Commit Graph

121 Commits

Author SHA1 Message Date
7400dc2700 missing file 2010-02-22 08:53:47 -08:00
981b0cd1e1 roll recovery tools into a single binary. fix up nandroid to work without deviceid requirement 2010-02-22 08:53:34 -08:00
225c6b4673 do error checking of nandroid operations 2010-02-21 22:02:24 -08:00
1fa52ecdbf add mount sdcard option 2010-02-21 21:29:10 -08:00
bcdd00359c Functional Nandroid 2010-02-21 21:10:25 -08:00
466e67a582 generate fstab using mkfstab.sh. implement restore of system data and cache in nandroid 2010-02-21 19:29:32 -08:00
1a7ee5384d add mkfstab.sh script to create an fstab on any device. clean up dump_image to use libmtdutils 2010-02-21 17:52:30 -08:00
7e5a661064 move unyaffs into external/yaffs2 2010-02-21 01:09:52 -08:00
f9476fbfe8 fix build break 2010-02-21 01:02:30 -08:00
c788c26397 more fixes 2010-02-20 17:25:03 -08:00
8ce0be4956 nearly working nandroid, built against libc 2010-02-20 15:59:06 -08:00
3ab130fa9e add missing files 2010-02-19 16:50:36 -08:00
bf055bb1be intetrate nandroid into recovery 2010-02-19 16:47:53 -08:00
a3c2f735d7 fix up back button and menu toggling 2010-02-19 14:17:22 -08:00
841b2bf352 Add static flash_image for recovery 2010-02-15 01:45:38 -08:00
3a976a79bf power button toggles menu 2010-02-13 15:33:49 -08:00
e923487ff6 working recovery image! 2010-02-12 00:43:24 -08:00
1f14c9a1f1 Need to call register_package_root prior to installing the package. The subsequent unregistering made me think that it was being registered at some point, but it was removed in the transition to eclair. The unregistration was dead code. It is being used now though. 2010-02-11 23:08:23 -08:00
6060e5c6df update.zip somewhat working now... 2010-02-11 22:27:06 -08:00
4c1eed2573 Initial reintegration of legacy update process that used update-script 2010-02-11 18:59:58 -08:00
0d4ff2fcc6 allow alt-l or menu to bring up the recovery menu 2010-02-10 12:10:02 -08:00
89d385c4d9 allow mouse/trackball click for enter on touch only devices. 2010-02-10 12:10:01 -08:00
22d79a5c5e make offsets in firmware update header not point to bad blocks
(This is being cherry-picked from master.)

hboot will apparently fail to install if the first block of the image
(the one pointed to by the offset in the block 0 header) is a bad
block.  (Hopefully it handles subsequent bad blocks.)

This change makes the MTD write code keep track of the bad blocks it
has skipped over, so that the offset in the header can be adjusted to
be the address of the first successfully written block.

http://b/2358012 - passion: failure to flash hboot (bad blocks?)
2010-01-13 10:07:28 -08:00
17a47098d2 use MEMGETBADBLOCK to look for bad blocks when reading MTD partitions 2009-12-14 18:27:03 -08:00
c652e41d91 fix cut-and-paste error in verifier
Oops.
2009-12-08 15:30:09 -08:00
54ec81fe86 Merge change I1c8ca2e4 into eclair
* changes:
  replace generic recovery icons
2009-10-07 20:09:16 -04:00
988500b615 add terminator to recovery's getopt_long options array
http://b/2170691 - recovery argument parsing is broken
2009-10-06 14:44:54 -07:00
1c8ca2e40b replace generic recovery icons 2009-10-02 15:42:31 -07:00
f93d8166ef confirm before wiping user data in recovery
When using the hidden menu to wipe data in recovery, confirm before
starting the wipe.  (This does not affect booting with the --wipe_data
flag, or using Alt+W on dream with the menu hidden -- those still wipe
data immediately.)
2009-09-22 18:16:56 -07:00
fd8fb0c492 reduce fraction of progress bar for verification
Reduce the fraction of the progress bar used for package verification
from 50% to 25%:
  - verification is faster than before due to sha1 improvements
  - in eclair we're now verifying the compressed data rather than
    decompressing it
  - incremental packages (which is what most installs use) write more
    data than is contained in the package.
2009-09-20 14:10:27 -07:00
60babf8ba7 delete files before symlinking; log error messages
The symlink() function should remove existing files before creating
symlinks, so scripts are idempotent.  Log messages when various system
calls fail (but don't make the whole script fail).
2009-09-18 15:11:24 -07:00
196c25c777 don't complain if recovery icon is short
If the a recovery icon file is so short that we can't even read the
8-byte header, put a message in the log but not on the device screen.
We intentionally have zero-length files for some icons on some devices,
if they're never shown (eg, the firmware installation icons are only
used on HTC devices).
2009-09-15 08:50:04 -07:00
a3f89eabb7 add a run_program() function to edify
Handy for producing debugging OTA packages (eg, running sqlite3 or
whatever in recovery).
2009-09-10 14:20:53 -07:00
34c98df78a do not merge: cherry-pick of c2d666bd4f from master 2009-08-18 12:05:45 -07:00
54e2e86c57 do not merge: cherry-picked 60151a295c from master branch 2009-08-17 13:21:04 -07:00
0cf0e33b5f merge from open-source master 2009-08-04 09:00:15 -07:00
b9955b8373 am 20697b96: only build flash_image for eng
Merge commit '20697b965cd923211daeea2037f60218358b9659'

