| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
| |
In A/B, the device check isn´t done by the updater script, but using
metadata for it.
Let´s search if the device codename/assert is included in the string.
Change-Id: Ie856ac699aaa83de2b364bc85a510a037d36edf9
Signed-off-by: Hernán Castañón <herna@paranoidandroid.co>
Signed-off-by: theimpulson <aayushgupta219@gmail.com>
|
|
|
|
|
|
| |
* Avoid dangling pointer. Instead of pointing to fstab_rec create copy.
Change-Id: I41685dd9a2e5cc8ba1c49f63115368868279e6c4
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* toybox's `mount` does not support multiple entries like we do,
so if we can detect the filesystem of an fstab enrtry print that
and only that one to /etc/fstab, so that mounting via toybox has
a better chance of succeding.
* as a bonus, this patch also gets rid of duplicates in /etc/fstab
caused by the fact that ReadDefaultFstab() combines entries from
DT and from recovery.fstab
Change-Id: Iec4ab38044054555d2a33da6f5d53de7716e7bee
|
|
|
|
|
|
|
|
|
|
|
| |
* Too low of a value causes out of memory exceptions when
sideloading big files (compared to the device's available ram)
* 400MB was determined by increasingly trying different values,
100MB at a time, on a 2GB device when sideloading 1.9GB
of data.
Change-Id: I8faac2deb7b1c2ddbd9dcb21914e34d6cd672575
|
|
|
|
|
|
|
|
|
| |
When running on an A/B device, split the top version line horizontally
to show the Android version on the left and the boot slot on the right.
When not running on A/B device, center the Android version.
Change-Id: Ica18af5e46dadf1088fdfd6d23d96db090195a6a
|
|
|
|
| |
Change-Id: I5a32a59691dea5fa2aa867e6594cec15b1b24ccf
|
|
|
|
|
|
|
| |
Cast device parameter to void to shut up the compiler.
Change-Id: Ifc6496345d02e83f6bacb8308c4714e1f8b46bf3
Signed-off-by: shagbag913 <sh4gbag913@gmail.com>
|
|
|
|
|
|
|
| |
* Let's show our logo during install
and dialog to ask to continue unverified install
Change-Id: Idd362a25efa185a2c46c764557758f3d15d60b39
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* If userdata is initially unencrypted and user performs a wipe,
the option to install from emulated storage would still be
shown as available.
* This patch sets userdata_mountable = false after wiping data.
* That alone though is not enough, because if we reboot straight
into recovery after wiping data, userdata_mountable will be true
since "mounting" would indeed succeed, but no media/0 folder
would be available. That is only created after booting system.
To solve this we just add a check for mountability via the
volume manager, which fails if it cannot bind-mount media/0 to
/storage/emulated/0.
Change-Id: Ia4a6f12db89bfbf21bbec38cc2f7b9598c8157a5
|
|
|
|
|
|
|
|
|
|
|
| |
* In volume manager, check if new volumes are mountable.
* Check volumes for mountable for inclusion into update list.
* Erase unmountable volumes from volumes vector for consistency with
the item array.
Change-Id: I89ff6cc05a93afffe5e46b24d70fc368bccaf020
|
|\
| |
| |
| |
| |
| |
| |
| |
| |
| | |
* aosp/pie-gsi:
uncrypt: write permission for f2fs_pin_file
uncrypt: fix f2fs ioctl argument for pin_file
f2fs: support f2fs by setting unmovable bit for package file
Revert "f2fs: support f2fs by setting unmovable bit for package file"
Change-Id: I458ebf8f58e5dbe0f43685909d5e2c85018be9c4
|
| |\
| | |
| | |
| | | |
pi-platform-release am: 94baa21254" into pie-gsi
|
| |/|
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
pi-platform-release
am: 94baa21254
Change-Id: I4877ebcdca81b329cf7b61d4e61c81afc696f24a
|
| | |\
| | | |
| | | |
| | | |
| | | |
| | | | |
pi-platform-release
Change-Id: I02180574c059ce8af1f031195ee497540a82cd65
|
| |\ \ \
| | | | |
| | | | |
| | | | | |
pi-platform-release am: 96f7786965" into pie-gsi
|
| |/| | |
| | |/ /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
pi-platform-release
am: 96f7786965
Change-Id: I785deaa02b405efa21abf5f8713a1bccc8fd0829
|
| | |\ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
pi-platform-release
Change-Id: I18ffadbdab03199e33f4476135b95da658880541
|
| |\ \ \ \ |
|
| |/ / / /
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
We need a write permission to set a flag in the file.
