Commit Graph

83 Commits

Author SHA1 Message Date
topjohnwu
ed027ec3ee Refactor build flags 2019-02-12 05:17:02 -05:00
topjohnwu
ed4df87b57 Remove imgtool 2019-02-12 02:44:46 -05:00
topjohnwu
1321f097b8 Remove usage of magisk.img
Mounting ext4 images causes tons of issues, such as unmountable with broken F2FS drivers.
Resizing is also very complicated and does not work properly on all devices.
Each step in either measuring free space, resizing, and shrinking the image is a
point of failure, and either step's failure could cause the module system completely broken.

The new method is to directly store modules into /data/adb/modules, and for module installation
on boot /data/adb/modules_update. Several compatibility layers has been done: the new path is
bind mounted to the old path (/sbin/.magisk/img), and the helper functions in util_functions.sh
will now transparently make existing modules install to the new location without any changes.

MagiskHide is also updated to unmount module files stored in this new location.
2019-02-12 02:14:57 -05:00
topjohnwu
71ecbb3af3 Clean/refactor includes 2019-02-10 03:57:51 -05:00
topjohnwu
3a422c3f15 Remove magisklogd, use threads and BlockingQueue 2019-02-10 01:05:19 -05:00
topjohnwu
b3242322fd Harden socket verification
- Do not allow connections to magiskd from binaries other than the one started the server
- Do not allow connections to magisklogd without root access
2019-02-09 15:02:46 -05:00
topjohnwu
c8491d008f Move sbin overlay creation to magiskinit 2019-02-09 01:51:46 -05:00
topjohnwu
fdf167db11 Get API level from build.prop 2019-01-20 15:20:34 -05:00
topjohnwu
3e4c12cf56 Migrate to STL 2019-01-19 23:59:37 -05:00
topjohnwu
03c39e692a Switch to libc++ 2019-01-19 13:47:33 -05:00
topjohnwu
23f8f35098 Stop using system STL since it is no longer supported 2018-12-25 19:38:44 +08:00
topjohnwu
2ab999f4ca Fix bug in DB query wrapper 2018-11-20 02:20:49 -05:00
topjohnwu
ab5fedda0b Prevent Magisk database race condition
The database should only be accessed by a single process, which is magiskd.
This means 'magisk --sqlite [SQL]' has to be updated to pass the SQL command to the daemon.
In addition, open the database connection with SQLITE_OPEN_FULLMUTEX to support multithread in magiskd.
2018-11-16 03:20:30 -05:00
topjohnwu
ab74290fe3 Move magiskhide config into database 2018-11-16 01:15:34 -05:00
topjohnwu
572e078d87 Fully deprecate <mount_point>/.core folder
Symlinks are preserved for backwards compatibility
2018-11-15 22:55:28 -05:00
topjohnwu
ee4548230b Disable native systemless hosts, add built-in systemless hosts module 2018-11-15 13:57:41 -05:00
topjohnwu
376e7977f0 Deprecate path /sbin/.core, switch to /sbin/.magisk
Symlink is preserved for backwards compatibility
2018-11-15 01:36:03 -05:00
topjohnwu
3a2a2a4ffa Micro optimizations 2018-11-13 02:07:02 -05:00
topjohnwu
5743c72cca Minor cleanup 2018-11-08 15:23:36 -05:00
topjohnwu
3d269fe8be Migrate MagiskInit to C++ 2018-11-08 04:20:16 -05:00
topjohnwu
803c5377a6 Clean init.c 2018-11-07 02:21:15 -05:00
topjohnwu
7c12bf7fa1 Modernize code base 2018-11-07 02:10:38 -05:00
topjohnwu
0742901cd2 Modernize database code 2018-11-04 18:24:08 -05:00
topjohnwu
5e4d2dedbe Minor log_daemon changes 2018-11-04 17:23:08 -05:00
topjohnwu
cda57dd4b4 Fully migrate Magisk to C++ 2018-11-04 04:15:51 -05:00
topjohnwu
a7824af5a8 Expose persist prop API 2018-11-03 00:15:21 -04:00
topjohnwu
b8a3cc8b60 Separate magiskhide logic from main daemon 2018-11-01 14:08:33 -04:00
topjohnwu
27c688252d Store hidelist in magisk database 2018-11-01 13:23:12 -04:00
topjohnwu
dbb8b8a439 Handle magisk.db completely natively
Prevent database corruption due to different Android application sqlite default settings
2018-10-27 17:54:48 -04:00
topjohnwu
a9121fa28f Reorganize libutils and cleanups 2018-10-12 21:46:09 -04:00
topjohnwu
acf7c0c665 Minor reorganization of daemons 2018-10-12 00:50:47 -04:00
topjohnwu
d119dd9a0c Rewrite su daemon and client 2018-10-04 04:59:51 -04:00
topjohnwu
09ef19f7ec Code cleanups 2018-10-04 01:49:52 -04:00
topjohnwu
5020cd1bbf Small cleanup 2018-09-28 01:25:43 -04:00
topjohnwu
cce636224c Reorganization 2018-09-27 18:26:41 -04:00
topjohnwu
60b3b8ddce Better incremental builds 2018-09-27 03:56:56 -04:00
topjohnwu
41446ec9ba Separate libutils and libsystemproperties 2018-09-27 03:30:16 -04:00
topjohnwu
df8b047bca Generalize logging interface 2018-09-27 03:11:10 -04:00
topjohnwu
1d53335ae5 Dynamic load libselinux 2018-09-27 00:09:59 -04:00
topjohnwu
906b4aad9e New method of communication
Introduce a new communication method between Magisk and Magisk Manager.

