Specify per-service rules for PDX transport. Now being able to
grant permissions to individual services provided by processes,
not all services of a process.

Also tighter control over which permissions are required for
client and server for individual components of IPC (endpoints,
channels, etc).

Bug: 37646189
Change-Id: I78eb8ae8b6e08105666445a66bfcbd2f1d69d0ea
Merged-Id: I78eb8ae8b6e08105666445a66bfcbd2f1d69d0ea

This reverts commit 8c60f74d.

Bug: 38242876
Change-Id: Iba5a94d16901dc0c52f1941972c26877baa4805c

Node for /dev/uhid driver needs to be accessible
by shell for the 'hid' command in frameworks/base/cmds.
This CL is in support of another CL c/2048848, topic
'Refactor hid command in /frameworks/base/cmds'
in internal master.

Bug: 34052337
Test: CTS test for GamepadTestCase#testButtonA; Checked that
cat /dev/uhid does not raise permission error.

Change-Id: I861c1226b4a67272af7c2a93d7811bf87a083478

Remove SELinux access from domain_deprecated. Access to SELinux APIs can
be granted on a per-domain basis.

Remove appdomain access to SELinux APIs. SELinux APIs are not public and
are not intended for application use. In particular, some exploits poll
on /sys/fs/selinux/enforce to determine if the attack was successful,
and we want to ensure that the behavior isn't allowed. This access was
only granted in the past for CTS purposes, but all the relevant CTS
tests have been moved to the shell domain.

Bug: 27756382
Bug: 28760354
Test: Device boots and no obvious problems. No collected denials.
Change-Id: Ide68311bd0542671c8ebf9df0326e512a1cf325b

The service "storaged" implememnts a dump() interface for
dumpsys, and thus it needs to write its state to the fd
provided by dumpstate.

To correct this, and fix dumpstate, allow the permission.

Fixes:
avc: denied { use } for pid=3298 comm="dumpsys" path="pipe:[33470]" dev="pipefs" ino=33470 scontext=u:r:storaged:s0 tcontext=u:r:dumpstate:s0 tclass=fd permissive=0

Test:
With a device that has storaged, issue the command:
$ adb shell dumpstate

Change-Id: I515e20f0328b6edc01ea2a7c53b1d3c4ca0e72ac
Signed-off-by: William Roberts <william.c.roberts@intel.com>

This was accidentally omitted from all_untrusted_app

While I'm here, split across mutiple lines and alphabetize.

Test: policy compiles.
Change-Id: I7fe1d1d0a4ef2ed3ab010931ee2ba15637c2be51

These were missing when the sepolicy was migrated.

Addresses denials:

E SELinux : avc:  denied  { find } for service=drm.drmManager pid=11769
uid=10018 scontext=u:r:mediaprovider:s0:c512,c768
tcontext=u:object_r:drmserver_service:s0 tclass=service_manager

W kworker/u16:2: type=1400 audit(0.0:1667): avc: denied { use } for
path="/storage/emulated/0/DCIM/Camera/IMG_20170425_124723.jpg"
dev="sdcardfs" ino=1032250 scontext=u:r:kernel:s0
tcontext=u:r:mediaprovider:s0:c512,c768 tclass=fd permissive=0

Bug: 37685394
Bug: 37686255
Test: Sync files
Test: Open downloaded file

Change-Id: Ibb02d233720b8510c3eec0463b8909fcc5bbb73d

iptables recently changed its behavior to strictly require xtables.lock.
dumpstate selinux policy must be updated to allow access.

Bug: 37648320
Test: dumpstate succeeds with no avc: denied ... xtables.lock messages
Change-Id: Ic7e243739f375a60fa14fe67fac910d31d978ffd
(cherry picked from commit ca097979)

Bluetooth needs the capability to set audio-related threads to be RT
scheduled.  Grant it sys_nice.

system_server needs to set priority for the Bluetooth HAL.  Allow it.

