs" ino=17612 scontext=u:r:hal_nfc_default:s0
tcontext=u:object_r:default_prop:s0 tclass=file permissive=0
Test: check logcat
Bug: 79417308
Change-Id: If2d0a1d3403851d819305f18c96c18eca35db7a8
system properties must not be used as a communication channel in between
system and vendor processes. However, there has been no enforcement on
this: system process could write system properties that are owned and
read by vendor processes and vice versa. Such communication should be
done over hwbinder and should be formally specified in HIDL.
Until we finish migrating the existing use cases of sysprops to HIDL,
whitelisting them in system_writes_vendor_properties_violators so that
the violators are clearly tracked.
These violators are allowed only for P, but not for Q.
Bug: 78598545
Test: m -j selinux_policy
Change-Id: I60b12f1232c77ad997c8c87e6d91baa14c626e94
Setting PRODUCT_COMPATIBLE_PROPERTY_OVERRIDE to true blocks
System-only-OTA from OMR1 devices because it does not allow some
propertis set by the OMR1 vendor partition.
Only if we include the VNDK snapshot v27 to wahoo devices, remove
PRODUCT_COMPATIBLE_PROPERTY_OVERRIDE from wahoo as well as ag/3694012
for rild.
Bug: 74505580
Test: Check if Bluetooth could be turned on in P-OMR1 walleye device.
Change-Id: I5113988b10f0cdde394ed2fc31ebc9d1ea811e1c
AU031 graphics drivers introduce SELinux denials for
"vendor_default_prop".
Pending a proper fix, tracking this bug so the new graphics driver
can be merged.
Test: Check that presubmit succeeds.
Bug: 79426077
Change-Id: I775de870c6fae32f35acaa7017192ef12254dd7f
Wahoo devices don't have keymaster4 hardware, just a software
implementation that was used for testing.
Bug: 77307896
Test: Check that the KM4 binary and RC are no longer on the device
Change-Id: I607937921c09bf24bb971bb7ef089aab2346debd
This should help fix presubmit tests.
Bug: 79541095
Test: Built policy.
Change-Id: Ide4401527cce5473288092a6c44fc446e9c1fc27
(cherry picked from commit 703a55c3a9)
Test:
- delete file /persist/battery/qcom_charge_full
- adb logcat to check code flow related health@2.0
- adb bugreport
- no "avc: denied" on health vendor service
- check learned capacity in both
/sys/class/power_supply/bms/charge_full
/persist/battery/qcom_charge_full
-- decrease /sys/class/power_supply/bms/charge_full
to check backup to /persist/battery/qcom_charge_full
-- increase /sys/class/power_supply/bms/charge_full
to check restore from /persist/battery/qcom_charge_full
(simluate the reset condition)
Bug: 78883741
Change-Id: Ib17bc0938d3a4b78376140a8a1008b4fc82220af
Signed-off-by: Jack Wu <wjack@google.com>
b/71936484 has audio accessories failing to connect
when there's a transition to suspend when they're attempting
to enumerate.
Avoid this problem by increasing autosuspend delay
from the default 2s to 7s.
The longest I've seen a device take to enumerate after
vbus is ~3 seconds, this gives a 2x margin.
Test: booted on Walleye, saw no more messages:
[ 472.412555] c0 5675 msm-dwc3 a800000.ssusb: DWC3 in low power mode
between device connection and enumeration.
Bug: 71936484
Change-Id: Ic655f2ce6dec429510cc7c00452da4d49f2392b0
After disabling vendor driver override, supplicant needs to specify
the max channel width and pass it to driver.
Bug: 79236591
Test: manually check channel width is HT80 when operating as GO on walleye
add p2p_go_vht=1, ch_width should be 80MHz
Note: need HIDL wpas_p2p_connect argument fix
Change-Id: I14d04556470c5efa66287f2327d7c6324885e822
