Original discussion here: http://review.cyanogenmod.org/45559

Patchset 2: really remove the option
Patchset 3: fix options that I broke on ps2

Signed-off-by: Michael Bestas <mikeioannina@gmail.com>

Change-Id: I54877a24444c24a296c8c044efd977f30cc66de4

Also:
Cleanup leftovers from previous poweroff option
Always ensure a clean reboot (Thanks to PhilZ for his implementation)

Bootloader option defaults to "bootloader" (fastboot mode), it can be overriden
by defining "ro.bootloader.mode=download" prop for samsung download mode
We leave reboot to bootloader option always enabled, that way we can
override the get_filtered_menu_selection bug that breaks the order
of the options when NULLifying menu options.

Signed-off-by: Michael Bestas <mikeioannina@gmail.com>

Change-Id: I6d674b37ab41ab342b105993719277360b4707e4

Change-Id: Ic5716e5f2c311406bc6384e63fb0f217113e6eeb

the releasetools want this now

Change-Id: Ibc3cdbc8f5f8022bf51479a4702d1d0d043f70a9

v1 is the old recovery.fstab format that everybody is still
using.
Devices using v2 should set RECOVERY_FSTAB_VERSION=2 in
their board files

Change-Id: Ie7221f110ecc594f6f92973d2c27da10f62b6431

Original discussion here: http://review.cyanogenmod.org/45559

Patchset 2: really remove the option
Patchset 3: fix options that I broke on ps2

Signed-off-by: Michael Bestas <mikeioannina@gmail.com>

Change-Id: I54877a24444c24a296c8c044efd977f30cc66de4

Also:
Cleanup leftovers from previous poweroff option
Always ensure a clean reboot (Thanks to PhilZ for his implementation)

Bootloader option defaults to "bootloader" (fastboot mode), it can be overriden
by defining "ro.bootloader.mode=download" prop for samsung download mode
We leave reboot to bootloader option always enabled, that way we can
override the get_filtered_menu_selection bug that breaks the order
of the options when NULLifying menu options.

Signed-off-by: Michael Bestas <mikeioannina@gmail.com>

Change-Id: I6d674b37ab41ab342b105993719277360b4707e4

Change-Id: I86a280088ee8767e02ba51b72f4d011cf4658f6f

Change-Id: I86a280088ee8767e02ba51b72f4d011cf4658f6f

Change-Id: Idce9a6d4ca18ddc9b49029024c26bc114f6d3c15

Change-Id: Iab792171fdbe62fe85ac3084f49cbd38cb172fba

Change-Id: Ibb9be9e764d0ccb842a26a0795a0bb6cb6113b58

Change-Id: Ib5e3df51fa7a383f3699968e899e06d579480564

Change-Id: Ie7e8bfea57bed207c3c2d57bf3cd23daf6cd4c20

See http://review.cyanogenmod.org/#/c/46642/ for why this is necessary.

Change-Id: Ib97484755569fcafc062d3c39b2841581987ca72

Change-Id: Ie68daba5e014dff0c4998da85576f35abe8ad270

Add an option to verifier_test to load keys from a file, the way the
recovery does.

Conflicts:
	verifier_test.c

Change-Id: Icba0e391164f2c1a9fefeab4b0bcb878e91d17b4

Extends the last_log mechanism to save logs from the last six
invocations of recovery, so that we're more likely to have useful logs
even if the device has repeatedly booted into recovery.

Conflicts:
	recovery.c

Change-Id: I08ae7a09553ada45f9e0733fe1e55e5a22efd9f9

This reverts commit 1a6c91f.

Change-Id: I5fb1105bfb640f0552ddcc4a99b6f98487a1c946

Voldmanaged volumes use sysfs path instead of device
path which is not supported by recovery, so a separate
entries for /sdcard and /external_sd are required.

We ignore these volume so we don't show the menu for
"mount /storage/sdcard0", etc.

Change-Id: I6bf26e0ad3fb59579c986cf5537f9e70448891e9

This commit allows the wipe data/factory reset option to recognize the
.one_confirm or .no_confirm files in the clockworkmod folder.