Bug 37518404
Test:  Play Bluetooth audio, confirm RT scheduling with systrace
Change-Id: Iaf7b85a11a51883744d72a50addfd320b6fbbc2f

Bug: 37476041
Test: make, pair and connect to HID device
Change-Id: Ic7e81382994769e3f3a91255dcf3624edeaf6bfd

MediaProvider requires permissions that diverge from those
of a typical priv_app. This create a new domain and removes
Mtp related permissions from priv_app.

Bug: 33574909
Test: Connect with MTP, download apps and files, select ringtones
Test: DownloadProvider instrument tests, CtsProviderTestCases

Change-Id: I950dc11f21048c34af639cb3ab81873d2a6730a9

Remove domain_deprecated from bluetooth. This removes some unnecessarily
permissive rules.

Bug: 25433265
Test: All of the permissions being removed were being audited. Verify
      that no audited (granted) avc messages for bluetooth exist in
      in the logs.

Change-Id: Ifa12a0f1533edcb623bbb9631f88f1ff1d6d7085

These permissions allow the system server to create and
bind a UDP socket such that it gains the SOCK_BINDPORT_LOCK.
(ref: af_inet.c - inet_bind()) This prevents the user from
disconnecting the socket, which would create a security
vulnerability. The user may then use the provided socket,
which is always IPv4/UDP, for IKE negotiation. Thus, an
un-trusted user app must be able to use the socket for
communication.

-ALLOW: read, write, connect, sendto, and recvfrom.
-NEVERALLOW: anything else

Bug: 30984788
Test: CTS tested via IpSecManagerTest:testUdpEncapsulationSocket

Change-Id: I045ba941797ac12fd14a0cce42efdd2abc4d67e0

This was marked deprecated in 2014 and removed in 2015, let's remove
the sepolicy now too.

Test: see that logging still works on bullhead

Change-Id: I4caa0dbf77956fcbc61a07897242b951c275b502

With build/core eaa9d88cf, system_server should not be loading code
from /data. Add an auditallow rule to report violations.

Bug: 37214733
Test: Boot marlin, no SELinux audit lines for system_server.
Change-Id: I2e25eb144503274025bd4fc9bb519555851f6521

Add /dev/kmsg_debug on userdebug devices, to allow crash_dump to log
crashes to dmesg when logd isn't up yet (or is the one crashing).

Bug: http://b/36574794
Test: stop tombstoned; crasher; dmesg
Change-Id: I249e11291c58fee77098dec3fd3271ea23363ac9

This change extends the recovery mode modprobe sepolicy
to support loadable kernel module in normal mode by using
statement below in init.rc:

exec u:r:modprobe:s0 -- /system/bin/modprobe \
    -d /vendor/lib/modules mod

Bug: b/35653245
Test: sailfish  with local built kernel and LKM enabled
Change-Id: I827e2ce387c899db3e0e179da92e79c75d61f5ae

Test: works on internal marlin
Bug: 34274385
Change-Id: Idd35e5cdccb595b4e5994eb1d78fdeece0aec0a6

logcatd is the same as logcat, except that the -L flag, if supplied,
runs once, then the command re-runs itself without the -L flag with
the same argument set.  By introducing a logcatd daemon executable
we can solve the problem of the longish reads from pstore that
sometimes occur when the system is excessively busy spinning in a
foreground task starving this daemon as we absorb the delay in
an init service, rather than in an init exec.  This would not have
been efficiently possible without the introduction of liblogcat.

