The following variables in recovery ui were protected by
key_queue_mutex. But the purpose of key_queue_mutex is to protect the
key_queue, which will be changed after we already have a key code. So
getting the key pressed should be orthogonal to the key queue. And
adding a mutex will help to avoid deadlocks in b/135078366.
Variables include:
char key_pressed[KEY_MAX + 1];
int key_last_down;
bool key_long_press;
int key_down_count;
bool enable_reboot;
Bug: 135078366
Test: boot into recovery and press keys
Change-Id: Ie2cfcf1f2fec49b53f8fac97aa9a2c60f15b84f9
As system_ext partition will be linked to system/system_ext, update
the path to system/system_ext.
Bug: 134359158
Test: build
Change-Id: I5a49adfe7e045e24bf6dfbf6990a965b8a5e1ce1
We may fail to memory map the package on 32 bit builds for packages with
2GiB+ size. This cl tries to install the package with fuse when memory map
fails in such cases.
Bug: 127071893
Test: build 32 bit version sailfish, push package and block.map, reboot into recovery with
the corresponding update_package argument.
Change-Id: I5dae4f3e27ccaf8d64ff3657d36f0e75db2330b0
Corresponds to GetMappedPackageAddress.
There is at least one custom device recovery with an extension to copy
the package somewhere else as a backup. Ability to do this was removed
in change ag/955273.
Test: lunch cf_x86_tv-eng; make
Test: atest recovery_unit_test recovery_component_test
Change-Id: I0e61d8a8839c47721874526504ea03b9ca7bafa9
Stop building libimgdiff on device because we are only running
patching there.
Test: unit tests pass
Change-Id: I4225c6b52a536617301a64c405e325799a303b40
Make sure the simulator succeeds executing common non-A/B update
functions.
Bug: 131911365
Test: run unit tests
Change-Id: I520ce6a8827539b88a9e36f9e67eec30d8b586d4
The ShowMenu in StubRecoveryUI used to return initial selection
immediately. This leads to an immediate reboot if we enter recovery
with a stub UI and wait for user's input. This cl changes the behavior
of stub UI's ShowMenu so that it waits keys in loop, and only returns in
the case of interruption or time out.
Bug: 130535924
Test: start recovery with a stub UI, wait for the timeout
Change-Id: I96ad8aad8930756b09101bb98ba5fbe7b53bcb37
The simulator skips executing the operations for dynamic partitions, and
will use the logical images under target_files/IMAGES directly. (Similar
to the targets without DAP enabled)
Bug: 131911365
Test: run update on cuttlefish, run simulator
Change-Id: Id318d97ece4560df9f20dc5cabeb8b2e261bdf9c
Prior to the change, SystemUpdaterSample was built as privileged app.
It must be installed along with the matching privapp-permissions
whitelist change (as noted in updater_sample/README.md), otherwise would
block device booting if installed unintentionally (e.g. with `mm` or
`mmma`).
This CL avoids putting the device in a bad state due to a sample app.
Fixes: 135703777
Test: `lunch aosp_taimen-userdebug && m -j installclean &&
mmma -j bootable/recovery && m -j`
Flash on device and boot successfully.
Test: Run the tests per instructions in updater_sample/README.md.
Change-Id: Ib8b587c77570f05f7db748fad84744fa45016aab
Factor out a new function from ApplyFromSdcard that installs a package
from a local path. Inside this function, we start the fuse and choose the
type of data provider depending on the path string. And similar to the
existing logic, we treat the package as a block map if the path starts
with a '@'.
This is part of the effort to install larger than 2GiB packages on ILP32
devices.
Bug: 127071893
Test: Build a 32 bit sailfish and create a 3GiB OTA package. Sideload
the package, uncrypt and install the package from sdcard.
Change-Id: I328ea34fa530731acbce7554bfc3059313ad6ece
Therefore InstallPackage() doesn't need to worry about the details of a
given Package.