Signed-off-by: Michael Bestas <mikeioannina@gmail.com>

Change-Id: Ib0f82b4448e3b9df95a5e59a49428f3e9da58578

Taken from PhilZ Touch recovery

Signed-off-by: Michael Bestas <mikeioannina@gmail.com>

Change-Id: I079617fc32f8ad0aea754a60898cd4f9cc482524

…lesystem corruption if some data is not sync when mount sdcard

Change-Id: Ib3ad32842fcbf3b058c986df6901c7744bd5cf1f

Change-Id: I762269d696a194889e74c101b9338920e924ebb5

Change-Id: I00556341b746f186f424ebe46c94fd02f10e6536

Change-Id: I7e70d0e2e367f703fe420d4300eb5fecdc247099

Change-Id: I73b4112983b7b118d738eb224b32f4f0beb5d147

--wipe_data issues via /cache/recovery/command will now completely format data.

All other manual wipe commands will preserve data on /data/media.
There is another mounts/storage option that will do a true data format.

Change-Id: Ie8ecd2b0e14c3bb1d8a404ea868cdf703455d2ab

…rage" into cm-10.2

…to cm-10.2

…e data." into cm-10.2

…card, silent failure while formatting device (every nandroid restore or wipe data)

Change-Id: I3ee2f891899be8e4fe0d722e64357dc5e7bc7428

 * Recovery is currently calling setup_adbd when the process starts to
   fetch the keys from the data partition and (re)start adbd.
 * This of course has the heinous side effect of making it
   impossible to use gdbserver to debug recovery, making me a sad panda.
 * Make setup_adbd a pseudo-binary, and run it as a oneshot service
   instead.

Change-Id: Ib4b9e342564bb60bdc2df4cee91566f31c7db8f1

 * Vold has been refactored to be friendly to an environment such as
   recovery. Using Vold will eliminate many hacks and add previously
   unavailable features.
 * Link with libminivold and create a pseudo-binary to launch as a
   service in recovery.
 * Libvoldclient is an event-driven client for controlling the daemon.
 * Handles device hotplug and automatic mounting, if desired.
 * Manages USB storage
 * Can check and format volumes using any supported filesystem
 * Dynamically generate all storage paths
 * Dynamically refresh menus affected by storage changes
 * Fixed wall-of-compile-warnings and const-correctness in extendedcommands.c
 * Misc cleanups
 * Say goodbye to "apply /sdcard/update.zip". You have served us well.

Change-Id: I13a3d72493f8fce7c19f192f9e94ad22cb60a2d2

 * This causes problems with some hosts. Only enable the composition if
   the user actually turns UMS on.

Change-Id: I18b109a3416d55ef711d346685d46a93851da332

 * Do it for hotplug and on-demand only.

Change-Id: I9c0ca96203a8ff1f16fa42e31f426cd25bdaba1d

 * Our custom ext4 support for sdcards needs it.

Change-Id: Ie1c22c1d6a3255cbe3a7ff3c072243f4a5461a17

Nexus 4 has flash errors that manifest during large writes (eg, of the
radio partition).  Writes of some blocks seem to be dropped silently,
without any errors being returned to the user level.

Make two changes to the partition-writing code:

- break it up into 1MB writes instead of writing partitions with a
  single fwrite() call.  Pause for 50ms in between every chunk.

- read the partition back after writing and verify that we read what
  we wrote.  Drop caches before reading so we (hopefully) are reading
  off the actual flash and not some cache.

Neither of these should be necessary.

Bug: 9602014

Change-Id: Ice2e24dd4c11f1a57968277b5eb1468c772f6f63

Write and verify partitions using write(2) and read(2) rather than the
stdio functions.  Read and write in 4kb blocks.  When writing, fsync()
every 1MB.

Bug: 9602014
Change-Id: Ie98ce38e857786fc0f4ebf36bb5ffc93b41bc96f

Another speculative attempt to get everything we write actually stored
to the device.

Change-Id: Icf40b0741b4c535e55ea34848073a97d90dc0e70

 * Systems are getting quite large nowadays, and performing backups
   without using compression is wildly inefficient. A good compression
   algorithm also gives us a big speedup on backup and restore times since
   the process is entirely I/O bound.
 * Bring in pigz (via libminipigz), which is a multithreaded version of
   gzip and pipe tar backups and restores thru it.
 * Autodetect compressed images where appropriate, and add a menu option
   to select the compressed format by default.
 * Also fixed various trivial warnings and const-correctness issues.

Comparison on my Flo device right now:

Uncompressed:
  1868512   2013-08-24.07.10.08

Compressed:
  1149300   2013-08-24.06.51.14

Change-Id: I55025c993662142f5b647e600daf6885f9be6667

Self-nominating for the most impactful commit CM has seen
since 2009.