Test: gTest logcat-unit-tests
Test: Manual check logpersist operations
Bug: 28788401
Bug: 30041146
Bug: 30612424
Bug: 35326290
Change-Id: I3454bad666c66663f59ae03bcd72e0fe8426bb0a

Test: adb kill-server && adb shell dumpsys storaged
Bug: 36492915
Change-Id: I3a1a2ad2f016ddd5770d585cae82c8be69001df9

This CL changes the policy for ASAN files on-disk to support the
changes made by the following CLs -
https://android-review.googlesource.com/#/c/359087/
https://android-review.googlesource.com/#/c/359389/

which refactor the on-disk layout of sanitized libraries in the following
manner -
/data/lib* --> /data/asan/system/lib*
/data/vendor/* --> /data/asan/vendor/*

There are a couple of advantages to this, including better isolation
from other components, and more transparent linker renaming and
SELinux policies.

Bug: 36574794
Bug: 36674745
Test: m -j40 && SANITIZE_TARGET="address" m -j40 and the device
boots. All sanitized libraries are correctly located in /data/asan/*,
and have the right SELinux permissions.

Change-Id: Ib08e360cecc8d77754a768a9af0f7db35d6921a9

Per loop(4), this device is the preferred way of allocating new
loop devices since Linux 3.1.

avc: denied { read write } for name="loop-control" dev="tmpfs" ino=15221 scontext=u:r:vold:s0 tcontext=u:object_r:device:s0 tclass=chr_file permissive=0

Bug: 34903607
Change-Id: I1f5f62cf0a1c24c6f6453100004812af4b8e1503

Bug: 36546152
Bug: 36278706

Test: `adb shell screencap ...` and pull and visually verify image.
Change-Id: Iab2ddcfc145cb7f55104cd8f1ce0d58286bca282

Test: `adb shell am hang --allow-restart` -> Watchdog dumps
  hal traces (eventually)
Bug: 36414311

Change-Id: I57e6875998b1f06a7deec1b8774facb75148d2c1

This change associates all domains which are clients of Allocator HAL
with hal_allocator_client and the, required for all HAL client
domains, halclientdomain.

This enables this commit to remove the now unnecessary hwallocator_use
macro because its binder_call(..., hal_allocator_server) is covered by
binder_call(hal_allocator_client, hal_allocator_server) added in this
commit.

Unfortunately apps, except isolated app, are clients of Allocator HAL
as well. This makes it hard to use the hal_client_domain(...,
hal_allocator) macro because it translates into "typeattribute" which
currently does not support being provided with a set of types, such as
{ appdomain -isolated_app }. As a workaround, hopefully until
typeattribute is improved, this commit expresses the necessary
association operation in CIL. private/technical_debt.cil introduced by
this commit is appended into the platform policy CIL file, thus
ensuring that the hack has effect on the final monolithic policy.

P. S. This change also removes Allocator HAL access from isolated_app.
Isolated app shouldn't have access to this HAL anyway.

Test: Google Play Music plays back radios
Test: Google Camera records video with sound and that video is then
      successfully played back with sound
Test: YouTube app plays back clips with sound
Test: YouTube in Chrome plays back clips with sound
Bug: 34170079
Change-Id: Id00bba6fde83e7cf04fb58bc1c353c2f66333f92

This is a follow-up to f5446eb1 where
I forgot to associate su and perfprofd domains with coredomain.

Test: mmm system/sepolicy
      sepolicy-analyze $OUT/root/sepolicy attribute coredomain
Bug: 35870313
Change-Id: I13f90693843f7c6fe9fea8e5332aa6dd9558478a

On PRODUCT_FULL_TREBLE devices, non-vendor domains (except vendor
apps) are not permitted to use Binder. This commit thus:
* groups non-vendor domains using the new "coredomain" attribute,
* adds neverallow rules restricting Binder use to coredomain and
  appdomain only, and
* temporarily exempts the domains which are currently violating this
  rule from this restriction. These domains are grouped using the new
  "binder_in_vendor_violators" attribute. The attribute is needed
  because the types corresponding to violators are not exposed to the
  public policy where the neverallow rules are.

