This gives us finer control over the partitions to wipe on the host
side.
Bug: 127492427
Test: unit tests pass, install a wipe package on sailfish
Change-Id: I612f8bac743a310f28e365b490ef388b278cfccb
Found these FDs unintentionally propagated into forked processes (e.g.
updater) during my local debugging. Use O_CLOEXEC to avoid that.
Test: Dump open FDs in updater process.
Test: Sideload a package on taimen (which uses framebuffer).
Test: Run graphics test on taimen.
Change-Id: I1fb65dd4a103c47d28b15973def9e9359c251ca3
Also clean up the comments regarding updater/recovery protocol.
Test: Sideload a package on taimen. Check that /tmp/recovery.log
contains the log from update_engine_sideload.
Test: Factory reset on taimen. Check that log file has the info from
mke2fs.
Change-Id: If560269ce7007769fc85a63bd228813bdd75bebd
fastbootd looks too much like recovery, even if you're carefully reading
the menu. It's not obvious the device is in a flashing mode, and it's
too tempting to reboot or unplug the device in this state.
As a first step, this patch adds a big red "fastbootd" logo so it's less
obviously in recovery mode.
Bug: 120429730
Test: manual test
Change-Id: I73359f1fdfdc0b1694993f760fe7f35c5713b24e
This moves actually marking the slot as successful to a later point
so that on devices with checkpointing enabled we can still roll back to
the previous version if we fail to boot to the point that the checkpoint
is marked as successful.
Test: When taking an update on a checkpoint enabled device, it
defers marking the slot as successful instead of directly
marking it. Visible in logs.
Bug: 123260515
Change-Id: I7ed3595c1b0904ddbfe20d1cad4f69ecbf1ea351
The intended string constructor is supposed be
basic_string(size_type count, CharT ch). But the parameter is
accidentally reversed when calling the constructor in install_test.
Test: A failed unit test pass
Change-Id: Id9765bfa7d2368ff0d7fbeea45c9c8357864e060
Also add libfstab dependencies where needed. Previously the
`typedef struct FstabEntry Volume;` line served to both define a
`struct FstabEntry` as well as alias Volume to it. With the new
namespace for android::fs_mgr::FstabEntry, `struct FstabEntry` isn't
compatible anymore, so we need to alias Volume to the real
android::fs_mgr::FstabEntry.
In doing so, we need to include <fstab/fstab.h> and this requires
libfstab as a library, which a few modules did not have before.
Test: treehugger
Change-Id: I655209a0efb304b3e0568db0748bd5cf7cecbdb7
This reverts commit 9ce1d14ef6.
Reason for revert: AOSP is fixed with new vendor image
Change-Id: Ie5a9748acdae22a2b9862cb2ecedda7031f77264
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
This will remove spurious error messages in logcat such as:
08-02 00:27:21.580 600 600 E /system/bin/recovery-persist: Failed to
unlink /data/misc/recovery/last_install: No such file or directory
since the install file doesn't exist in the first place.
Bug: None
Test: m
Change-Id: Ifaa95729c50efae4e641286dfbe1718aceb5f50a
Because set_sched_policy is moved into libprocessgroup an additional
dependency is requred for recovery_component_test to build.
Exempt-From-Owner-Approval: janitorial
Bug: 111307099
Test: builds, boots
Merged-In: I7cf75e473ee1e2837940606c71d15be26db0c3f2
Change-Id: I7cf75e473ee1e2837940606c71d15be26db0c3f2
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Switch to 64 bit integers since the size of the entire src/tgt images may
not fit in size_t of ILP32.
There are other theoretical overflow cases in memory allocation and I/O
functions. However, they reside within a single transfer command and are
less likely to happen. I will evaluate and address them in separate
cls.
Test: unit tests pass
Bug: 122461124
Change-Id: Ib719ee695920877458fcfaa25c6ac058a5bbabf2
Remount will instead use overlayfs. If the device is not configured
to use overlayfs or if the device does not support dynamic partitions
but has chosed to setup an ext4-dedupe filesystem, e2fsck_static must
then be included in recovery.
Bug: 122115286
Test: fastboot flashall
Change-Id: I931282dbc67ff4ef53291bcbc5a078b500e5e807
Test: sideload full OTA on cuttlefish
Test: sideload incremental OTA on cuttlefish (that grows
system, shrinks vendor, and move vendor to group foo)
Test: verify that /cache/recovery/cc46ebfd04058569d0c6c1431c6af6c1328458e4
exists (sha1sum of "system")
Bug: 111801737
Change-Id: Ibdf6565bc1b60f3665c01739b4c95a85f0261ae5
When applying an OTA package onto the device in OTA mode,
if the recovery logs haven't been viewed, there is a chance
that /cache/recovery does not exist. Then, stash creation will
fail. Create stash directories recursively to avoid this error.
Test: without /cache/recovery, sideload the OTA on cuttlefish
Change-Id: I5cc01a067d866476a3594e795dcb5b15649e817b
The caller of daemon_service_to_fd() in core adb has switched to
std::string_view in [1]. The mismatch breaks the sideload service, as it
picks up the wrong daemon_service_to_fd() when serving sideload.
[1] https://android-review.googlesource.com/c/platform/system/core/+/850392
Bug: 122171762
Test: `adb sideload` on taimen.
Change-Id: Ie828400768523c35c5576e2c029e38fc0ad0aff9
It also reduces the space cost for devices using f2fs (e.g. crosshatch).
/sbin/mkfs.f2fs 722560 => /system/bin/make_f2fs 49568
/sbin/sload.f2fs 1182456 => /system/bin/sload_f2fs 150032
Test: Build and boot recovery on crosshatch. Factory reset.
Test: Install a non-A/B OTA package that formats a f2fs partition.
Change-Id: Ibe70c8d91a1d07e1c78ff9eac19b1f7955800161
Test: Build and boot into recovery on marlin. Factory reset.
Test: Build and install a non-A/B OTA that calls format.
Change-Id: I72416e775e237fc15ca5eff1036175a9eef43b76
Convert code to use C++ Fstab struct and C++ std::strings.
Bug: 62292478
Bug: 118634720
Test: boots
Change-Id: Ibdc1df5831bc885d7c1574419f41af026e49a137