Change-Id: I450167573b7cf6db1a92227aef7ff3ed95856dd5

Otherwise recovery will restart itself (gets killed by vold for
leaving a file open) when reboot is selected.

Change-Id: I915c3082bd99ca994c3a4cbe5ba913866051bc0e

 * Fixing warnings, const-correctness
 * Split the multi-call list out into recovery_cmds.h

 * This is part of an effort to improve security in recovery mode.

Change-Id: I1e913345c94537225639183434ea4803348ad335

Otherwise when formatting /sdcard/.android_secure, it will be
matched to /sdcard and vold will format the entire /sdcard.

Change-Id: Ifc26121f982b391b4ccf582b6162ab76feb72cae

We simply need the space to fill both strings and a final null.

a char pointer is a lot larger than a char ;)

Change-Id: I3472e1a9978fb2f6d439d4af9c072a98cec0cd19

When using legacy path, recovery will not be able to mount the
correct device. Fix by resolving the symlink before installing.

Change-Id: Ibb268908458c49de09204a1a0ba3143e83ccc0c4

Backing this out for now to prevent a tragic recovery loop
where it'll keep restarting to flash from /cache/recovery/command
and crash from signature verification bug. The loop can only
be ended by removing the command file via adb.

This reverts commit ca02861.

Change-Id: Iea29ee04871e136c680df699f5d19570aba44113

…o cm-10.2

 * There are issues with conflicting symbols in OpenSSL that cause some
   nasty problems. Let's just build minivold as a standalone binary so
   we avoid anything like this in the future as well.

Change-Id: I691a403bb377171989cba4ff74a515050862be40

This reverts commit ea09a67.

Change-Id: I172ec887dd52e373d249d61b22eccaaae03dfb75

Add support for automatically checking and making loki images
for locked bootloader bypass on loki supported devices. Loki
support is enabled with ENABLE_LOKI_RECOVERY.

Change-Id: I81fe267d58f0dda18a790e25df223f011dca3e37

 * It is ancient.

Change-Id: Ie2e31f9e3c8a2753dccb1bd42425cd5b0ce7780b

Change-Id: Ice466de2bddd6050335978d70f31831f4d428576

Change-Id: I7b083a0a083fae487d4b48bef5e79371b3173f7a

This allows devices like Moto X, which have f2fs formated
data partition to be used with CWM recovery. Format data
works as enteded as well as backup/restore.

Change-Id: I8b56252cc462c9e9ff2824bc817ea00ddd4ccacc

Change-Id: I5637e796e4dc65ae527b0b1984b295d69ccc0707

Change-Id: I14af245a05422d25364236da86a96d48b7216b46

Change-Id: Iaf86e006ff4a67e72fe6c8141068ec2e36360671

Change-Id: I152ba81c03a9386897a4ad28849ab28ac35e7c0f

Remove roots.c dead code
New lines
Misc formatting

Change-Id: I445fe240b3e5d1fff4c4dbce25e286cec66f69fc

Patchset 3 - Indigo has been removed from the rainbow.

(Rainbow Mode - Easter Egg)

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.

Change-Id: Ie9f6f1b6ed94ee69d4d8d9fe27f9a5417cc3f212

Change-Id: Id117adcc2ab96a32ca61937863f260d760496e19

Change-Id: Ia1efbe35ac5f26f2a6d03fa505ae1982756541a7

Change-Id: I8b35a010e5fdb36cec991c56d5cd0def72865308

Change-Id: If31be58f4ecb9ab95cca7af0cde5ce32f25c8573

Change-Id: I4ac989ed7efcfd8988057813a94e153d15aad3a3

Array max is incorrectly calculated, can lead to missed events for ids > MAX_DEVICES

Change-Id: I5bbd6bd7239680f20880ac6fdf7744e2e47b39ac

Change-Id: Ia6a077ea71d9f03e90e039ce28eebe26d28a221f

all the extra no options.

Change-Id: I1362815b18f4f1aeecb7426689f38bf8f1f1f008

Enabled via BOARD_RECOVERY_SWIPE := true in BoardConfig.mk.
Does not interfere with hardware keys, they continue to work.

