The build is broken. Reverting temporarily to fix breakage.

libsepol.check_assertion_helper: neverallow on line 4758 violated by allow init kernel:security { setenforce };
Error while expanding policy
make: *** [out/target/product/mako/obj/ETC/sepolicy_intermediates/sepolicy] Error 1
make: *** Waiting for unfinished jobs....


This reverts commit bf12e225.

Change-Id: I78a05756d8ce3c7d06e1d9d27e6135f4b352bb85

As per the discussion in:
https://android-review.googlesource.com/#/c/71184/



init sets the enforcing mode in its code prior to switching to
the init domain via a setcon command in the init.rc file.  Hence,
the setenforce permission is checked while still running in the
kernel domain.  Further, as init has no reason to ever set the
enforcing mode again, we do not need to allow setenforce to the
init domain and this prevents reverting to permissive
mode via an errant write by init later.  We could technically
dontaudit the kernel setenforce access instead since the first
call to setenforce happens while still permissive (and thus we
never need to allow it in policy) but we allow it to more accurately
represent what is possible.

Change-Id: I617876c479666a03167b8fce270c82a8d45c7cc6
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

lmkd low memory killer daemon

The kernel low memory killer logic has been moved to a new daemon
called lmkd.  ActivityManager communicates with this daemon over a
named socket.

This is just a placeholder policy, starting off in unconfined_domain.

Change-Id: Ia3f9a18432c2ae37d4f5526850e11432fd633e10

Limit the ability to write to the files that configure kernel
usermodehelpers and security-sensitive proc settings to the init domain.
Permissive domains can also continue to set these values.

The current list is not exhaustive, just an initial set.
Not all of these files will exist on all kernels/devices.
Controlling access to certain kernel usermodehelpers, e.g. cgroup
release_agent, will require kernel changes to support and cannot be
addressed here.

Expected output on e.g. flo after the change:
ls -Z /sys/kernel/uevent_helper /proc/sys/fs/suid_dumpable /proc/sys/kernel/core_pattern /proc/sys/kernel/dmesg_restrict /proc/sys/kernel/hotplug /proc/sys/kernel/kptr_restrict /proc/sys/kernel/poweroff_cmd /proc/sys/kernel/randomize_va_space /proc/sys/kernel/usermodehelper
-rw-r--r-- root     root              u:object_r:usermodehelper:s0 uevent_helper
-rw-r--r-- root     root              u:object_r:proc_security:s0 suid_dumpable
-rw-r--r-- root     root              u:object_r:usermodehelper:s0 core_pattern
-rw-r--r-- root     root              u:object_r:proc_security:s0 dmesg_restrict
-rw-r--r-- root     root              u:object_r:usermodehelper:s0 hotplug
-rw-r--r-- root     root              u:object_r:proc_security:s0 kptr_restrict
-rw-r--r-- root     root              u:object_r:usermodehelper:s0 poweroff_cmd
-rw-r--r-- root     root              u:object_r:proc_security:s0 randomize_va_space
-rw------- root     root              u:object_r:usermodehelper:s0 bset
-rw------- root     root              u:object_r:usermodehelper:s0 inheritable

Change-Id: I3f24b4bb90f0916ead863be6afd66d15ac5e8de0
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

Denials seen on hammerhead but seem
appropriate for general policy.

<5>[ 8.339347] type=1400 audit(3731546.390:17): avc: denied { ioctl } for pid=314 comm="rild" path="socket:[7996]" dev="sockfs" ino=7996 scontext=u:r:rild:s0 tcontext=u:r:rild:s0 tclass=socket
<5>[ 8.339065] type=1400 audit(3731546.390:16): avc: denied { create } for pid=314 comm="rild" scontext=u:r:rild:s0 tcontext=u:r:rild:s0 tclass=socket
<5>[ 11.232121] type=1400 audit(3731549.289:22): avc: denied { read } for pid=620 comm="rild" scontext=u:r:rild:s0 tcontext=u:r:rild:s0 tclass=socket

Change-Id: Ieaca5360afbb44d5da21c7c24bdd5e7c5758f0a2

This label was originally used for Motorola
Xoom devices. nvmap is the tegra gpu memory
manager and the various nvhost drivers are
for tegra graphics related functionality,
i.e. display serial interface, image signal
processor, or media processing stuff.

Only grouper and tilapia presently need this
policy.

Change-Id: I2a7000f69abf3185724d88d428e8237e0ca436ec

Also make su and shell permissive in non-user builds to allow
use of setenforce without violating the neverallow rule.

Change-Id: Ie76ee04e90d5a76dfaa5f56e9e3eb7e283328a3f
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

Change-Id: I10006f43c142f07168e2ea0f4f5f7af68d03e504
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

Otherwise sockets that have no specific entry match the /dev(/.*) entry
instead, leaving them in device type rather than socket_device type.
Every socket should get its own entry regardless, but this at least puts
it into a more specific type by default.

