Fixed function definition and restored missing ksu_try_umount calls that were unintentionally removed in previous commits. This includes unmounts for:
- /debug_ramdisk
- /sbin
- /system/etc/hosts
- /apex/com.android.art/bin/dex2oat64
- /apex/com.android.art/bin/dex2oat32
Fixes: b443b11df9 ("kernel: Refactoring the mount point unloading logic")
Fixes: da7632470d ("kernel: 使用 try_umount 替换 ksu_try_umount 以统一卸载逻辑")
Signed-off-by: schqiushui <orochi9999@gmail.com>
Skip directories that does NOT have the same magic as /data/app.
This is to avoid scanning incfs and any other stacked filesystems.
While this is way dumber, it's way cheaper.
no kern_path(), no missable path_put(), no ref handling.
This supercedes
`throne_tracker: avoid cross fs access
(https://github.com/tiann/KernelSU/pull/2626)`
- upstream
0b6998b474
Signed-off-by: backslashxx
<118538522+backslashxx@users.noreply.github.com>
context: this is known by many as `selinux hook`, `4.9 hook`
add is_ksu_transition check which allows ksud execution under nosuid.
it also eases up integration on 3.X kernels that does not have check_nnp_nosuid.
this also adds a `ksu_execveat_hook` check since this transition is NOT needed
anymore once ksud ran.
Usage:
if (check_ksu_transition(old_tsec, new_tsec))
return 0;
on either check_nnp_nosuid or selinux_bprm_set_creds (after execve sid reset)
reference: dfe003c9fd
taken from:
`allow init exec ksud under nosuid`
- 3df9df42a6
- https://github.com/tiann/KernelSU/pull/166#issue-1565872173
Signed-off-by: backslashxx <118538522+backslashxx@users.noreply.github.com>
Signed-off-by: rsuntk <rsuntk@yukiprjkt.my.id>
* Some people reports about undefined reference to `sys_umount`
* Since ksys_umount exist on Linux 4.17-rc1, then we gonna use that one.
Rev 2: Use correct int instead of long for ksys_umount
Signed-off-by: rsuntk <90097027+rsuntk@users.noreply.github.com>
I am repasting here what I posted on the source code originally:
/*
* turns out path_umount backport is completely unneeded
* we copy the trick used on strncpy_from_unsafe_user / strncpy_from_user_nofault
* https://elixir.bootlin.com/linux/v4.4.302/source/mm/maccess.c#L184
* basically
*
* mm_segment_t old_fs = get_fs(); // remember original fs segment
* set_fs(USER_DS); // or KERNEL_DS *
* do_whatever_in_userspace();
* set_fs(old_fs); // restore fs segment
*
* * kernel -> user, KERNEL_DS, user -> kernel, USER_DS
*
* so yes, we can try to straight up call a syscall from kernel space
*
* NOTE: on newer kernels you can use force_uaccess_begin + force_uaccess_end
* ref: https://elixir.bootlin.com/linux/v5.10.237/source/mm/maccess.c#L250
*
*/
path_umount backport now optional — neat trick, werks, what can I say.
Backports? Nah, we’re good.
Signed-off-by: backslashxx <118538522+backslashxx@users.noreply.github.com>
On plain ARMv8.0 devices (A53,A57,A73), strncpy_from_user_nofault() sometimes
fails to copy `filename_user` string correctly. This breaks su ofc, breaking
some apps like Termux (Play Store ver), ZArchiver and Root Explorer.
This does NOT seem to affect newer ARMv8.2+ CPUs (A75/A76 and newer)
My speculation? ARMv8.0 has weak speculation :)
here we replace `strncpy_from_user_nofault()` with another routine:
- access_ok() to validate the pointer
- strncpy_from_user() to copy and validate string
- manual null-termination just in case, as strncpy_from_user_nofault also does it
- remove that memset, seems useless as it is an strncpy, not strncat
Kind of mimicking _nofault, but yes with this one we allow pagefaults.
Tested on:
- ARMv8.0 A73.a53, A57.a53, A53.a53
- ARMv8.2 A76.a55
Tested-by: iDead XD <rafifirdaus12bb@gmail.com>
Signed-off-by: backslashxx <118538522+backslashxx@users.noreply.github.com>
- We don't know if KPM can run on arm32-bit devices, so to avoid some problems, add a dependency on 64-bit architectures
Signed-off-by: ShirkNeko <109797057+ShirkNeko@users.noreply.github.com>