Magisk used to hardcode classnames and send broadcast/start activities to
specific components. This new method makes no assumption of any class names,
so Magisk Manager can easily be fully obfuscated.

In addition, the new method connects Magisk and Magisk Manager with random
abstract Linux sockets instead of socket files in filesystems, bypassing
file system complexities (selinux, permissions and such)
2018-09-16 04:16:18 -04:00
topjohnwu
6f41d9855b Randomize service names
Fix Aniplex Game detections.
Close #502, close #513
2018-09-08 23:17:00 -04:00
topjohnwu
7a0025673c Use libsystemproperties in resetprop
Upstream to latest Android Pie
2018-09-06 02:57:02 -04:00
topjohnwu
72cf5f3f9f Temporary disable module bootloop prevention
Some devices don't like it, need further tests before pushing to production
2018-08-31 03:23:59 -04:00
topjohnwu
5bac442b18 Reorganize sources 2018-08-10 03:49:25 +08:00
topjohnwu
6add682705 Remove high compression mode 2018-08-10 03:49:25 +08:00
topjohnwu
d3858b81e2 Add new boot service: boot-complete 2018-08-09 14:52:44 +08:00
topjohnwu
ada0f93686 Apply all sepolicy patches pre-init
Boot services tend to fail in the middle when the kernel loads a sepolicy live.
It seems that moving full patch (allow magisk * * *) to late_start is still not enough to fix service startup failures.
So screw it, apply all patched in magiskinit, which makes sure that all rules are only loaded in a single step.
The only down side is that some OEM with a HUGE set of secontexts (e.g. Samsung) might suffer a slightly longer boot time, which IS the reason why the rules are split to 2 parts in the first place.
2018-08-09 03:20:28 +08:00
topjohnwu
7b5d79d313 Kill all processes using the same UID of the target
To workaround OOS embryo optimization
2018-08-08 05:47:58 +08:00
topjohnwu
c38533e0f8 Prevent problematic modules causing device stuck in bootloop
If boot failed after 2 times, it will enable core only mode (which disables all modules)
2018-08-07 04:41:48 +08:00
topjohnwu
5be035fd44 Try logging a little harder 2018-08-03 01:58:56 +08:00