system-core

Fork

Android 9.0.0 release 52
-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQRDQNE1cO+UXoOBCWTorT+BmrEOeAUCXhOwsgAKCRDorT+BmrEO
eKSaAJwLXTCNxEZHaZpzYp+Hdinp26lDSgCdFO9DqBHx6z8HZ4HQeeLnTT3o1Dc=
=nnJH
-----END PGP SIGNATURE-----

Merge tag 'android-9.0.0_r52' into pie-x86

Android 9.0.0 release 52

init: set default selinux mode to permissive

To support selinux enforcing mode, we still have a long way to go.
Let's set the default mode to permissive.

libsuspend: add sleep.state=force

Some devices have bad wakeup_count that prevents them from sleeping.
To workaround that, add sleep.state=force to ignore the wakeup_count
interface to put the devices into deep sleep forcibly.

FlattenableUtils::align memsets

Bug: 141890807
Test: boot, check data is zero'd
Change-Id: I45aaeac369f4c5cf3eb44f61c233e00f870a5c79
(cherry picked from commit bf824f8fa50d9f7390057278fefb58e83323a644)
(cherry picked from commit e62a9d7669ac85844d4424474f5c88c096f7160b)

Android 9.0.0 release 48
-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQRDQNE1cO+UXoOBCWTorT+BmrEOeAUCXXL99AAKCRDorT+BmrEO
eF9MAJ9VPLFhPQ58Y95hKLIBg+x4RkBgaQCgisyz5Lcp4vguVcjWDTdqaFBQ2n4=
=8LTQ
-----END PGP SIGNATURE-----

Merge tag 'android-9.0.0_r48' into pie-x86

Android 9.0.0 release 48

Fix a memory leak in gatekeeper.

In violation to the documentation of GateKeeper::GetAuthTokenKey and
GateKeeper::GetPasswordKey, the implementations in SoftGateKeeper
allocate and return buffers and relinquish ownership causing a memory
leak, because the caller expects the implementation to retain ownership.

Bug: 129768470
Bug: 134557251
Test: gatekeeper-unit-tests
Change-Id: I0af9539d3dcd47dfd1e7d80cdee700ea0c2d6d0f
Merged-In: I0af9539d3dcd47dfd1e7d80cdee700ea0c2d6d0f
(cherry picked from commit 6a9c4e7968e73393110b169b33fb636531fe7fc2)

Revert "Remove FUSE logic; it's only a sdcardfs wrapper."

This reverts commit 68e1093028185cde31dcacf34a59f6c08848d3d4.

Revert "Add support for esdfs mounting"

This reverts commit c336f86284ee817a2e846452ad80e5d014d11414.

ueventd: auto load modules on uevents (pie-x86)

This is an almost rewritten patch which implements the auto modules
loading function for pie-x86.

The commits from nougat-x86 and oreo-x86 are squashed:

* ueventd: defer modules loading if explicitly listed

* ueventd: auto load modules on uevents

* init: add modprobe

* ueventd: load modules in the main process only

* modprobe: handle module options correctly

* modprobe: fix return value

libcutils: ignore a trivial error

Sometimes it complains: "add_tid_to_cgroup failed to write '1900'
(No space left on device); fd=11". I don't understand why.
The device should have enough space.

Just ignore this error.

logd: show the concrete error if cap_set_proc fails

ueventd: fix the buggy logging

LOG(xxx) is totally broken. Use KLOG_xxx to actually show the log.

init: handle ctrl-alt-del event

By setting /proc/sys/kernel/ctrl-alt-del to 1, kernel will trap
ctrl-alt-del event and send a SIGINT signal to init.

Start the ctrl-alt-del service on receiving the signal.

fs_mgr: check ro.* property first

fs_mgr reads default fstab via ro.hardware property before Android 8.0.
However, Android 8.0 changed it to use ro.boot.hardware. The incompatible
change is bad.

Android-x86 doesn't set ro.boot.hardware anymore. Let's check ro.*
properties as before.

init: don't bail out even no SELinux domain defined

libsuspend: add powerbtnd thread

Also send wakeup key on resume to ensure the system wakes up normally.

init: set default value of ro.hardware to the target name

People usually forget to set the value of androidboot.hardware to
the target name in kernel cmdline that results in boot failure.

The change tries to reduce such an error in the future. The target
name is hardcoded to the image. This eliminates the necessity of
androidboot.hardware.

init.rc: symlink /lib to /system/lib

Currently more linux drivers use request_firmware_nowait() to load
firmwares. It requires the firmwares exist in the standard path.

Create the symlink before ueventd started to ensure firmwares
could be loaded correctly.