Bug: None
Test: pre-submit
Change-Id: I4896ecbe0fc04374e01d006b1c8acdb932e5d16d
Merged-In: I4896ecbe0fc04374e01d006b1c8acdb932e5d16d
Signed-off-by: Jaegeuk Kim <jaegeuk@google.com>
(cherry picked from commit 051b9d87e2bf1550ab6495493b9a2cfab2aa54f1)
|
| |\ \ \ \ |
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
This patch fixes missing f2fs ioctl call.
Change-Id: Id840b76b9d5c580041aaee4501ac8e69fc3fb818
Signed-off-by: Jaegeuk Kim <jaegeuk@google.com>
|
| |\| | | |
| | | | | |
| | | | | |
| | | | | | |
pie-gsi
|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | | |
This enables to use uncrypt for f2fs update-on-reboot.
It requires kernel patch named:
"f2fs: add an ioctl to disable GC for specific file"
If any operation fails during uncrypt, please delete package file as soon as
possible, and create the file again to move forward. IOWs, don't leave the
package file for a long time.
Bug: 70309376
Bug: 30170612
Change-Id: I3b4233e7da756f107be35364521699deaf2e7139
Signed-off-by: Jaegeuk Kim <jaegeuk@google.com>
|
| |\| | | |
| | |/ / /
| |/| | |
| | | | | |
file"" into pie-gsi
|
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* Replaced with upstream aosp/master version
* With this change flashing fails if either one of the following cases are met:
1) Kernel f2fs stack doesn't support the ioctl ( errno isn't checked for ENOTSUP )
2) If the device has multiple userdata entries in fstab ( f2fs and ext4 in this order )
and userdata is formatted as ext4, recovery will attempt to send the ioctl
without checking if it's supported ( same as 1, errno isn't checked for ENOTSUP )
3) The ioctl, when supported, needs an extra argument ( __u32 set )
This reverts commit 848f227eea524a48208bdd4181f90fb63a31ed2b.
Change-Id: I5871d3da4594f7da44546a2979e72dd87434f821
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
No longer will we get bad reviews on CWM recovery for its lack
of rainbows! Originally design by gweedo767, this patch brings
back the wrap count mechanism from v2.0.2.4 to toggle Rainbow
Mode for the user. In the original design, every character has
a different color, but that method requires modifying graphics.c
which will cause breakage in devices with custom graphics.c.
This Rainbow Mode has a different color for each line, instead.
The wrap_count has been set to 5 instead of the previous 3 so
the toggle won't occur as often, if on accident.
(port to CM12.1 by cyanogen)
Change-Id: Ie9f6f1b6ed94ee69d4d8d9fe27f9a5417cc3f212
improve Rainbow Mode
* Require 5 consecutive menu "wraps" in the same direction to help
prevent accidental toggling
* Align colors into stripes (i.e. columns)
a) less prone to induce dizziness in some people when they change
b) better resembles a rainbow
* Move the stripes to the right when the selection moves up and
move the stripes to the left when the selection moves down
[yodevil]
* Offset different rows by 1 color, so that colorss are diagonally
aligned instead of vertically.
* Refractor SelectMenu code to avoid clogging it and preserve
readability
Change-Id: I3feae173b22f5703c554ca33e634881749ff54cf
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* We may use a custom offset to:
a) preserve data that oem wrote to the first bytes of misc
b) skip recovery flags written by the bootloader (e.g. --wipe_data)
For case a) one should set the offset 'x' to be at least greater than
the size of bootloader_message struct (2048 bytes). If this is the case,
then we zero out bytes x ~ x + 2047
For case b) one should set the offset to be strictly smaller than
the size of bootloader_message struct. If this is the case, then we
zero out bytes 0 ~ 2047.
This allows to clear any additional flag set by the bootloader,
that would otherwise be forgotten in misc.
This also guarantees that we do not involountarily wipe any data that
the oem may have written starting at byte 2048 (coff coff LG)
Change-Id: I2d4e0702a2d8cbbef6274a87ce9499b0f69310dd
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* Fixes touch on some devices like chiron/sagit
* Some other devices (e.g. most MTK) need this because their kernel driver
implementations do not perform as expected (initial unblank is only
happening when altering the panel brightness x -> 0 -> y). In regular
Android, that "unblanking" is done on boot when the lights HAL loads and
sets the initial brightness.
