The README.md states that this ordering is not guaranteed to give
flexibility for the future, however it's time to state that this
ordering is guaranteed, especially since:
1) We have a tests, EventTriggerOrder and
EventTriggerOrderMultipleFiles, which have guaranteed this ordering
since 2017.
2) We have users requesting and depending on this order
Also update some slightly out of date parts of the documentation:
1) We import /system/etc/init/hw/init.rc instead of /init.rc as the
first import
2) We additionally import /system_ext/etc/init and /product/etc/init
Test: n/a
Change-Id: I6d7b8d9e52f0d52bee320d5074ebb74a537f9150
These logs are not generally useful and can be considered to be spam
by some. Drop them to VERBOSE so that we can be selectively enabled
when debugging logd.
Test: these logs do not appears by default.
Change-Id: Ic2bb12552d6f191478777c21f500794de29c8b0a
Implement OpenSnapshotWriter for non-compressed Virtual A/B. This is
done by adding an OnlineKernelSnapshotWriter class, which forwards all
writes to a dm-snapshot block device.
This also introduces a new ISnapshotWriter class which extends
ICowWriter, and adds features specific to libsnapshot (versus ICowWriter
which is intended only for the new COW format). The OpenSnapshotReader
call has been moved here since the writer retains all the information
needed to create the reader.
To test the new call, vts_libsnapshot_test has been modified to use
OpenSnapshotWriter.
As part of this change, all consumers of libsnapshot must now link to
libsnapshot_cow.
Bug: 168554689
Test: vts_libsnapshot_test
Change-Id: Ieedfadc557833c1e0540922aabc6e95c80266a64
Once COWs are allocated, their size is fixed, and we don't want to
write beyond the end. To make this validation less tedious, the virtual
methods of ICowWriter have been made internal. The user-facing API calls
do validation before calling the internal variants.
Bug: 168554689
Test: cow_writer_test
Change-Id: Ic9ebb5bc4b601180d35d915c47cd9c537bc423fa
adb has been built without -Wnon-virtual-dtor because Soong doesn't turn
the flag for the host targets, and the flag can't coexist with
-Wexit-time-destructors.
However, the new host target linux_bionic_arm64 turns the flag on,
because it inherits most of the cflags from the device config.
That is forcing us to implement the virtual destructor, which isn't
possible due to -Wexit-time-destructors.
Solving the issue by explicitly turning -Wnon-virtual-dtor off so that
it is disabled also for the linux_bionic_arm64 target.
Bug: 159685774
Test: HOST_CROSS_OS=linux_bionic HOST_CROSS_ARCH=arm64 m \
out/soong/host/linux_bionic-arm64/bin/adb
Test: run the adb on an ARM emulator
Change-Id: I655b340e4f199a3a75b7df7cc3fe18832e26a7ae
This simplifies some of the logic and removes the need to pass an
Arch value to functions that should already know about the arch
it is operating on.
Includes fixes for debuggerd/libbacktrace.
Added new unit tests to cover new cases.
Test: All unit tests pass.
Test: Faked unwinder failing to verify debuggerd error messages display
Test: properly in backtrace and tombstone.
Change-Id: I439fcae0695befcfb1cb4c0a786cc74949d33425
This patch changes the variable name `uid` to `userId` where it
denotes an Android user id. This makes it harder to mistake the user id
for an actual UID.
Test: Trivial refactoring no logical changes.
Bug: 163636858
Change-Id: I9e240367bcd30bbfa1816014a7f44bbcb58ede4b
I didn't expect this to ever be hit, but apparently it is, so add
additional logging to track it down.
Bug: 168870781
Test: build
Change-Id: Ia472bc7458a3727e4d407365672f7394426ae515
To make the transition to these new calls easier, give them a very
similar signature to MapUpdateSnapshot. Also, allow them to work in
non-compression mode.
Bug: 168554689
Test: builds
Change-Id: I2eb6b41ba8a294dbde1763aabd7701a80c0789f3
Added missing partitions (/product and /vendor).
APEXes can be put in /system, /system_ext, /product, and /vendor.
Binaries within "flattened" apexes should be marked as executable.
Bug: 167383491
Bug: 169025459
Test: OVERRIDE_TARGET_FLATTEN_APEX=true m vendorimage with a vendor apex
simg2img $OUT/vendor.img vendor.img
debugfs -R 'ls -l /vendor/apex/com.android.foo/bin' vendor.img
=> check if binaries are executable
Change-Id: I67284166d70c98a2a184b47e7052ade8b9039df9
To be consistent with other prebuilt modules within APEXes, the location
for firmware files is switched from /apex/*/firmware to
/apex/*/etc/firmware.
Bug: 167942098
Test: loading vibrator firmware from vibrator apex(sunfish)
Change-Id: Ia05735bf5a54482fded26525fa4a8f795dcfc029
In addition to "firmware_directories", ueventd scans /apex/*/firmware/
directory as well to find firmware files.
Bug: 167942098
Test: loading firmware from vibrator apex successfully.
(sunfish)
Change-Id: I90fc8f9ad843a08b1ca98a2be1b5d22c0c5954a3
android_filesystem_config.h is found since system/core/include is on
the include path for all projects and contains a symlink to the real
android_filesystem_config.h. This is fragile and the below bug seeks
to remove this symlink and have users correctly depend on
libcutils_headers.
Bug: 165825252
Test: build
Change-Id: I1aad6a2e8fbe87635e97b0ac1be3414403353382
There was a bug in SerializedFlushToState::Prune() caused by an access
to a SerializedLogEntry raw pointer as a member of a MinHeapElement,
which was deleted earlier in the function.
Instead of just fixing the order of the access and the deletion, I
sought out to remove the raw pointer entirely. In doing so, I noticed
that the min heap doesn't provide significant benefit, since we'll
only ever have 8 log buffers so scalability is not an issue.
Therefore this change removes the min heap entirely and uses the
existing log_position_ and logs_needed_from_next_position_ members to
keep track of which are the next unread logs.
It also adds a smoke test for SerializedFlushToState::Prune() and
additional CHECK() statements to help prevent future errors.
Bug: 168869299
Test: unit tests
Change-Id: Id4d5fdbaff2fe6dc49c38f01e73f900f84d3696b
Since we can't provide a single device or fd anymore, we need to expose
a CowWriter directly. Additionally, we expose an API for reading
snapshots through the FileDescriptor abstraction.
Bug: 168554689
Test: builds
Change-Id: If7e8adbfe69c2a84d34c63d4b0adff2b3365fd82
block->payload and its size are not valid when it is used to check
against bytes_left due to std::move() performed on its just prior
to the check. Hence check will always fail to detect the case where
received data is more than expected. To detect this condition and
allow error handling with std::move(), remove extra payload variable
and directly use block->payload.
Bug: http://b/168917244
Change-Id: I992bbba9d9a9861a195834f69d62e69b90658210
There are libraries built with the NDK but bundled with the platform
that were otherwise directly including the header path. It's better
for them to include liblog_headers instead.
Test: build
Change-Id: Icca3684ba5b2539789d30de9df840da4ade0dc67
