Commit Graph

206 Commits

Author SHA1 Message Date
topjohnwu
01b985eded Remove more pre SDK 21 stuffs 2021-04-09 21:29:42 -07:00
Shaka Huang
3da318b48e
Fix random return value of faccessat() in x86
faccessat() should return 0 when success, but it returns random number with errno == 0 in x86 platform.

It’s a side effect of commit bf80b08b5f when magisk binaries ‘corretly’ linked with library of API16 .. lol

Co-authored-by: John Wu <topjohnwu@gmail.com>
2021-04-04 03:04:09 -07:00
topjohnwu
66e30a7723 Build libc++ ourselves 2021-03-25 01:00:10 -07:00
vvb2060
6865652125 Fix process name in MagiskHide
close #3997
2021-03-24 04:43:45 -07:00
vvb2060
0783f3d5b6 Fix mount rules dir
close #4006
2021-03-13 21:10:02 -08:00
topjohnwu
39982d57ef Fix logging implementation
- Block signals in logging routine (fix #3976)
- Prevent possible deadlock after fork (stdio locks internally)
  by creating a new FILE pointer per logging call (thread/stack local)
2021-03-06 13:55:30 -08:00
Shaka Huang
ba1ce16b8b Fix error in pure 64-bit environment
In Android S preview, there’s no 32-bit libraries in x86_64 system image for emulator.

Signed-off-by: Shaka Huang <shakalaca@gmail.com>
2021-02-22 03:28:54 -08:00
topjohnwu
79140c7636 Proper xxread and xwrite implementation 2021-01-17 01:42:45 -08:00
topjohnwu
67e2a4720e Fix xxread false negatives
Fix #3710
2021-01-16 21:43:53 -08:00
topjohnwu
f2cb3c38fe Update mmap implementation
Always map memory as writable, but private when read-only
2021-01-12 22:50:55 -08:00
topjohnwu
d2acd59ea8 Minor code refactoring 2021-01-12 00:07:48 -08:00
topjohnwu
b36e6d987d Reorganize MagiskHide code
Prepare for zygote injection hiding
2021-01-10 17:11:00 -08:00
topjohnwu
4060c2107c Add preliminary zygote code injection support
Prototyping the injection setup and a clean "self unloading" mechanism.
2021-01-06 22:21:17 -08:00
topjohnwu
cd23d27048 Fix remote_write implementation 2021-01-06 21:56:29 -08:00
topjohnwu
5f2e22a259 Support remote function call with ptrace
End up not used for anything, but keep it for good
2021-01-02 21:29:45 -08:00
topjohnwu
f9bde347bc Convert indentation to spaces
The tab war is lost
2020-12-30 22:11:24 -08:00
topjohnwu
8e61080a4a Preparation for hiding isolated processes 2020-12-30 15:55:53 -08:00
topjohnwu
db4f78d463 Unblock signals before executing commands 2020-12-27 15:05:39 -08:00
topjohnwu
eee7f097e3 Make post-fs-data scripts block at most 35 secs 2020-12-17 16:54:53 -08:00
topjohnwu
2a694596b5 Better error handling and logging 2020-12-05 10:23:49 -08:00
topjohnwu
f152e8c33d Directly log to log file 2020-12-03 20:15:18 -08:00
topjohnwu
797ba4fbf4 Make sure all logging ends with newline 2020-12-02 00:55:22 -08:00
topjohnwu
0064b01ae0 Trim out \r from string
Fix #3490
2020-11-15 06:30:29 -08:00
topjohnwu
afebe734b8 Fix several things regarding scripting 2020-11-07 14:36:13 -08:00
topjohnwu
4c94f90e5d Templatize function callbacks 2020-11-03 01:16:55 -08:00
topjohnwu
ffb4224640 Don't use reserved symbols 2020-11-03 01:12:33 -08:00
topjohnwu
4fd04e62af Remove compressed ramdisk support
It is causing more issues than it addresses
2020-09-24 02:49:09 -07:00
topjohnwu
43146b8316 Update su request process
Due to changes in ec3705f2ed, the app can
no longer communicate with the dameon through a socket opened on the
daemon side due to SELinux restrictions. The workaround here is to have
the daemon decide a socket name, send it to the app, have the app create
the socket server, then finally the daemon connects to the app through
the socket.
2020-06-19 03:52:25 -07:00
topjohnwu
1e2f776b83 Move logging.hpp 2020-06-17 01:17:28 -07:00
topjohnwu
ec3705f2ed Redesign of MagiskSU's sepolicy model
Introduce new domain `magisk_client` and new file type `magisk_exec`.

Connection to magiskd's always-on socket is restricted to magisk_client
only. Whitelisted process domains can transit to magisk_client through
executing files labelled magisk_exec. The main magisk binary shall be
the only file labelled as magisk_exec throughout the whole system.
All processes thus are no longer allowed to connect to magiskd directly
without going through the proper magisk binary.