Change-Id: I2fffac508d09d07355a7a7f087805b0dceb5f97e
|
|\ \ \ \
| | |_|/
| |/| |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Android 9.0.0 Release 37 (PQ3A.190505.002)
* tag 'android-9.0.0_r37':
Import translations. DO NOT MERGE
Change-Id: Id5f82253387da4dde498a1498e91d6500b02b0e8
|
| |\ \ \
| | | | |
| | | | |
| | | | |
| | | | |
| | | | | |
pi-qpr3-release
Change-Id: I341bfa4c39eee9f4e21387e2e6f0af2c75d19364
|
| |/ / /
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Auto-generated-cl: translation import
Bug: 64712476
Change-Id: Ie357c8e7a6445223a5a657cd0e96ded4689837f6
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: Iad020439a007a28351719774b641038b076601a2
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* The text array will be handled by always writing to the
last row, and will be moved up on a new line.
* Update the screen at each print.
* N.B. This does not show text on OTA updates,
but rather only on updates launched from the menu.
Change-Id: Ia62c361da17c5e24af508c29a04f045e81306e1c
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* the first fstab entry for data will be used.
* this also prevents strcmp to be called on a null parameter,
which ended up crashing some devices
Change-Id: I1df70d9df462ab30c857e834152b75d083d77652
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: Ib1928acfb598803b4dbc676d9f6d666b0f7e2881
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* Also avoid generating a refresh event when mounting
a storage volume to browse it, it would end up being
ignored anyways.
Thanks to @tdm
Change-Id: Idabce4e90b484f34f83d310abdb345778cfc3bf0
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: If783711ecd0d20728ec27ebc572b57e6d9bee760
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I42bb91a83c44a4da92c1e85bcea9849e046253dd
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: Ieedfc3a56cd341583700bfad31993879eb312ba6
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: If6b5b91cdcc36003a17ed24e58b820d44a810070
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I45775c1ebbd4a84e0d67d0f573b9aae97ab55a8b
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I4c7623004fc45e936c73bbcf0f818c5fc4e11bea
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I0ec2c9f3c44b3aa69e07aa97e45a0ee3214ffd05
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I9d94b611ab9a873b6b57a6d22bf78de131e3c8e5
|
| | | |
| | | |
| | | |
| | | | |
Change-Id: I8684e188577fa16073723e0093e221b767a3826d
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Backported from upstream I7891bb84427ec734a21a872036629b95ab3fb13c
"Add fastboot mode to recovery"
Change-Id: I252f99f0e44f6be53c5b7577c19c8c1f0f616ad8
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
In the old code, the recovery only enumerated the input devices at the
startup, and read the input events from these devices.
So if a USB input device is probed after the recovery startup, then the
recovery can't read the events from this device.
This patch use inotify to monitor /dev/input for new added input
device, then support input device hotplug in recovery mode.
Bug: 111847510
Test: can use USB keyboard hotplugged in recovery mode
Change-Id: I7e7dcbd619d3c66a2f40a43418f5dac6a50c859e
Signed-off-by: Liu Shuo A <shuo.a.liu@intel.com>
Signed-off-by: Ming Tan <ming.tan@intel.com>
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
The input_cb in ev_init() should be shared among the detected input
devices.
Test: Boot into recovery. Check key inputs.
Change-Id: I0315f6a23625bacedd333fefacbf4ac6797bc1fd
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
* Multiple fstab lines (supported in android) cause recovery to fail
to mount partitions if the fs type is not the same as the first
fstab entry. So when we attempt to find an fstab entry that matches
a path for an f2fs, ext4 or vfat type, check it against blkid's
determination of what filesystem type it is. If there is a discrepancy,
query fs_mgr for the next possible entry that matches that path until
either we find one that is good, or run out of fstab entries.
* Also attempt to autodetect the filesystem type for mounting from
update.zips.
Change-Id: Ib6f4535dd88ef714ae1ca6fb0ffae1c7dac0f7ce
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
This allows A/B devices to install legacy (non-payload) style update.zip
packages like our addonsu or gapps.
Change-Id: I907a92732470b947de007dde11fb71db64d94c9b
|