Patchset 1) Initial set
Patchset 2) change Board Config variable
            don't add if BOARD_TOUCH_RECOVERY set
Patchset 3) more ifdef's for all!
Patchset 4) moved instructions
            even more stuff ifdef'ed
            no more abs name issues
Patchset 5) clean up commit message
Patchset 6) Move things to swipe.c to clean up ui.c
Patchset 7) white space

Change-Id: Id9844c38c31bd1985dd78a6d1dd81e71aa1f87f3

Change-Id: I576c054dc5ab6a080e499f26c8cb2a66ecb8ae42

- do not show partition sdcard entries for vold managed volumes until we have a valid blk_device2. That also caused a memory overflow
- make it possible again to partition old style blk_device volumes
- enable format voldmanaged devices to ext2 and ext3 with existing binaries by using optional blk_device2 (defined in recovery to NULL but not currently implemented by fs_mgr
- fix unexpected behavior when in format/partition menus and insert an usb key (REFRESH return code)
- ensure_path_unmounted() could not unmount /sdcard path on /data/media devices
- format_device() and format_volume() properly handle /sdcard path argument on /data/media devices

Change-Id: I2d5872690805e70175ba1346fd03978e3c2b879e

This will preserve logging for format/usb mount and for all USB OTG storage

Change-Id: I0ab41632b4b97d23a43471755e21d9750c6cfa04

Change-Id: I43ad3d2600b9529d726aac435f0091713786ac4f

Change-Id: Ie94b28694c12182652464c4b8e61ab4c67905b07

 * AOSP added a "bold" flag for image based fonts. Stay compatible
   with that API since there are other components using minui.

Change-Id: I6c0f12f7f8f1df8dd27779b47883ccff96f2d291

Change-Id: I6ec438f61cafce69c5083977c4330277f56aa060

 * Breaks every device because of their deprecated graphics.c overloads.
 * Revert until we clean this mess up.

This reverts commit 48ccdce.

Change-Id: I8179607517f55b2941924b5726e0e57b012f8e18

* Some devices rely on the LCD backlight to blank/unblank the display,
  support them by allowing a custom LCD backlight path to be used.

Change-Id: I8406b4b55cd7a2afc4f8f8ba65de2c53b949489d

…o cm-10.2

 * Clean all the stuff that conflicts with AOSP recovery.

Change-Id: I175329c446a014c451f4c0fc5a9a4472941afc10

 * We need to clean everything, otherwise nasty things happen when
   switching to alternate recovery module.

Change-Id: I9ea88179f5329112c587b649328c1c83df0bd5d9

This reverts commit 2457cc0.

Introduce two new updater functions:

* set_metadata
* set_metadata_recursive

Long term, these functions are intended to be more flexible replacements
for the following methods:

* set_perm
* set_perm_recursive

Usage:

  set_metadata("filename", "key1", "value1", "key2", "value2", ...)
  set_metadata_recursive("dirname", "key1", "value1", "key2", "value2", ...)

Description:

set_metadata() and set_metadata_recursive() set the attributes on a file/directory
according to the key/value pairs provided. Today, the following keys are
supported:

* uid
* gid
* mode (set_perm_extd only)
* fmode (set_perm_extd_recursive only)
* dmode (set_perm_extd_recursive only)
* selabel
* capabilities

Unknown keys are logged as warnings, but are not fatal errors.

Examples:

* set_metadata("/system/bin/netcfg", "selabel", "u:object_r:system_file:s0");

This sets the SELinux label of /system/bin/netcfg to u:object_r:system_file:s0.
No other changes occur.

* set_metadata("/system/bin/netcfg", "uid", 0, "gid", 3003, "mode", 02750, "selabel", "u:object_r:system_file:s0", "capabilities", 0x0);

This sets /system/bin/netcfg to uid=0, gid=3003, mode=02750,
selinux label=u:object_r:system_file:s0, and clears the capabilities
associated with the file.

* set_metadata_recursive("/system", "uid", 0, "gid", 0, "fmode", 0644, "dmode", 0755, "selabel", "u:object_r:system_file:s0", "capabilities", 0x0);

All files and directories under /system are set to uid=0, gid=0,
and selinux label=u:object_r:system_file:s0. Directories are set to
mode=0755.  Files are set to mode=0644 and all capabilities are cleared.

Bug: 10183961
Bug: 10186213
Bug: 8985290
Change-Id: Ifdcf186a7ed45265511dc493c4036e1ac5e3d0af

Bug: 10183961
Bug: 10186213
Bug: 8985290
Change-Id: I57cb14af59682c5f25f1e091564548bdbf20f74e

Add a copy of reboot.c from system/core/reboot, since it is no
longer a cleanly usable tool in toolbox.
minor API updates