Bug: 127071893
Test: run update from /bin/recovery --update_package=@path, sideload a package
Change-Id: I0caa36785b43924f884ee398e7ea640d7472a92e
Waitkey() is designed to obtain lock "key_queue_mutex" in
the very beginning of function.
int RecoveryUI::WaitKey() {
std::unique_lock<std::mutex> lk(key_queue_mutex);
...
}
However, there's case "key_queue_mutex" being applied again in
waitkey(), thus cause deadlock. There are two reproduce
scenario:
1.Executing "fastboot reboot recovery" in userspace
fastboot
2.Executing "adb reboot fastboot" in recovery os
When entering userspace fastboot/recovery, waitkey()
will wait there for user action. fastboot/adb commands
will trigger ui->interruptkey() to notify the thread
waitkey() in. In the next, waitkey() will move on and call
SetScreenSaveState(), which do LOG(ERROR) in fail case of
brightness set. LOG(ERROR) is designed to print log on
UI. Unfortunately, UI->print() applies lock "key_queue_mutex"
too, so deadlock happen.
Note:
Here is details how lock "key_queue_mutex" applied in
UI->print():
Function Print() call
Function PrintV() call
Function update_screen_locked() call
Function draw_screen_locked() call
Function draw_menu_and_test_buffer_locked() call
Function IsLongPress()
bool RecoveryUI::IsLongPress() {
std::lock_guard<std::mutex> lg(key_queue_mutex);
bool result = key_long_press;
return result;
}
Bug: 135078366
Test: no errors when running "fastboot reboot recovery" in userspace
fastboot & "adb reboot fastboot" in recovery os
Change-Id: Ida6b3c4ba9896a70021373f02a94954f0a60cf31
Signed-off-by: Zhang, GaofengX <gaofengx.zhang@intel.com>
Signed-off-by: Xihua Chen <xihua.chen@intel.com>
This change addresses the comment in [1], which makes the results of
`adb shell getprop foo` and `adb rescue getprop foo` more consistent.
That is, both will return newline-terminated results now.
[1] https://r.android.com/c/platform/bootable/recovery/+/976340/3/minadbd/minadbd_services.cpp#188
Test: Run the following commands on taimen (under rescue mode):
`adb rescue getprop ro.build.fingerprint`
`adb rescue getprop ro.nonexistent`
`adb rescue getprop`
Change-Id: I5af47f8ea4d569b8507e259daef87749c0945f47
It dumps all the allowed properties, similar to `adb shell getprop`.
Bug: 134027350
Test: Run the command under rescue mode.
Change-Id: Ic0864ca0fb51505ec1e4f38af2464591aa576201
Most of these properties are already part of the fingerprint. This CL
allows querying them directly, instead of encouraging users to decode
from fingerprints.
Bug: 134027350
Test: Boot into rescue mode on taimen. Run `adb rescue getprop` with new
props.
Change-Id: Id4667fcaf0e908c391085b22e22c957acd01d9c4
The TargetFile class parses a target-file and provides functions to read
its contents. And the BuildInfo tries to simulate the device with files
on host. Some work it does includes parsing the build properties,
and extracting the image files for partitions specified in the fstab.
Bug: 131911365
Test: unit tests pass, run simulator with cuttlefish, wear devices and from extracted TF.
Change-Id: Iefe4a96d619d2e4b3d038e31480f11a0f9a70afa
Disable building the simulator and its support libraries to resolve the breakage
on mac host targets. Because the simulator is not intended to use on mac anyway.
Bug: 134047992
Test: mma
Change-Id: I488ab50cab1282f03250010b5334f1895d44f98b
Implement the simulator runtime and build the updater simulator as a host
executable. The code to parse the target-files and mocks the block devices
will be submitted in the follow-up.
Bug: 131911365
Test: unit tests pass
Change-Id: Ib1ba939aec8333ca68a45139514d772ad7a27ad8
