The syntax of init's mount command changed in April 2008 but
recovery's init.rc was never updated, so recovery's /tmp has been on
the root fs all this time. Fix.
Also add /system/bin to the PATH in recovery, which is handy for
debugging.
Change-Id: I39f7ae435a8ce3bad691e4b7c307db0bd8de1302
Otherwise everything is left running in the kernel domain when
booting recovery.
Change-Id: Ie3d86547d5be0b68dd1875a97afe1e00fc3e4da1
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
minzip had some features that were used when reading APKs, but APK
handling now uses libziparchive instead of minzip. Remove these
unused functions.
Change-Id: Iead89209a716bfe9e3d339bf85b3e97e33a41f35
On n8013 at least, this event is sometimes/often sent on touch start which
causes touches to be randomly rejected and throws you back to the start screen.
Touch releases still work fine for me, since it sends ABS_MT_TRACKING_ID == -1
for that purpose.
Adds a flag TW_IGNORE_MAJOR_AXIS_0 to configure this behavior.
Change-Id: I14bb036a2c2d4ee676b642c3ab297c55bb6c2084
These were attempts to write partitions "conservatively" in hopes of
fixing the problems with writing the radio partition on Nexus 4. They
didn't work (a kernel patch was needed), but got left in. They make
writing of partitions unnecessarily slow (ie, we really shouldn't need
to sync() after every 4kb). Roll back most of them, but leave the
verification read-back in.
Change-Id: I94badc0979e88816c5aa0485f6316c02be69173c
If the kernel does not use SELinux security then the ramdisk
contents will not have contexts set on the files, however we may
still have everything we need in recovery if the kernel supports
EXT4 security labels, so we try to check /cache/recovery if it
exists first.
Change-Id: I96a4ae16418e07b0a2e83cd56e19893898d0b394
- libmincrypt renamed to libmincrypttwrp that is an static library
- libjpegtwrp does not exist
- libfs_mgrtwrp is for JB decryption methods
This fixes making full builds when TW_INCLUDE_CRYPTO_SAMSUNG := true and TW_INCLUDE_CRYPTO := true are set.
Somehow typing make recoveryimage doesnt push the mentioned issue.
Change-Id: I7cad5db4f51152a1a8209e619b188ca88d7c74d1
example userdata line (as needed by MotoX):
/data f2fs /dev/block/platform/msm_sdcc.1/by-name/userdata flags=fsflags="inline_xattr"
Code for parsing flags/options originally based on AOSP fs_mgr sources.
Change-Id: I5fb2b5d5cdd08137e6bf71f0085a3f8aebd889a8
This assumes that the metadata is correctly defined in fstab.
Which apparently some devices don't do.
Bug: 8766487
Bug: 12112624
Change-Id: I1b14b9d4c888e9348527984be3dce04bdd9f4de0
* Add support for variables with persist="1", which will cause
them saved into settings file and reloaded on next boot.
* This is handy for theme makers, who can use this to
implement "theme settings".
Signed-off-by: Vojtech Bocek <vbocek@gmail.com>
Change-Id: I6415979011b38cbb15dec9909c445b02da6eadc0
If the button has a conditional, the image render function will
return a 0 which indicated that an image was present even if it
was not if the condition was not being met during boot. When the
condition is met, mButtonImg was not set to NULL causing a render
request failure.
Change-Id: I32be0ddcdc1bf7ad22440d66018ef09484280afa
In order to support multi-stage recovery packages, we add the
set_stage() and get_stage() functions, which store a short string
somewhere it can be accessed across invocations of recovery. We also
add reboot_now() which updater can invoke to immediately reboot the
device, without doing normal recovery cleanup. (It can also choose
whether to boot off the boot or recovery partition.)
If the stage string is of the form "#/#", recovery's UI will be
augmented with a simple indicator of what stage you're in, so it
doesn't look like a reboot loop.
Change-Id: I62f7ff0bc802b549c9bcf3cc154a6bad99f94603
A system/core change made in Mar 26 2012 6ebf12f "init: Change umask
of forked processes to 077" changed the default umask of services
forked from init.
Because recovery is forked from init, it has a umask of 077. Therefore
when update-binary is forked from recovery, it too has a umask of 077.
This umask is overly restrictive and can cause problems for scripts
relying on minzip to extract binaries directly into the target
filesystem. Any directories updated by minzip will have their
permissions reset to r-x------ and created files will have similarly
restrictive permissions.
As it seems unlikely this security measure was intended to have this
side effect on legacy sideloads that do not have chmods to repair
the damage done by minzip, this change reverts the umask to 022 in
the fork made for update-binary.
Change-Id: Ib1a3fc83aa4ecc7480b5d0c00f3c7d0d040d4887
Because backgroundIcon[] is not initialized by the ScreenRecoveryUI
constructor, it should be initialized explicitly to NULL in Init().
If it is not initialized, ScreenRecoveryUI::SetBackground() can fail
for the NONE icon because the NULL test can fail and junk will be
dereferenced.
Change-Id: I4e3738d2e241ed90df43c984fb41e0072933f50a
* commit '40dfc474c457f5be24e204e6a0c657c97acb8f3c':
Enable incremental builder to find files that moved, and try to process them via patch + rename, instead of delete + add.
Doug Zongker
Kra1o5
Nick Kralevich
Stephen Smalley
Doug Zongker
Ibrahim Awwal
bigbiff bigbiff
Michael Runge
Dees Troy
OliverG96
Hashcode
Koushik Dutta
Vojtech Bocek
synergy dev
Alistair Strachan