Test: mmm system/sepolicy
Test: Device boots, no new denials
Test: In Chrome, navigate to ip6.me, play a YouTube video
Test: YouTube: play a video
Test: Netflix: play a movie
Test: Google Camera: take a photo, take an HDR+ photo, record video with
      sound, record slow motion video with sound. Confirm videos play
      back fine and with sound.
Bug: 35870313
Change-Id: I0cd1a80b60bcbde358ce0f7a47b90f4435a45c95

Test: no relevant denials on marlin while booting
Test: no relevant denials on angler while booting
Bug: 36278706
Change-Id: Ieba79e1c8fca4f74c63bc63e6dd0bdcf59204ca2

Add a new type and context for IpSec to system SEPolicy

Bug: 35923241
Test: service starts + talks to NetD
Change-Id: I69356c8525b426d344fcc4858fc499ab12405b20

vndservicemanager is the context manager for binder services
that are solely registered and accessed from vendor processes.

Bug: 36052864
Test: vendorservicemanager runs
Merged-In: Ifbf536932678d0ff13d019635fe6347e185ef387
Change-Id: I430f1762eb83825f6cd4be939a69d46a8ddc80ff

Rules in clients of NFC HAL due to the HAL running (or previously
running) in passthrough mode are now targeting hal_nfc. Domains which
are clients of NFC HAL are associated with hal_nfc only the the HAL
runs in passthrough mode. NFC HAL server domains are always associated
with hal_nfc and thus get these rules unconditionally.

This commit also moves the policy of nfc domain to private. The only
thing remaining in the public policy is the existence of this domain.
This is needed because there are references to this domain in public
and vendor policy.

Test: Open a URL in Chrome, NFC-tap Android to another Android and
      observe that the same URL is opened in a web browser on the
      destination device. Do the same reversing the roles of the two
      Androids.
Test: Install an NFC reader app, tap a passive NFC tag with the
      Android and observe that the app is displaying information about
      the tag.
Test: No SELinux denials to do with NFC before and during and after
      the above tests on sailfish, bullhead, and angler.
Bug: 34170079

Change-Id: I29fe43f63d64b286c28eb19a3a9fe4f630612226

ASAN builds may require additional permissions to launch processes
with ASAN wrappers. In this case, system_server needs permission to
execute /system/bin/sh.

Create with_asan() macro which can be used exclusively on debug
builds. Note this means that ASAN builds with these additional
permission will not pass the security portion of CTS - like any
other debug build.

Addresses:
avc: denied { execute } for name="sh" dev="dm-0" ino=571
scontext=u:r:system_server:s0 tcontext=u:object_r:shell_exec:s0
tclass=file

Test: lunch aosp_marlin-userdebug;
      cd system/sepolicy; mm SANITIZE_TARGET=address;
      Verify permissions granted using with_asan() are granted.
Test: lunch aosp_marlin-userdebug;
      cd system/sepolicy; mm;
      Verify permissions granted using with_asan() are not granted.
Test: lunch aosp_marlin-user;
      cd system/sepolicy; mm SANITIZE_TARGET=address;
      Verify permissions granted using with_asan() are not granted.
Bug: 36138508
Change-Id: I6e39ada4bacd71687a593023f16b45bc16cd7ef8

Now that the android wifi framework has fully switched over to HIDL,
remove the sepolicy permissions for accessing wpa_supplicant using
socket control interface.

While there, also removed the redundant |hwbinder_use|.

Bug: 35707797
Test: Device boots up and able to connect to wifi networks.
Test: Wifi integration tests passed.
Change-Id: I55e24b852558d1a905b189116879179d62bdc76c

Prevent app domains (processes spawned by zygote) from acquiring
locks on files in /system. In particular, /system/etc/xtables.lock
must never be lockable by applications, as it will block future
iptables commands from running.

Test: device boots and no obvious problems.
Change-Id: Ifd8dc7b117cf4a622b30fd4fffbcab1b76c4421b

Test: Boots, runs
Bug: 32713782
Change-Id: Ia58db3c4c0159482f08e72ef638f3e1736095918

Apps should be able to access the configstore HAL since framework
libraries which are loaded into app process can call configstore.