Change-Id: I97f7999af7f9f83484d3a51440dda791d3726f1a
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

Change-Id: I11be7d1713dd7cb35b8046503a09e42567e53d86
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

Allow tmpfs_domains the ability to write to ashmem allocated
regions. At least one Google internal app does this, and switching
untrusted_app into enforcing causes the following denial:

<5>[  291.791423] type=1400 audit(1385587240.320:79): avc:  denied  { write } for  pid=3774 comm="XXXXXXXXXXXX" path=2F6465762F6173686D656D202864656C6574656429 dev="tmpfs" ino=16937 scontext=u:r:untrusted_app:s0 tcontext=u:object_r:untrusted_app_tmpfs:s0 tclass=file

path=/dev/ashmem (deleted)

Bug: 11891764
Change-Id: I64d414c055cd02481ebf69994fad65d777d8381d

* commit '043b9027':
  Confine watchdogd, but leave it permissive for now.

Change-Id: If2285e927cb886956b3314dd18384145a1ebeaa9
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

* commit '1ed3caf7':
  Add support for duplicate allow rule detection (-D / --dups).

Usage:
sepolicy-analyze -D -P out/target/product/<board>/root/sepolicy

Displays duplicate allow rules, i.e. pairs of allow rules that grant
the same permissions where one allow rule is written directly in terms
of individual types and the other is written in terms of attributes
associated with those same types.  The rule with individual types is
a candidate for removal.  The rule with individual types may be directly
represented in the source policy or may be a result of expansion of
a type negation (e.g. domain -foo -bar is expanded to individual allow
rules by the policy compiler).  Domains with unconfineddomain will
typically have such duplicate rules as a natural side effect and can
be ignored.

Also add a tools/README with a description of all of the tools.

Change-Id: I07838dbd22c5cc8a4a65b57003ccae38129050f5
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

* commit '006260e5':
  Confine hostapd, but leave it permissive for now.

* commit '2b939e8c':
  Confine ping, but leave it permissive for now.

Change-Id: I23a2c568e9fdd51c6c09c6c80a7ce9f2b5bd4966
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

* commit 'ae49e7a3':
  Confine tee, but leave it permissive for now.

* commit '6ce3d60c':
  Confine rild, but leave it permissive for now.

* commit 'b1cb3205':
  Confine wpa_supplicant, but leave it permissive for now.

Change-Id: I6df9981b2af0150c6379a0ebdbe0a8597c994f4a
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

Change-Id: Id69b1fe80746429a550448b9168ac7e86c38aa9f
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

Change-Id: Iaa4ed5428d1c49cb4cff3a39c48800cb108f2ac3
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

* commit '91ebcf33':
  netd: allow tcp_socket name_connect

The patch in 36a5d109 wasn't
sufficient to address DNS over TCP. We also need to allow
name_connect.

Fixes the following denial:

<5>[   82.120746] type=1400 audit(1830030.349:5): avc:  denied  { name_connect } for  pid=1457 comm="netd" dest=53 scontext=u:r:netd:s0 tcontext=u:object_r:port:s0 tclass=tcp_socket

Public Bug: https://code.google.com/p/android/issues/detail?id=62196
Bug: 11097631

Change-Id: I688d6923b78782e2183a9d69b7e74f95d6e3f893

* commit '36a5d109':
  netd: allow tcp connections.

DNS can use TCP connections, in addition to UDP connections.
Allow TCP connections.

Addresses the following denial:

[ 1831.586826] type=1400 audit(1384129166.563:173): avc:  denied  { create } for  pid=11406 comm="netd" scontext=u:r:netd:s0 tcontext=u:r:netd:s0 tclass=tcp_socket

Public Bug: https://code.google.com/p/android/issues/detail?id=62196
Change-Id: Ia542a9df3e466a8d409955bab6a23a524ff3d07b
Bug: 11097631

* commit '8510d31e':
  Rename camera_calibration_file and audio_firmware_file.

Use more general type names for the contents of /data/misc/camera and
/data/misc/audio.  These were the names used in our policy until 4.3
was released, at which point we switched to be consistent with AOSP.
However, the Galaxy S4 4.2.2 image, Galaxy S4 4.3 image, and
Galaxy Note 3 4.3 image all shipped with policies using _data_file names
because they were based on our older policy.  So we may as well switch
AOSP to these names.

Not sure if in fact these could be all coalesced to the new media_data_file
type for /data/misc/media introduced by
Ic374488f8b62bd4f8b3c90f30da0e8d1ed1a7343.

Options to fix already existing devices, which would only apply
to Nexus devices with 4.3 or 4.4 at this point:
1) Add restorecon_recursive /data/misc/audio /data/misc/camera to either
the system/core init.rc or to the device-specific init.*.rc files.
-or-
2) Add a typealias declaration in the policy to remap the old type names.
to the new ones.  Then existing types on persistent storage will be
remapped internally to the new ones.
-or-
3) Some sort of relabeld.

Option #2 is implemented by this change.

Change-Id: Id36203f5bb66b5200efc1205630b5b260ef97496
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>

* commit 'bc1388d3':
  Make kernel / init enforcing

* commit '56f39193':
  Confine debuggerd, but leave it permissive for now.