134 Commits

Author SHA1 Message Date
topjohnwu
a63696836c Proper addon.d-v2 support 2018-08-03 22:40:49 +08:00
topjohnwu
46aad00f16 Use buffer on stack 2018-08-03 21:30:44 +08:00
topjohnwu
252afe8932 Use mirror in post-fs-data scripts 2018-08-03 17:09:24 +08:00
topjohnwu
82e8375957 Respect filesystem type when mounting mirrors
Close #405
2018-08-03 04:45:07 +08:00
topjohnwu
169c0fe4af Stop use clashing names 2018-08-03 03:43:02 +08:00
topjohnwu
cd6918e6eb Stop altering PATH to mirror 2018-08-03 03:38:36 +08:00
topjohnwu
5be035fd44 Try logging a little harder 2018-08-03 01:58:56 +08:00
topjohnwu
f1edc8443c Make root shell always use dev_pts
Close #433
2018-08-02 20:29:18 +08:00
topjohnwu
d9564bd04c Delay full sepolicy patch loading time 2018-08-02 05:35:01 +08:00
topjohnwu
91818cfa1a Support compiling split cils via magiskpolicy CLI 2018-07-21 05:12:22 +08:00
topjohnwu
4ffc388491 Allow bootctl to run 2018-07-20 22:22:49 +08:00
topjohnwu
3ae959af95 Do not early mount on symlinks on half Treble devices
Fix #399
2018-07-19 17:43:37 +08:00
topjohnwu
e11e88a9c5 Huawei hardcodes the slot suffix, don't append the suffix twice 2018-07-19 03:12:28 +08:00
topjohnwu
7cec8baa55 Merge magiskpolicy into Magisk main repo 2018-07-18 18:45:21 +08:00
topjohnwu
e987db9fb5 Move magiskpolicy sources to native/jni/magiskpolicy 2018-07-18 18:44:43 +08:00
topjohnwu
c603b9084f Remove magiskpolicy as submodule 2018-07-18 18:43:36 +08:00
topjohnwu
492d6dfcf0 Merge MagiskSU into Magisk main repo 2018-07-18 18:40:13 +08:00
topjohnwu
a3e0f2dcc3 Remove MagiskSU as submodule 2018-07-18 18:40:10 +08:00
topjohnwu
cf211e26f4 Move MagiskSU sources to native/jni/su 2018-07-18 18:23:36 +08:00
topjohnwu
d08f326990 Log fatal errors in debug mode only 2018-07-18 03:25:36 +08:00
topjohnwu
adf95ce3a0 Read fstab from device tree
In previous versions, magiskinit will not early mount if /sepolicy is detected. However on OP5/5T latest betas, the devices are fully trebelized,
but for some reason the file /sepolicy still exists, making magiskinit think it is NOT a treble device and doesn't work properly.

So to properly fix this issue, I will have to use the "official" way - check fstab in device trees. Any block mentioned in the fstab in device trees
are supposed to be early mounted. Currently magiskinit will only mount system and vendor even if other partitions exists in the dtb fstab, since other
partitions are not used to construct sepolicy (currently).