Letting apps have direct access to this HAL is OK because: 

(1) the API of this HAL does not make clients provide any sensitive 
information to the HAL, which makes it impossible for the HAL to 
disclose sensitive information of its clients when the HAL is 
compromised, 

(2) we will require that this HAL is binderized (i.e., does not run 
inside the process of its clients), 

(3) we will require that this HAL runs in a tight seccomp sandbox 
(this HAL doesn't need much access, if at all) and,

(4) we'll restrict the HALs powers via neverallows.

Test: apps can use configstore hal.

Change-Id: I04836b7318fbc6ef78deff770a22c68ce7745fa9

This switches Allocator HAL policy to the design which enables us to
identify all SELinux domains which host HALs and all domains which are
clients of HALs.

Allocator HAL is special in the sense that it's assumed to be always
binderized. As a result, rules in Camera HAL target hal_allocator_server
rather than hal_allocator (which would be the server and any client, if
the Allocator HAL runs in passthrough mode).

Test: Device boots up, no new denials
Test: YouTube video plays back
Test: Take photo using Google Camera app, recover a video, record a slow
      motion video
Bug: 34170079
Change-Id: Ifbbca554ec221712361ee6cda94c82f254d84936

Every client of Graphics Allocator HAL needs permission to (Hw)Binder
IPC into the HAL.

Test: Device boots, no denials to do with hal_graphics_allocator
      (also, removing the binder_call(hal_graphics_allocator_client,
      hal_graphics_allocator_server) leads to denials)
Test: GUI works, YouTube works
Bug: 34170079

Change-Id: I5c64d966862a125994dab903c2eda5815e336a94

This switches Boot Control HAL policy to the design which enables us
to conditionally remove unnecessary rules from domains which are
clients of Boot Control HAL.

Domains which are clients of Boot Control HAL, such as update_server,
are granted rules targeting hal_bootctl only when the Boot Control HAL
runs in passthrough mode (i.e., inside the client's process). When the
HAL runs in binderized mode (i.e., in another process/domain, with
clients talking to the HAL over HwBinder IPC), rules targeting
hal_bootctl are not granted to client domains.

Domains which offer a binderized implementation of Boot Control HAL,
such as hal_bootctl_default domain, are always granted rules targeting
hal_bootctl.

P. S. This commit removes direct access to Boot Control HAL from
system_server because system_server is not a client of this HAL. This
commit also removes bootctrl_block_device type which is no longer
used. Finally, boot_control_hal attribute is removed because it is now
covered by the hal_bootctl attribute.

Test: Device boots up, no new denials
Test: Reboot into recovery, sideload OTA update succeeds
Test: Apply OTA update via update_engine:
      1. make dist
      2. Ensure device has network connectivity
      3. ota_call.py -s <serial here> out/dist/sailfish-ota-*.zip
Bug: 34170079
Change-Id: I9c410c092069e431a3852b66c04c4d2a9f1a25cf

This switches most remaining HALs to the _client/_server approach.
To unblock efforts blocked on majority of HALs having to use this
model, this change does not remove unnecessary rules from clients of
these HALs. That work will be performed in follow-up commits. This
commit only adds allow rules and thus does not break existing
functionality.

The HALs not yet on the _client/_server model after this commit are:
* Allocator HAL, because it's non-trivial to declare all apps except
  isolated apps as clients of this HAL, which they are.
* Boot HAL, because it's still on the non-attributized model and I'm
  waiting for update_engine folks to answer a couple of questions
  which will let me refactor the policy of this HAL.

Test: mmm system/sepolicy
Test: Device boots, no new denials
Test: Device boots in recovery mode, no new denials
Bug: 34170079
Change-Id: I03e6bcec2fa02f14bdf17d11f7367b62c68a14b9