* commit '20697b965cd923211daeea2037f60218358b9659':
  only build flash_image for eng
2009-08-04 08:44:43 -07:00
34109de24e merge from donut 2009-07-29 14:56:48 -07:00
eb1ac27bf0 reconcile korg/master into goog/master 2009-07-26 11:43:53 -07:00
7bd5c66075 Merge korg/donut into korg/master 2009-07-25 17:48:00 -07:00
20697b965c only build flash_image for eng
With the recovery image being installed by applypatch, the flash_image
tool isn't needed any more.  Continue to build it for eng just in case
it's handy for debugging.
2009-07-23 15:17:00 -07:00
23412e6f14 fix compile warnings in recovery, change images
gcc 4.4 complains about some of the recovery ui functions not being
declared.  To include the header, we have to fix the 'volatile'
declaration (otherwise there's a compiler error).

Move the dream-specific images to vendor/htc/dream, make the default
images a generic phone.
2009-07-23 10:36:43 -07:00
608fa02e1a resolved conflicts for merge of 64893ccc to master 2009-07-15 18:10:28 -07:00
64893ccc09 remove amend
Yank all the code to install OTA packages out of the recovery binary
itself.  Now packages are installed by a binary included in the
package (run as a child of recovery), so we can make improvements in
the installation process without waiting for a new release to use
them.
2009-07-14 16:58:42 -07:00
c457ff6436 am bec02d57: skip over all-zero blocks when reading MTD partition
Merge commit 'bec02d57fb85cc7dd0196a54b0e9530e306623ac'

* commit 'bec02d57fb85cc7dd0196a54b0e9530e306623ac':
  skip over all-zero blocks when reading MTD partition
2009-07-02 15:24:19 -07:00
bec02d57fb skip over all-zero blocks when reading MTD partition
We fail to detect certain bad blocks (marked in the factory as bad, I
think?) when reading mtd partitions.  These come back as a block of
all zeros.  Since it's fairly unlikely a legitimate boot or recovery
block will contain 128k of zeros, change mtdutils to skip over such
blocks.

Arve says https://review.source.android.com/10535 may be a long-term
fix for this, but he isn't yet sure.
2009-07-01 12:09:29 -07:00
fc382dfc75 am d8f7c9b8: Merge change 5545 into donut
Merge commit 'd8f7c9b85e25fab93fef2221a84b60edc2e7b837'

* commit 'd8f7c9b85e25fab93fef2221a84b60edc2e7b837':
  remove updater from the user system image
2009-06-29 16:39:15 -07:00
d8f7c9b85e Merge change 5545 into donut
* changes:
  remove updater from the user system image
2009-06-26 14:42:37 -07:00
ad3db099d5 remove updater from the user system image
updater (which is only needed in OTA packages) is getting included in
/system/bin, where it just takes up (quite a bit of) space.  Use the
hack of including it only in eng builds so it's not there for user
builds.
2009-06-26 13:38:55 -07:00
898ef399d3 am 0bbfe3d9: fix off-by-one error in set_perm()
Merge commit '0bbfe3d901885c1f0ab006e8d4cc1029c44a7376'

* commit '0bbfe3d901885c1f0ab006e8d4cc1029c44a7376':
  fix off-by-one error in set_perm()
2009-06-25 14:34:03 -07:00