Connection failures are silenced from audit logs with dontaudit rules,
so crazy processes which traverse through all unix domain sockets to try
connection can no longer check logcat to know the actual reason behind
EACCES, leaking the denied process policy (which is u:r:magisk:s0).

This also allows us to remove many rules that open up holes in
untrusted_app domains that were used to make remote shell work properly.
Since all processes establishing the remote shell are now restricted to
the magisk_client domain, all these rules are moved to magisk_client.
This makes Magisk require fewer compromises in Android's security model.

Note: as of this commit, requesting new root access via Magisk Manager
will stop working as Magisk Manager can no longer communicate with
magiskd directly. This will be addressed in a future commit that
involves changes in both native and application side.
2020-06-03 23:29:42 -07:00
topjohnwu
2f824f59dc Better logging system
Use C++ magic to strip out debug logs at compile time
2020-06-01 04:15:37 -07:00
topjohnwu
e02e46d0fc Detect volume down key combo for safe mode
It is possible that a module is breaking the device so bad that zygote
cannot even be started. In this case, system_server cannot start and
detect the safe mode key combo, set the persist property, and reboot.

Also on old Android versions, the system directly goes to safe mode
after detecting a key combo without rebooting, defeating the purpose of
Magisk's safe mode protection if we only check for the persist property.

Directly adding key combo check natively in magiskd allows us to enter
Magisk safe mode before the system is even aware of it.
2020-05-19 04:57:47 -07:00
topjohnwu
99ef20627a Remove unused code 2020-05-18 05:45:08 -07:00
topjohnwu
122b4d66b6 Move Android logging out of libutils 2020-05-10 00:48:41 -07:00
topjohnwu
c113f854a2 Fix overlay.d on SAR again 2020-05-07 02:30:43 -07:00
topjohnwu
0c99c4d93f More complete support for fstab in dt 2020-05-03 22:49:54 -07:00
topjohnwu
8ab045331b Workaround realpath FORTIFY crashes 2020-05-03 22:11:39 -07:00
topjohnwu
947e3b06b4 Use template to get lambda for RAII 2020-04-30 01:27:48 -07:00
topjohnwu
c7ed0ef5eb Fix SAR support for overlay.d 2020-04-25 23:19:36 -07:00
topjohnwu
43029f37b1 Cleanup our tracks 2020-04-19 04:57:18 -07:00
topjohnwu
4c959cd983 Support cases when /sbin does not exist 2020-04-18 23:19:19 -07:00
topjohnwu
dcf07ad8c7 Directly filter '.' and '..' in xreaddir 2020-04-18 04:20:21 -07:00
topjohnwu
ed6cdb2eb4 Rename file 2020-04-18 04:10:19 -07:00
topjohnwu
ab853e1fcf Update dir traversal code 2020-04-12 13:38:57 -07:00
topjohnwu
d739dcac2b Remove dependency on magisk.hpp in libutils 2020-04-11 04:40:40 -07:00
topjohnwu
93ef90cd24 Fix FORTIFY crashes 2020-04-11 04:05:34 -07:00
topjohnwu
4066e5bf14 Update Makefiles 2020-04-06 22:45:08 -07:00
topjohnwu
9820296e92 Update files.cpp in libutils 2020-04-02 02:17:45 -07:00
topjohnwu
dbfde74c1e Clean rootfs in switch_root 2020-04-01 23:37:11 -07:00
topjohnwu
5f1174de27 Introduce new boot flow to handle SAR 2SI
The existing method for handling legacy SAR is:
1. Mount /sbin tmpfs overlay
2. Dump all patched/new files into /sbin
3. Magic mount root dir and re-exec patched stock init

With Android 11 removing the /sbin folder, it is quite obvious that
things completely break down right in step 1.

To overcome this issue, we have to find a way to swap out the init
binary AFTER we re-exec stock init. This is where 2SI comes to rescue!

2SI normal boot procedure is:
1st stage -> Load sepolicy -> 2nd stage -> boot continue...

2SI Magisk boot procedure is:
MagiskInit 1st stage -> Stock 1st stage -> MagiskInit 2nd Stage ->
-> Stock init load sepolicy -> Stock 2nd stage -> boot continue...

As you can see, the trick is to make stock 1st stage init re-exec back
into MagiskInit so we can do our setup. This is possible by manipulating
some ramdisk files on initramfs based 2SI devices (old ass non SAR
devices AND super modern devices like Pixel 3/4), but not possible
on device that are stuck using legacy SAR (device that are not that
modern but not too old, like Pixel 1/2. Fucking Google logic!!)

This commit introduces a new way to intercept stock init re-exec flow:
ptrace init with forked tracer, monitor PTRACE_EVENT_EXEC, then swap
out the init file with bind mounts right before execv returns!

Going through this flow however will lose some necessary backup files,
so some bookkeeping has to be done by making the tracer hold these
files in memory and act as a daemon. 2nd stage MagiskInit will ack the
daemon to release these files at the correct time.

It just works™  ¯\_(ツ)_/¯
2020-04-01 04:39:28 -07:00