Fix build break on x86_64

When TARGET_ARCH=x86_64, the LOCAL_SRC_FILES will be wrongly set to
codegen.cpp.arm, but the android make will 'smartly' remove the suffix
".arm" and try to compile the codegen.cpp without the libenc dependency.

NO_REF_TASK
Tested: local build

Change-Id: I25df0692886fd8b7c56d467d3b6f674919b09b56

Support use local time for RTC

When default timezone isn't UTC, there will no persist.sys.timezone
under /data/property/, so init won't get the default timezone for
setting time from rtc.

This change adds a fallback to read the property when the persist file
does not exists.

Notice, the default property of persist.sys.timezone should be set in
/default.prop instead of /system/build.prop

NO_REF_TASK
Tested: set default timezone with Asia/Shanghai, make sure bios time
is correct in local time, reboot to android, the local time should
be correct.

Change-Id: Ifbd20cb3710f833ab65852b4e5d51e38cc7c2d79

init: support usbmisc devices

The patch allows to create usbmisc device nodes like /dev/cdc-wdm0.

libsuspend: make sleep state configurable and add a fallback

This patch allows the user to set the sleep state target from
Android properties for wakeup_count method.

It also includes a fallback state if the default state is not
available and the user didn't set the sleep.state property.

Signed-off-by: Jaap Jan Meijer <jjmeijer88@gmail.com>

Mount debugfs earlier regardless build variant

ag/835596 can be reverted after this CL is merged.

BUG:26455292
Change-Id: I453b0f1d08d822f8cd2d4d33f1c1b21f0d7f543a

init: expand variables on mount_all command

This allows mount_all to handle fstab.${ro.hardware} instead
of hardcoding the file name.

Change-Id: Ia7fb4e5a84dd50a46afdd4239b9b06204f449cfe

Software-accelerated Pixel Flinger

The typical graphic workloads on VirtualBox are improved 3X.

Supports both x86 and x86_64.

Similar change by HazouPH <jgrharbers@gmail.com>:
http://review.cyanogenmod.org/#/c/70896/

And by Quanganh pham <quanganh2627@gmail.com>:
http://review.cyanogenmod.org/#/c/97125/

Merge branch 'android-ia' into pie-x86

UPSTREAM_PENDING newfs_msdos: Handle alignment failure

When trying to create a new FAT32 file system using alignment, the resulted
MBR structure is corrupted, because the total sectors reported does not
correspond to the actual available sectors. This is due to the reserved
sectors field which is updated during the alignment.

When fsck_msdos will try to check this file system, it will fail because it
cannot match the reported sectors with the actual available sectors.
Due to the failed check, the file system cannot be mounted by vold.

I/fsck_msdos( 2180): FAT size too small, 242822 entries won't fit into 1897 sectors
I/fsck_msdos( 2180): fsck_msdos terminated by exit(8)
E/Vold ( 2180): Filesystem check failed (unknown exit code 8)
E/Vold ( 2180): /dev/block/vold/179:49 failed FS checks (I/O error)
D/Vold ( 2180): Volume sdcard1 state changing 3 (Checking) -> 1 (Idle-Unmounted)

In order to fix this issue, we will save the initial bpb structure and restore
it if the alignment fails.

Change-Id: I58d5078551c689b28c9070585bb330c45bab9c62
Signed-off-by: Irina Patru <irina.patru@intel.com>
Reviewed-on: https://android.intel.com/256992
Reviewed-by: Dolca, Robert <robert.dolca@intel.com>
Reviewed-by: cactus <cactus@intel.com>
Reviewed-by: Popa, Valentin <valentin.popa@intel.com>

diskconfig: Fix disk_layout.conf size parsing

If the user was appending 'K' to their lengths (which is valid
but redundant) it was being treated as megabytes instead of
kilobytes.

Change-Id: I979ec4d01f1342de1f3209b946201ca87bad32a7

Use progressive offset in buffer for successive writes

The code for copying from one raw file to another looped
back and tried again with a suitably smaller number of bytes
if the entire buffer was not written on the first try.
However, it failed to advance an offset in the buffer past
the bytes that were written. This could conceivably happen
under rare circumstances, leading to a corrupted disk image.

This fix defines an offset into the buffer, advancing it by
the number of bytes written on each successive retry.

Signed-off-by: Paul Drews <paul.drews@intel.com>

Change-Id: I76b8e0b3c0d3b0c32bd2114d0867ac7489808279

libdiskconfig: HACK Don't do space checks on zero-size images

If an image file is of zero size, assume we just want to expand it
to fit the available images.

Change-Id: If8f22eb12d3a5179960d4359307bb177e3323e8a
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>