These changes can also fix #373, since we dynamically detect PARTNAME from device trees.
2018-07-18 00:45:10 +08:00
topjohnwu
3c1aca114f Wrap accept4 since some device does not have it 2018-07-17 06:52:23 +08:00
topjohnwu
18d0fd9d2a MagiskInit optimizations 2018-07-17 06:18:36 +08:00
topjohnwu
2bde8a1975 EMUI's logcat will change output file's permission and break ADBD 2018-07-16 06:42:36 +08:00
topjohnwu
bf9927c7dd Sync source with AOSP 2018-07-14 04:34:48 +08:00
topjohnwu
f339a087a2 Let Magisk compile against SDK 16 2018-07-13 22:14:32 +08:00
topjohnwu
1affb91f17 Support compiling against lower SDK
Reduce even more size for static binaries
2018-07-13 05:41:29 +08:00
topjohnwu
c6fc0e587e Support hardlink based proc ns 2018-07-11 23:41:38 +08:00
topjohnwu
7ed2c077de Support deodexed ROM on Oreo 2018-07-07 01:37:04 +08:00
topjohnwu
1283167595 Maintain our own set of loop devices 2018-07-07 01:32:58 +08:00
topjohnwu
23c2e22910 Update image functions 2018-07-06 22:04:06 +08:00
topjohnwu
f44b2dbd45 Rename log_monitor -> log_daemon 2018-07-06 07:57:18 +08:00
topjohnwu
46ee2c3f4e Improve handshake between the 2 daemons 2018-07-06 07:51:17 +08:00
topjohnwu
5d5ec08566 Test the log buffers before running command 2018-07-06 07:32:16 +08:00
topjohnwu
ef175e3cbe Open log file in append mode 2018-07-04 23:46:40 +08:00
topjohnwu
e7a2144def Optimize magiskhide to work with the log daemon 2018-07-04 01:52:23 +08:00
topjohnwu
52a2c6958b Optimize log daemon 2018-07-03 22:25:39 +08:00
topjohnwu
6c4d81b1e9 Invincible mode implemented in magisklogd 2018-07-03 01:38:19 +08:00
topjohnwu
c88dc8795b Single log file 2018-07-02 22:48:26 +08:00
topjohnwu
a8030c39b1 Separate logging into its own daemon 2018-07-02 22:11:28 +08:00
topjohnwu
7243b9e72f Improve log_monitor implementation 2018-07-01 18:18:12 +08:00
topjohnwu
d149af9628 Fix bootloop when upgrading from older Magisk 2018-07-01 14:58:31 +08:00
topjohnwu
528634d755 Remove unused code 2018-06-27 06:04:16 +08:00
topjohnwu
e579f314a6 Fix MagiskSU force denying root access 2018-06-26 18:46:18 +08:00
topjohnwu
e837bdc8ad Update BusyBox
BusyBox is unable to run properly on non-root applications due to seccomp introduced in Android 8.0.
The SDK-21 libc.a has system call wrappers that uses the system calls on the whitelist, so binaries compiled with the updated libc can work properly.
2018-06-25 16:11:31 +08:00
topjohnwu
7265450e2e Precise free space calculation for magisk.img
1. Introduce new applet: imgtool for better separation from the main program
2. Actually mount the image and check statvfs for free space in the image

This shall eliminate any possible module installation failure from image resizing issues.
2018-06-22 06:18:06 +08:00
topjohnwu
058dbc9f9e Add more indentation 2018-06-21 18:11:43 +08:00
osm0sis
a82ef6bd35 MagiskHide: add ro.boot.vbmeta.device_state=locked 2018-06-19 14:23:06 -07:00
npes87184
312466aaf8 Prevent setting zero over than bound
The &cmd will return a pointer which point to a pointer of cmdline.
It is a memory address which is usually 8 bytes in 64 bits machine.

However, the struct cmdline is 4 bytes. This will cause setting zero
beyond the bound.

Below is a simple example to show the differentiation:

struct cmdline {
        char skip_initramfs;
        char slot[3];
};

static void parse_cmdline(struct cmdline *cmd)
{
        printf("%lu\n", sizeof(*cmd)); /* 4 */
        printf("%lu\n", sizeof(&cmd)); /* 8 */
}

int main()
{
        struct cmdline cmd;
        parse_cmdline(&cmd);
        return 0;
}

This patch prevents this.

Signed-off-by: npes87184 <npes87184@gmail.com>
2018-06-19 14:22:11 -07:00
topjohnwu
44395e8ff0 Fix root loss issue when MagiskHide is enabled
In previous implementations, proc_monitor checks whether the mount namespace of an app is actually separated from zygote using a list generated at startup.
However, for some unknown reason, some devices (e.g. Samsung) has multiple zygote servers running in the background.
This means that app processes spawned from the unlisted zygotes are not checked whether the separation is done or not, causing MagiskHide unmount stuffs in the namespace of zygote, and since zygote is the "mother" of all apps, all apps will no longer have root access.

Since I'm not sure of the reason why multiple zygotes exists, so instead of checking the namespace against a list, compare the current namespace against the parent process's namespace.
This will make sure the namespace is NOT the same as the parent process, which is supposed to be the zygote server.
2018-06-20 04:55:34 +08:00